Home

EMA001082-00 - MPCdrivers.com

1. x Name Will be assigned Type active backup Y Description NA Mac Address NA IP Address 192 168 0 7 Subnet Mask 266 255 255 0 Default Gateway 192 168 0 2 Ea Figure 64 Creating a NIC Bond 5 To create an active backup bond select active backup from the Type list or To create a NIC aggregation bond select 802 3ad from the Type list 6 Enter a static IP address for the bond The default value for the bond IP address is the IP address of one of the physical interfaces in the bond 7 Enter the Subnet mask The default value for the bond subnet mask is the subnet mask of one of the physical interfaces in the bond 8 Optional Enter the default gateway The default value for the bond default gateway is the gateway of the one of the physical interfaces in the bond 9 Click OK A confirmation message opens 10 Click OK to confirm the TCP IP changes A message opens prompting you to search for the bonded DataFRAME on the network 11 Search for the DataFRAME by subnet and mask or by IP address host name A message opens listing the manager IP addresses that must be set on the application servers 12 Click OK 13 Verify the new bond interface The TCP IP tab displays the new list of interfaces as shown in Figure 65 MPC DataFRAME 440 Storage System Software User Manual 95 Managing the Network Edit Configuration for DF140 192 168 0 7 Module T Storage
2. e ee Pass Fail Diagnostic Test Description Criteria Criteria Motherboard Compares the mother board Within Outside temperature temperature against the accepted range range temperature range for normal operation CPU temperature Compares the processor temperature Within Outside against the accepted temperature range range range for normal operation Mother board Compares the power supply voltages Allvoltages One or voltages against the accepted voltage range are within more for normal operation the range voltages outside range Enclosure Sends a passive command to the Backplane Backplane ER backplane and verifies that the returns times out communication response from the backplane expected or does not matches criteria string return expected string Hard drive status Checks the status of all installed All drives One or drives are On more and drives not Secured On and Secured Hard drive For each of the drives compares the Temp ofall One or temperature temperature against an accepted drives are more range for normal operation within drives out range of range Fan status Checks the fan status Fan is Fan is normal faulty Power supply Checks the power supply status Power Power status supply is supply is normal faulty Hard drive S M A R T Self Monitoring Analysis All drives Warning or SMART health and Reporting Technology is pass Failed if implemented in all m
3. What to Configure in the What to Configure in the iSCSI CHAP Level Authentication Group Initiator CHAP not required Initiator node name only e No configuration requirements 1 way CHAP CHAP Name see note belowh Enter the target secret when e Target Secret logging on to available target 2 way CHAP e CHAP Name Enter the initiator secret Target Secret Enter the target secret e Initiator Secret Note If using CHAP with a single node only use the initiator node name as the CHAP name Sample iSCSI Configurations Figure 169 illustrates the configuration for a single host authentication with CHAP not required New Authentication Group iSCSI Initiator Properties xj New Authentication Group General Discovery Targets Persistent Targets Bound Volumes Devices Name AuthGroup_0O SS The iSCSI protocol uses the following information to uniquely Gr identify this initiator and authenticate targets Description Volume List No Volume List Initiator Node Name ign 1991 05 com microsoft kwegner wxp iscsi EBSD Fibre Channel Z Allow access via iSCSI To rename the initiator node click Change Authentication To authenticate targets using CHAP click Secret to specify a CHAP secret L SEG CHAP not required To configure IPSec Tunnel Mode addresses click Tunnel z Tunnel Initiator Node Name iqn 1991 05 com microsoft l kwegner wxp CHAP re
4. Management Group TransData_Remote v Volume Name Ke New Volume Snapshot Name Snapshot_O Snapshot Description Figure 217 Creating a New Remote Snapshot MPC DataFRAME 440 Storage System Software User Manual 339 Remote Copy Creating the Primary Snapshot 1 In the Primary section of the New Remote Snapshot window click New Snapshot The New Snapshot window opens shown in Figure 218 New Snapshot New Snapshot Snapshot Name Snapshot_o Description l Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Ges Figure 218 Creating a New Primary Snapshot 2 Type a name for the primary snapshot Names are case sensitive They cannot be changed after the snapshot is created Note Make the beginning of volume and snapshot names meaningful for example Snap 1Exchg_03 The Console displays volume and snapshot names under the icons If aname is longer than the wiath of the icon the end of the name is cut off however the full name does show on the corresponding Details tab and on other relevant tab views 3 Optional Type in a description of the snapshot 4 Optional Change the hard and soft thresholds for the snapshot 5 Click OK to return to the New Remote Snapshot window The information for the primary snapshot is filled in as shown in Figure 219 At thi
5. Managers Managers are DataFRAMEs within a management group that you designate to govern the activity of all of the DataFRAMEs in the group All DataFRAMEs contain the management software but you must designate which DataFRAMEs in the management group you want to act as managers These DataFRAMEs run managers much like a PC runs various services Functions of Managers Managers control data replication keep track of system status coordinate reconfigurations as DataFRAMEs are brought up and taken offline and re synchronize data when DataFRAMEs fail and recover Managers and Quorum Managers use a voting technology to coordinate DataFRAME behavior In this voting technology a strict majority of managers a quorum must be running and communicating with each other in order for the Storage System Software to function Therefore for optimal fault tolerance you should have three or five managers in your management group Three or five managers provide the best balance between fault tolerance and performance 158 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Number of Number fora Management Managers Quorum Fault Tolerance Explanation 1 1 None If the manager fails no data control takes place 2 2 None If one manager fails there is not a quorum Not Recommended 3 2 High If one manager fails 2 remain so there is still a quorum Note 2 managers are not
6. Frame Size Default Setto Figure 69 Editing TCP Speed Duplex and Frame Size 3 Select the combination of speed and duplex that you want 4 Click OK A series of status messages display Then the changed setting displays in the TCP status report Note You can also use the Configuration Interface to edit the TCP speed and duplex See Setting the TCP Speed Duplex and Frame Size on page 306 Editing the NIC Frame Size The frame size specifies the size of data packets that are transferred over the network The default Ethernet standard frame size is 1500 bytes The maximum allowed frame size is 9000 bytes Increasing the frame size improves data transfer speed by allowing larger packets to be transferred over the network and by decreasing the CPU processing time required to transfer data However increasing the frame size requires that routers switches and other devices on your network support that frame size Note Increasing the frame size can cause decreased performance and other network problems if routers switches or other devices on your network do not support frame sizes greater than 1500 bytes If you are unsure about whether your routers and other devices support larger frame sizes keep the frame size at the default setting Note If you edit the frame size on a disabled or failed NIC the new setting will not be applied until the NIC is enabled or connectivity is restored Best Practice T
7. cccesececeseseeeeseeeeceeeeneeaeeesneeeeseeeeesaaseeeseeseceseeeseseaeseneeeeseeseeseaneeseeneenseneens 28 Figure 12 DataFRAMEs Tab with DatatHAMts LUsted AAA 28 Figure 13 Management Groups Tab from Network Tab View 29 Figure 14 Viewing messages in the Alert Messages Tab 29 Figure 15 EBSD Hosts Tab 30 Figure 16 Viewing individual DataFRAME mtormatton AAA 30 Figure 17 Icon Showing RAID is Normal 31 Figure 18 Icon Showing RAID is Om 31 Figure 19 Icon Showing RAID is Deoraded AAA 32 Figure 20 Icon Showing RAID is Pebuding A 32 Figure 21 Management Group Information Tab 32 Figure 22 Availability Tab wee ssic ccces sige tpn aaae aee ai d a rora o ba aoe oriai pairan 33 Figure 23 Logging in to a Data wl EE 33 Figure 24 Automatic Log in Failures ssi sccseneiszecgenecccene sccte neice e ae arai i 34 Figure 25 DataFRAME Configuration Window ssseeseeseesieereettritsttetttsttttuetttttnttttttnttntttnttnstuntnateuenntunennenaee nenea 35 Figure 26 DataFRAME Copy Configuration Wimdow 36 Figure 27 DataFRAME configuration window for the MPC DataFRAME 440 Storage Gvstem 37 Figure 28 Logging in ERC el LEE 38 Figure 29 Automatic Log In Fails because DataFRAME User Name and Password are Ditterent A 39 Figure 30 Upgrading the DataFRAME Goitware A 43 Figure 31 Browsing for Upgrade or Patch File 43 Figure 32 Upgrade Status Messages 44 Figure 33 Viewing the Backup and Restore Window A 45 Figure 34 Backing up th
8. 0 Configuration ategories Power Off Feature Registration Boot Devices Module Information Host Name IP Address 192 168 0 7 Model DF 100 Series Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Logged In User admin Backup amp Restore bao Am Install Software Change Password Figure 27 DataFRAME configuration window for the MPC DataFRAME 440 Storage System Configuration Categories The left pane lists the configuration categories The right pane contains a set of tabs which you use to configure different functions for each specific category The configuration categories are described below e Module Use the module category to change the host name and login password for the DataFRAME You can also install software backup and restore the Storage System Software configuration reboot or power off the DataFRAME register the DataFRAME for add on features and activate the flash cards used for booting the DataFRAME Storage Manage RAID and the individual disks in the DataFRAME Time Configure the time zone and set the date and time on the DataFRAME The date and time settings are used to create a time stamp on volumes and snapshots The date and time settings also affect schedules for snapshots and remote copies TCP IP Network For each DataFRAME you can configure and manage the network settings including TC
9. vi allow access via Fibre Channel Authenticate hosts with the following WWWPNs Name Optional pete Figure 173 Creating a New Authentication Group for Fibre Channel Access Planning Volumes and Fibre Channel First you configure authentication groups for Fibre Channel Then when you add the volume to a volume list you assign a LUN number for that volume Assigning LUN Numbers to Volumes Some requirements for authentication group access and LUN numbers include the following e LUN numbers must be assigned to volumes when the volumes are added to a volume list LUN numbers are required before the hosts can access the volumes e LUN numbers must be unique per authentication group For example one authentication group can only access one LUN 0 If you plan to have one host accessing two volumes with the same authentication group each volume must be assigned a different LUN number MPC DataFRAME 440 Storage System Software User Manual 271 Controlling Client Access to Volumes Best Practice Hosts with Separately Numbered LUNs Figure 174 shows a typical best practice configuration for assigning LUN numbers to volumes and associating LUNs to hosts Each host is accessing two LUNs Each of the two LUN associations has a different LUN number Engineering Volume 1 Engineering Authentication Group Engineering Volume 2 Finance Volume 1 Finance Authentication Group Finan
10. Failed The device encountered an I O error and is not ready to be used MPC DataFRAME 440 Storage System If a flash card has a status of Failed select the card and click Activate If the card fails repeatedly it needs to be replaced Empty MPC DataFRAME 440 Storage System The flash card bay on the front of the DataFRAME does not contain a boot flash card or the card in the slot is unreadable Unformatted The device has not yet been used in an DataFRAME It is ready to be activated Not Recognized The device in the flash card bay is not recognized as a boot flash device Unsupported The flash card in the flash card bay cannot be used For example it is the wrong size or card type 1 Some statuses listed above only occur in a system with two boot devices Note When the status of a flash card changes an alert is generated See Using Active Monitoring on page 144 Replacing a Boot Device Only in Modules with Two Boot Devices DataFRAME 440 Storage System If a compact flash card fails first try to activate it on the Boot Devices window If the card fails repeatedly replace it with a new one 54 Note Warning You can also replace a boot flash card if you have removed the original card to store it as a backup in a remote location The replacement boot flash card must be a standard 256 Mb or larger compact flash memory card As of publication product approved car
11. eecceeceeeceeseeeeeeeeeeeeeeeeeeaeeeaeeesaeesaeeseeeesaesaeeeseeeaeeseeeesaeeseeeseaeeeeeeeeeensaes 219 Creating the Volume and Setting Thresholds AAA 219 Elle RT GOW E 220 How Auto Grow Works sieissanto eener anma na aea aa aaa Ro Eiaa OR EAA IE Eaa raa aE rana EAEE PASEU EEE 220 Creating a Melmer sissie esiaine nnee ie Eea Ee E E EE Eeer ASTEA EASE EN TER EAE OEO En aoia 225 The Vol me NEE 227 Editing EC Weu 229 EE is DR CET 230 Changing the Volume Description AAA 231 Changing the Cluster AAA 231 Changing the Replication Level tege dEE EEN AE E 232 Changing the Replication Priority 0 cceceeceeeeeeeeeeeeeeeeeeeeeeeeeeceaeeeeeeseaeeeaeeseeeeeaeeseaeeeaeesieeesaeeseaeseeeeneeees 232 CHANGING the EE 232 Changing the Hard Threshold ceccesceeeeeeeeeeeeeeeeeeeeeeeeeeeeseeseaeeeaeeeseeesaeessaeseaeeseaeeeaeecaeeeaeeseaeeeaeeeneeesas 232 Changing the Soft Threshold A 232 Fixing a Replica challenged Redundant Volume ceecceeeeseeeeeneeeeeeneeeeeeeesenneeeseneeeeseneeeeeneeeseaeeesnnneeeseeeeseeees 233 Deleting VOU EE 233 Selecting a Volume or Snapshot from the List cceeceeeeeceeeeeeeeeeeeeeneeeeeeeneeeseeeeeaeeesaeeeaeeseeeesaeeseeeesaeesieeeeaeenaes 234 Chapter 13 Working with Snapshots ccccccesssseeeeeeeeeeesseseeneeeesseeeeeeeseeneeeeeeeenseeeeeneeeeees 235 Using SnapShots sessir enere e en Eea EE E E REE aS EEEa ESSEE ie 235 Single Snapshots versus Scheduled Snapshots AAA 236 Requirem
12. Cancel Figure 225 Editing a Remote Management Group 2 In the Remote Management Groups section click Edit Bandwidth The Edit Remote Bandwidth window opens Edit Remote Bandwidth x Remote Bandwidth 32768 KbiSec Figure 226 Editing the Remote Bandwidth 3 Change the bandwidth setting as desired For example change the value to 93 KB to use no more than about one half the capacity of a T1 line Note Both bandwidth settings are configured in kilobytes Be careful when configuring this parameter as you may be used to using bits for networking settings Canceling a Remote Snapshot When you cancel a remote snapshot that is in progress the remote snapshot is deleted and the primary snapshot remains To cancel a remote snapshot that is in progress 1 Click the primary or remote snapshot The snapshot tab view opens 2 Click the Remote Snapshot tab 3 Select from the list the remote snapshot you want to cancel 4 Click Cancel Remote Snapshot A confirmation message opens 346 MPC DataFRAME 440 Storage System Software User Manual Remote Copy 5 Click OK Editing a Remote Snapshot You can edit the description of a remote snapshot You can also change the hard and soft thresholds but it is not recommended 1 Log in to the management group that contains the remote snapshot 2 Right click the remote snapshot and select Edit Snapshot from the menu The Edit Snapshot window opens shown in Figure
13. Feature Registration Scripting Evaluation In order to use scripting during a license evaluation period you must check the box below By checking this box you are agreeing to enter the license evaluation period C Scripting Evaluation Period Enabled Cancel Figure 197 Enabling Scripting Evaluation 5 Check the box to enable the use of scripts during a license evaluation period 6 Click OK For more information about scripting see Chapter 14 Working with Scripting Turn Off Scripting Evaluation The scripting evaluation period is turned off when e You purchase the add on feature or application you were evaluating or e You complete the evaluation and decide not to purchase any add on features or applications Select the management group Select the Register tab From the Tasks menu select Feature Registration Select the Scripting Evaluation tab shown in Figure 197 Clear the check box Click OK Table 55 The table below describes additional steps to safely back out of the scripting evaluation oar OON MPC DataFRAME 440 Storage System Software User Manual 295 Feature Registration Table 55 Safely Backing Out of Scripting Evaluation Feature Being Evaluated Steps to Back Out Any of the items below that are created by 1 Back out of any configurable snapshots an application based script scheduled snapshots or remote copying e Scheduled snapshots 2 Delete any scripts
14. eee eeeeeceeeeneeeeeeceeeeeeeeeeaeeceaeeeaeeeaeeesaeeseaeseaeeseaeeeaeesaeeseeeeeeesiaeseeeeeees 227 Editing Re tu 231 VOlUIMG Fab TE 240 Creating a DEEN EE 241 New Snapshot Shapshot Labs ss iecacavieyenet hiwktdaiep iene Aa oti nde a eal Viewing the Writable Space Used for a Gnapehot 244 MPC DataFRAME 440 Storage System Software User Manual xvii List of Figures Figure 162 Figure 163 Figure 164 Figure 165 Figure 166 Figure 167 Figure 168 Figure 169 Figure 170 Figure 171 Figure 172 Figure 173 Figure 174 Figure 175 Figure 176 Figure 177 Figure 178 Figure 179 Figure 180 Figure 181 Figure 182 Figure 183 Figure 184 Figure 185 Figure 186 Figure 187 Figure 188 Figure 189 Figure 190 Figure 191 Figure 192 Figure 193 Figure 194 Figure 195 Figure 196 Figure 197 Figure 198 Figure 199 Figure 200 Figure 201 Figure 202 Figure 203 Figure 204 Figure 205 Figure 206 Figure 207 Figure 208 Figure 209 Figure 210 Figure 211 Figure 212 Figure 213 Figure 214 Figure 215 Figure 216 Figure 217 xviii Editin a Snapshot ciresarii eenaa oae inaa e AEE TEE EA NEEE e Aa a aA Eeka Etine Eea Creating a Snapshot Gchedule ttt Akut EAtEnEn nEn at enten ntnnt ennenen enne List of Scheduled Snapshots wis sisisi iin Ahh e a E satis ea a eN Editing a Snapshot Schedule A Rolling Backa Volume scien tiene Ain nd eee eee Verifying the Volume Roll Back Creating a Ne
15. 260 Below is an example of scripting automatic threshold increases using java commandline CommandLine java commandline CommandLine lt admin name gt lt admin password gt lt manager ip gt volume_autogrow_set lt volume name gt lt grow size Megabytes gt lt failure timeout seconds gt Example Joe Jones creates a script to automatically increase the hard and soft thresholds for his X Rays volume The volume length is 10 GB with a hard threshold of 2 GB and a soft threshold of 1 GB Joe scripts the increases for increments of 512 MB MPC DataFRAME 440 Storage System Software User Manual Working with Scripting java commandline CommandLine jjones trumpet 10 0 111 212 volume_autogrow_set X Rays 512 600 Reviewing the Increment Size for Increasing the Thresholds You can run an operation to review the setting for automatic threshold increases using java commandline CommandLine java commandline CommandLine lt admin name gt lt admin password gt lt manager ip gt volume_autogrow_ get lt volume name gt lt failure timeout seconds gt Example java commandline CommandLine jjones trumpet 10 0 111 212 volume _autogrow_get X Rays 60 Scripted Commands for Remote Copy Scripting operations for Remote Copy use the same tools that are available for scripting snapshots with the addition of parameters specific to Remote Copy Using the command line parameters allows you to create scripts for e Creating a primary snapshot
16. Figure 130 Adding DataFRAMEs to Cluster in Alternating Site Order Create the Volume with 2 way Replication 186 Configure the volume with 2 way replication Two way replication ensures that two copies of the data are written to the volume The fact that you added the DataFRAMEs to the cluster in alternating order ensures that a complete copy of the data exists on each site See Planning Data Replication on page 214 MPC DataFRAME 440 Storage System Software User Manual Disaster Recovery Using A Virtual Manager mre Computers DataFRAME Management Console File Find Tasks View Help G Available DataFRAMEs Group_0 sau SSChuster 0 am DF140 1 DF140 DF420 Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Name Cluster D Figure 131 Cluster with DataFRAMEs Added in Alternating Order Configuring a Virtual Manager In order to use a virtual manager in a management group beyond the 30 day evaluation period you must purchase the Scalability Pak See Chapter 16 Feature Registration Adding a Virtual Manager Add a virtual manager to a management group 1 From the Tasks menu on the Details tab select Add or Delete Virtual Manager A confirmation dialog opens 2 Click OK to continue The virtual manager is added to the management group The Details tab lists the virtual manager as added and the virtual
17. In the MPC DataFRAME 440 Storage System for example you have 1 DataFRAME with four 4 disk sets configured as 3 plus a spare and 1 DataFRAME with two 8 disk sets configured as 8 disks per set Configuring RAID e If you are configuring RAID 1 or RAID 10 the DataFRAME must contain an even number of disks e If you are configuring RAID 5 or RAID 50 the MPC DataFRAME 440 Storage System must contain 4 8 12 or 16 disks Warning Changing the RAID configuration will erase all the data on the disks 3 On the Storage configuration category click the RAID Setup tab to bring it to the front shown in Figure 42 4 Select the RAID configuration from the list 5 Click Configure RAID A confirmation message opens 6 Click OK MPC DataFRAME 440 Storage System Software User Manual 67 Storage A warning message opens 7 Click OK RAID starts configuring Note If the DataFRAME contains a large number of disks it may take several hours for the disks to synchronize in a RAID 10 configuration When the RAID status on the RAID Setup tab shows Normal the disks provide fully operational data redundancy with the mirror in place The DataFRAME is ready for data transfer at this point Setting RAID Rebuild Rate for RAID 1 10 or RAID 5 50 Choose the rate at which the RAID configuration rebuilds if a disk is replaced The rate is a percentage of the throughput of the RAID card e Setting the rate high is good for rebuilding RAID
18. MPC DataFRAME 440 Storage System Software User Manual 63 Storage Planning RAID Configuration The RAID configuration you choose for the DataFRAME depends on your plans for data safety data availability and capacity growth If you plan to expand your network of DataFRAMEs and create clusters of DataFRAMEs choose your RAID configuration carefully Warning Once RAID is configured you cannot change the RAID configuration without deleting all data on the DataFRAME Data Replication Keeping multiple copies of your data can ensure that data will be safe and will remain available in the case of disk failure There are two ways to achieve data replication e Configure RAID 1 10 5 or 50 within each DataFRAME e Replicate data volumes across clusters of DataFRAMEs Using RAID for Data Replication Within each DataFRAME RAID 1 or RAID 10 can ensure that 2 copies of all data exist If one of the disks in a RAID pair goes down data reads and writes can continue on the other disk Similarly RAID 5 or RAID 50 provides redundancy by spreading parity evenly across the disks in the set If one disk in the set goes down data reads and writes continue on the remaining disks in the set RAID protects against failure of disks within a module but not against failure of an entire DataFRAME For example if network connectivity to the DataFRAME is lost then data reads and writes to the DataFRAME cannot continue Note If you plan to create
19. Replacing a failed disk requires removing the DataFRAME from the cluster and management group replacing the disk and returning the DataFRAME to the cluster Because of the replication level removing and returning the DataFRAME to the cluster would normally cause the remaining DataFRAMEs in the cluster to restripe the data twice once when the DataFRAME is removed from the cluster and once when it is returned Repairing the DataFRAME creates a placeholder in the cluster in the form of a ghost DataFRAME This ghost DataFRAME keeps the cluster intact while you remove the DataFRAME replace the disk configure RAID and return the DataFRAME to the cluster The returned DataFRAME only has to resynchronize with the other two DataFRAMEs in the cluster Repairing a DataFRAME When a DataFRAME in a cluster has a disk failure the Network View displays the DataFRAME and the cluster as blinking red and needing attention shown in Figure 144 206 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters a Transaction_Data a a TransData05 Failed LCL CES LH Denyer 1 Boulder 1 DataFRAME blinks red Figure 144 DataFRAME with Failed Disk If the DataFRAME is running a manager stop the manager See Stopping a A Manager on page 173 2 Select the DataFRAME in the Network View 3 Right click and select Repair DataFRAME A confirmation message opens 4 Click OK The DataFRAME leaves
20. The new disks will show a red X and be listed as Off Select the new disks and click Power On Disk The disk status of the new disks becomes Uninitialized Select the new disks and click Add Disk to RAID Shift click to select multiple disks to add to RAID pairs of disks to RAID 1 10 or MPC DataFRAME 440 Storage System Software User Manual 77 Storage sets of 4 or 8 disks to RAID 5 50 MPC DataFRAME 440 Storage System or RAID begins to rebuild on the new disks according to the RAID Rebuild rate configured on the RAID Setup tab As soon as the RAID Status shows Normal the disks provide fully operational data redundancy with the mirror in place The DataFRAME is ready for data transfer at this point The newly added disks display on the RAID Setup tab Powering Drives On or Off Powering drives on and off is part of removing and replacing disks in the DataFRAME A bad drive should be powered off from the Console before you remove it from the module Then after the replacement disk is inserted in the drive bay it must be powered on Warning Any time you must remove a disk you should power it off from the Console before you physically remove it from the DataFRAME unless the DataFRAME itself is powered off Powering Drives Off 1 Select the Storage configuration category 2 Click the Disk Setup tab 3 Select the disk in the list to power off If the disk is on all the columns are filled in 4 Click P
21. Value text Name of the snapshot to create primary volume name Value text Name of the primary volume to make remote remote volume name Value text Name of the remote volume created in the Console remote snapshot name Value text Name of the remote snapshot remote snapshot description Value text Description of the remote snapshot MPC DataFRAME 440 Storage System Software User Manual Working with Scripting Table 46 Parameters for commandline CommandLine Parameter What It Is soft threshold see note Value number Size of the volume s new soft threshold in MegaBytes MB May be the soft threshold of the new primary volume if using Remote Copy hard threshold see note Value number Size of the source volume s new hard threshold in MegaBytes MB May be the hard threshold of the new primary volume if using Remote Copy description see note Value text Optional Description associated with the snapshot failure timeout seconds Value number The number of seconds to wait until exiting with a failure grow size Value number The size in MegaBytes by which to increase the volume thresholds LUN Value number The LUN number assigned to the Fibre Channel volume when associating the volume to an auth group volume_snapshot Use this value as written This is verbatim Creates a snapshot from a volum
22. Volume Name Description Created Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Primary Remote TransData_BkUp e None X Off Auto Manual O Yes No Cancel Edit Volume Type Volume Name Description Created Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Remote volume wi enabled editable fields Selecting Primary changes the volume to a primary volume with all fields greyed out Figure 237 Making a Remote Volume into a Primary Volume 4 Click OK The Edit Volume window displays the editable fields enabled You can edit everything but the name and created date and time MPC DataFRAME 440 Storage System Software User Manual 359 Remote Copy 5 Make any required changes i e to the size and hard and soft thresholds 6 Click OK The volume becomes a primary volume Configuring Failover Configuring Remote Copy for failover provides for business continuance and high availability When configuring failover you take into consideration both the failover path and the recovery from failover Planning Failover To achieve failover you plan the following parameters e The location and structure of management groups and clusters e Configuration of primary and remote volumes and snapshots and scheduling
23. 0 MB7 Snapshot_0 10 GB70 MB Ha D TEE Usage Graphs iSCSI Remote Snapshot Dak 04 Tasks NEI 0 MB Max Writable space Tisting Max Used and Max Full can temporarily exceed 100 during data migration Max Full for snapshots is Max Used Writable Max Used Hard Threshold Wititable Hard Threshold Figure 161 Viewing the Writable Space Used for a Snapshot The additional writable space is deleted when the snapshot is deleted If you need to free up the extra space before the snapshot is deleted you can do so manually or through your snapshot scripts The next time an application or operating system accesses the snapshot the writable space will be recreated Deleting a Snapshot s Writable Space Prerequisite Stop any applications from accessing the volume 1 Select the snapshot for which you want to delete the writable space 2 Right click and select Delete Writable Space A warning message opens 3 Click OK Editing a Snapshot You can edit the description of a snapshot You can also change the hard and soft thresholds See Creating a Snapshot on page 240 1 Log into the management group that contains the snapshot that you want to edit 2 Select the snapshot you want to edit The snapshot Tab View opens 3 From the Tasks menu select Edit Snapshot The Edit Snapshot window opens shown in Figure 162 244 MPC DataFRAME 440 Storage System Softw
24. 4 Click OK when you are finished adding users Removing Users from a Group 1 Select the user to remove in the Users section 2 Click Remove The user is removed from the list Changing Administrative Group Permissions Change the management capabilities available to members of a group The default setting is Read Only for each category 1 Click the Groups tab to bring it to the front 2 Select a group and click Edit The Edit Administrative Group window opens 3 Click the Group Permission tab to bring it to the front 4 Click the management capabilities you want for the group you are editing 5 Click OK when you are finished MPC DataFRAME 440 Storage System Software User Manual 125 Administrative Users and Groups Deleting Administrative Groups 1 Select DataFRAME Administration from the configuration categories 2 Click the Groups tab to bring it to the front 3 Select the group to delete 4 Click Delete A confirmation message opens 5 Click OK Note When you delete a group the users who are members of that group are NOT deleted Managing Administrative Users Add administrative users as necessary to provide access to the management functions of Storage System Software Note The user who is created during DataFRAME configuration using the Configuration Interface becomes a member of the Full Administrator group by default Adding Administrative Users Administrative users are listed on the DataFRAME Ad
25. By Address IP Network Address 10 10 10 2 IP Netmask Single Host Vv Figure 95 Editing SNMP Client from the Access Control List 3 Change the appropriate information 4 Click OK 134 MPC DataFRAME 440 Storage System Software User Manual Using SNMP Deleting Access Control Entries Delete an SNMP client from the list 1 Select a host listed in the Access Control list shown in Figure 94 2 Click Delete A confirmation message opens 3 Click OK Entering System Information Optional 1 Enter System Location information such as address building name room number etc Normally this will be network administrator information the person you would contact if you could not connect to SNMP clients 2 Enter System Contact information such as name telephone email etc Using the SNMP MIB The Storage System MIB provides read only access to the DataFRAME The SNMP implementation in the DataFRAME supports MIB II compliant objects In addition MIB files have been developed for DataFRAME specific information These files when loaded in the Network Management System allow you to see DataFRAME specific information such as model number serial number hard disk capacity network parameters RAID configuration DNS server configuration details and more See Installing the Storage System MIB on page 135 Installing the Storage System MIB The Storage System MIB files are installed when you install the Storage Sys
26. Ron Edit Feature Registration EZ License Key for Boulder 2 Cancel Figure 200 Entering License Key 298 MPC DataFRAME 440 Storage System Software User Manual Feature Registration 5 Copy and paste the appropriate license key for that DataFRAME into the window 6 Click OK The license key information is updated in the Feature Registration window as shown in Figure 201 Feature Registration Si Feature Registration Registration is required for additional software features Current feature registration license keys are displayed below For additional feature registration please contact your storage provider www mpccorp com 1 888 224 4247 and be prepared to provide the serial number s indicated below Feature Registration Scripting Evaluation Lauer rey rene see E DF140 1 ia D Serial Number 00 30 48 2E B9 D2 D License Key none set E DF 420 DN Serial Number 423JMMJOO577 D License Key none set E DF 420 1 A Serial Number 427JMMJO0768 D License Key 0302 1005 1217 B295 F098 3545 76BD v D Edit License Key Cancel Figure 201 Viewing License Keys MPC DataFRAME 440 Storage System Software User Manual 299 Feature Registration 300 MPC DataFRAME 440 Storage System Software User Manual A Using the Configuration Interface The Configuration Interface is the command line interface that uses a direct connection with the DataFRAME You may need to access
27. mpc DataFRAME 440 SAN User Manual Copyright Limited Warranty Copyright Limited Warranty The information in this document is subject to change without prior notice in order to improve reliability design and function and does not represent a com mitment on the part of the manufacturer In no event will the manufacturer or seller of an MPC product be liable for direct indirect special incidental or conse quential damages arising out of the use or inability to use the product or documentation even if advised of the possibility of such damages Except as stated in the applicable MPC Computers LLC limited war ranty MPC and its affiliates by this manual make no other express warranties THE IMPLIED WAR RANTY OF MERCHANTABILITY AND THE IMPLIED WARRANTY OF FITNESS FOR A PAR TICULAR PURPOSE ARE HEREBY DISCLAIMED IF THESE LIMITATIONS DO NOT APPLY TO YOU THEN ALL IMPLIED WARRANTIES AND CONDI TIONS OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE LIMITED IN TIME TO THE TERM OF THE LIMITED WARRANTY PERIOD REFLECTED ON YOUR PACKING SLIP OR INVOICE NO WARRANTIES WHETHER EXPRESS OR IMPLIED WILL APPLY AFTER THE LIMITED WARRANTY PERIOD HAS EXPIRED This document contains proprietary information that is protected by copyright All rights are reserved No part of this document may be photocopied repro duced or translated to another language without the prior written consent of MPC Computers LLC 906 E
28. 264 MPC DataFRAME 440 Storage System Software User Manual 15 Controlling Client Access to Volumes Access to storage volumes by application servers is controlled using a combination of volume lists and authentication groups e Volume lists provide the connection between authentication groups and volumes They are created at the management group level and they link designated volumes with the authentication groups that can access those volumes e Authentication groups identify the person or entity accessing the volume Creating Access to Volumes After you have configured storage and created volumes you then create access to the volumes using authentication groups and volume lists 1 Create an authentication group 2 Create a volume list and associate the authentication group to the specific volume s it can access Types of Client Access The Storage System Software supports three modes by which clients can access volumes through an iSCSI initiator the EBSD driver or a Fibre Channel host You can configure authentication groups to use one two or all three modes if desired When creating an authentication group you choose the mode and level of client access you want Choices are e iSCSI Authentication based on the initiator node name single hosts or CHAP based authentication single or multiple hosts e EBSD Hosts with specific subnets and masks or hosts with all subnets and masks e Fibre Channel Hosts with
29. DataFRAME with the following parameters Begin DataFRAME configuration Machine generated file use minicom s to change parameters pr port dev cuaO pu baudrate 19200 pu bits 8 pu parity N pu stopbits 1 pu mautobaud Yes pu backspace DEL pu hasdcd No pu rtscts No pu xonxoff Yes pu askdndir Yes End DataFRAME configuration 2 Start xterm as follows xterm 3 In the xterm window start minicom as follows minicom c on I DataFRAME 4 Press Enter when the terminal emulation session is established A prompt appears asking you to type start and hit enter at the login prompt 5 Type start and press Enter When the session is connected to the DataFRAME the Configuration Interface window opens shown in Figure 202 302 MPC DataFRAME 440 Storage System Software User Manual Using the Configuration Interface Logging in to the DataFRAME Once you have established a connection to the DataFRAME using a terminal emulation program log in to the Configuration Interface 1 From the Configuration Interface entry window press Enter to start the log in process The Configuration Interface Login window opens shown in Figure 203 Figure 203 Enter User Name and Password 2 Type the user name and password of the administrative user established when the DataFRAME was first configured Note This user is viewable in the Storage System Console under DataFR
30. In this scenario system resources will be exhausted Therefore if there are other volumes in the cluster accessed by other applications those volumes will hang as well even though those volumes hard thresholds have NOT been reached Managing the Volume Growth Capacity When you receive the alert that the soft threshold has been reached you take the following actions e Increase the volume size For our example above you increase the volume size by 20 percent to 600 GB e Increase the hard threshold by about 20 percent to 590 GB e Increase the soft threshold to 585 GB See Editing a Volume on page 229 for information about changing the volume size and the soft and hard thresholds Over time as you near the capacity of the cluster you can increase the storage capacity of the cluster by adding more DataFRAMEs Note If you have file systems mounted on the host volume and you reach the soft or hard threshold deleting files from the volume does not create space on the DataFRAME volume Using Auto Grow You can use auto grow to automatically increase the hard and soft volume thresholds by a specific amount Note Auto grow is also available in the application based scripting described in Chapter 14 Working with Scripting How Auto Grow Works Auto grow is triggered when a soft threshold is reached Auto grow then raises both the soft and hard thresholds by either a calculated increment Auto or by an incr
31. Karcher Road Nampa Idaho 83687 Copyright Limited Warranty Trademark Notice Limitation of Liability Notices Life Support and Nuclear Facilities Windows Windows 2000 MS DOS and Microsoft are registered trademarks of Microsoft Corporation Pentium and MMX are trademarks or registered trademarks of Intel Corporation Other product names mentioned herein are used for identification purposes only and may be trademarks and or regis tered trademarks of their respective companies While reasonable efforts have been made to ensure the accuracy of this manual the manufacturer and seller assume no liability resulting from errors or omissions in this manual or from the use of the infor mation contained herein No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means mechanical photocopying recording or otherwise without the prior written permission of the manufacturer The information within this manual is subject to change without notice The manufacturer shall not be held liable for technical or editorial errors or omis sions contained herein nor for incidental or conse quential damages resulting from the furnishing performance or use of this material Products sold by MPC are not authorized for use as critical components in medical devices or systems or in nuclear facilities without the express written approval of the Chief Executive Officer of
32. Prerequisites e At least one management group has been created e At least one cluster has been created in that management group e At least one volume has been created in that cluster Warning When associating or deleting associations for Fibre Channel LUNs the host server s Disk Management or equivalent window must be closed Requirements for Volume Lists e An authentication group can contain only one volume list e Only one authentication group should have read write access to a volume e For Fibre Channel volumes or snapshots you must assign LUN numbers when you add the volume to the volume list Planning Volume Lists Planning volume lists takes into account multiple factors e What applications or clients will access the volume e How will those applications or clients access the volume through an iSCSI initiator the EBSD driver or through a Fibre Channel host e What permissions will you assign for those clients Creating a Volume List 1 Log into the management group and select that management group in the Network View The management group Tab View opens 2 Click the Volume Lists tab to bring it to the front 3 From the Tasks menu select New Volume List The New Volume List window opens shown in Figure 185 MPC DataFRAME 440 Storage System Software User Manual 281 Controlling Client Access to Volumes New Volume List New Volume List Name Volume_List_O Description Vol
33. Status Online Active Read DP Requests 0 Write DP Requests 0 Ops o sync 0 Cycles 147 BSize 512 Capacity 419430400 kb Table 57 Parameters for proc ebsd client Parameter What It Is Status Status of the device Starting Deleted Online may be either Active or Lost Manager Read Amount of data read and the number of read requests in bytes KiB MiB or GiB Write Amount of data written and the number of write requests in bytes Ops Combined total of read requests and write requests Cycles Internal the number of times the ebsd task looped BSize Block size in bytes MPC DataFRAME 440 Storage System Software User Manual 323 Using the EBSD Driver for Linux Table 57 Parameters for proc ebsd client Parameter What It Is Capacity Size of the attached volume in kilobytes NOTE KiB MiB and GiB are calculated in increments of 1024 bytes For example 1 KiB 1024 bytes 1 MiB 1024 KiB 1 GiB 1024 MiB 2 To verify that the block devices were created use Is la dev ebsd root lnx demo root ls la dev ebsd total 0 Crwxr xr xX 1 root root 176 0 Apr 4 14 39 ebsdctrl brwxr xr x 1 root root 177 0 Apr 10 17 53 disko brwxr xr x 1 root root 177 1 Apr 10 17 53 diskl You can format the block devices by using any of the OS filesystem utilities For example you can use mkfs t ext2 dev ebsd disko Mounting the Block Device EBSD Disk On
34. System Syntax Example Windows set CLASSPATH lt full path set CLASSPATH C Program Files MPC Storage to Ul jar gt Server UNUL jar Working with Scripting 256 Table 45 Setting the Environment for Using Scripting Tools Operating System Syntax Example Unix C Shell setenv CLASSPATH lt full setenv CLASSPATH opt MPCStorageServer UI type path to Ul jar gt Ul jar Unix Bourne export CLASSPATH lt full export CLASSPATH opt MPCStorageServer UI or Kshell or Bash path to Ul jar gt Ul jar 2 Run the tool commandline CommandLine Note Run this program twice to take a snapshot of both the journaling data and the application data if you have them stored in separate volumes Table 46 Parameters for commandline CommandLine Parameter What It Is admin name Value text Name of the administrator with full administrative privileges Can be either the primary or remote administrator if they are different admin password Value text The administrator s Storage System Console password Can be either primary or remote password if they are different manager ip Value IP address IP address of a DataFRAME running a manager in the management group containing either the source volume or the remote volume volume name Value text Name of the volume For volume_autogrow and FC_LUN commands this also could be a snapshot snapshot name
35. The Configure Variable wizard Step 3 opens Configure Yariable Step 3 of 3 xj The following actions will occur when the Threshold Active value of CPU Utilization exceeds 90 3 Y Storage System Console Alert Z SNMP Trap z Email enter email addresses below separated by commas e lt Back Finish cancer Figure 104 Setting Alerts for Monitored Variables 4 For each threshold listed select the type of alert you want to receive Table 30 Types of Alerts Available for Active Monitoring Type of Alert Where Alerts Are Sent For More Information Console alerts To the Alert Message area of the See Alert Messages Tab on Console and the Alerts tab in page 29 Reporting SNMP traps To the SNMP trap community See Enabling SNMP Traps on managers You must have page 137 configured the DataFRAME to use SNMP Email To specified email addresses Type See Setting Email Notification on the email addresses to receive the page 151 notification separated by commas Then configure Email Notification on the Email tab 5 Optional To apply the alert actions including the email addresses that you selected in step 4 to all variables that are monitored on the DataFRAME select the Apply Threshold Actions to All Monitored Variables checkbox MPC DataFRAME 440 Storage System Software User Manual 147 Reporting Note To save time w
36. Warning Disassociating the management groups Cancels any in progress remote snapshots Deletes all snapshot schedules that are shared between the selected management groups 5 Click OK The Edit Management Group window opens and the remote management group you disassociated from is gone from the list 6 Click OK to return to the Network view 366 MPC DataFRAME 440 Storage System Software User Manual E Sample Remote Copy Configurations Overview Because of the flexibility provided by Remote Copy you can use the functionality in a variety of configurations that are most suitable for your requirements The sample configurations described in this chapter are only a few possible ways to use Remote Copy for business continuance backup and recovery data migration and data mining Using Remote Copy for Business Continuance Business continuance comprises both disaster recovery and high availability of data Using Remote Copy for business continuance data is stored off site and is continuously available in the event of a site or system failure Achieving High Availability Creating remote snapshots in remote locations with application based scripting can ensure that database applications such as SQL Server Oracle and Exchange have continual access to data volumes if production application servers or data volumes fail Using off site remote snapshots of your production volumes you can configure a backup application server t
37. and then cluster the DataFRAMEs Warning Once RAID is configured you cannot change the RAID configuration without deleting all data on the DataFRAME Requirements for Configuring RAID Placement of Disks in the DataFRAME If there are empty drive bays only the disks to the left of the empty drive bay will be included in RAID The remaining disks will be inactive 66 MPC DataFRAME 440 Storage System Software User Manual Storage Because RAID 1 and RAID 10 create mirrored disk pairs there must be an even number of disks in the DataFRAME If you configure RAID 1 or RAID 10 ona DataFRAME that contains an odd number of disks RAID will be configured but the odd disk will not be included in RAID For example if the DataFRAME contains 9 disks then disks 1 8 will be included in 4 disk pairs Disk 9 will be inactive If you add a 10th disk later you can add disks 9 and 10 to RAID RAID 5 and RAID 50 can only be configured on completely populated sets of disks e The MPC DataFRAME 440 Storage System must contain 4 8 12 or 16 disks Management Groups and RAID You cannot configure RAID on a DataFRAME that is already in a management group If you want to change the RAID configuration for a DataFRAME that is ina management group you must first remove it from the management group Clusters and RAID All DataFRAMEs in a cluster must have the same RAID configuration However you can have mixed versions of RAID 5 50 within a cluster
38. fault tolerant See above 4 3 High If one manager fails 3 remain so there is still a quorum 5 3 High If one or two managers fail 3 remain so there is still a quorum Communication Mode The Storage System Console and Storage System Software support unicast communication among DataFRAMEs and application servers Unicast Communication Unicast is communication between a single sender and a single receiver over a network Unicast communication allows application servers to direct messages to DataFRAME managers which are located in different subnets When you configure application servers to access storage volumes you must use the IP addresses of the DataFRAMEs that are running managers Adding or Removing Managers Any time you add or remove managers in a management group a window opens which displays all the IP addresses of those managers along with a reminder to reconfigure the application servers that are affected by the change Note Unicast requires that the DataFRAMEs running managers have static IP addresses Creating a Management Group Creating a management group is the first step in the process of creating clusters and volumes for storage Tasks included in creating a management group are MPC DataFRAME 440 Storage System Software User Manual 159 Working with Management Groups e Adding DataFRAMEs to the management group e Starting managers on selected DataFRAMEs Getting There 1 Open the
39. remember that the replication level of the volume is duplicated in the snapshot Source Volumes for Data Mining or Tape Backups or Data Preservation Before Upgrading Software Plan to use a single snapshot and delete it when you are finished Consider the following questions in your planning e Is space available on the cluster to create the snapshot e Is space available in the cluster to accommodate the increase in the volume s hard threshold when the snapshot is deleted Remember that the hard threshold will never exceed the volume size Protection Against Data Corruption Plan to use a series of snapshots deleting the oldest on a scheduled basis Consider the following questions in your planning e What is the minimum size you can set for the hard threshold that will accommodate the changes likely to occur between snapshots e Is space available on the cluster to create the snapshots e Is space available in the cluster to accommodate the increase in the volume s hard threshold when the snapshot is deleted MPC DataFRAME 440 Storage System Software User Manual 239 Working with Snapshots Creating a Snapshot Create a snapshot to preserve a version of a volume at a specific point in time 1 Log into the management group that contains the volume for which you want to create a new snapshot The management group Tab View opens 2 Select the volume on which you want to create a snapshot The volume Tab View opens shown i
40. snapshots e Configuration of application servers and backup application servers e Task flow for failover recovery resuming production after failover Using Scripting for Failover Application based scripting provides the capability for creating mounting and deleting snapshots using scripts Remote Copy can be scripted as well Remote snapshots and snapshot schedules can be created and managed using scripts Detailed information about snapshot scripting is in Chapter 14 Working with Scripting Resuming Production After Failover After failover occurs three scenarios exist for resuming production e Failback Recovery return operations to the original primary site once it is restored e Make the backup site into the new primary site e Set up a new primary site and resume operations at that site The task flow for restoring or recovering data and resuming the original Remote Copy configuration are different for each scenario 360 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Synchronizing Data After Failover After a failover there will usually be two snapshots or volumes that have conflict ing data Recovering and synchronizing such data depends on multiple factors including the application involved Example Scenario The following example illustrates only one process for synchronizing data Remember that such synchronization is optional Time Event What Happens 1 00 p m
41. troubleshoot hardware problems To run diagnostic tests 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select Reporting from the configuration categories The Reporting Window opens 3 Select the Diagnostics tab The Diagnostics window opens shown in Figure 107 edit Configuration for DF140 192 168 0 7 i x Module TP Storage d Passive LogFiles Active Email Diagnostics Alerts Time ga TOPAP Network DataFRAME Administ gt SNMP Reporting Result Mother board temperature CPU temperature CPU fan speed Mother board voltages Power supply status Hard drive status Hard drive temperature Hard drive SMART health RAID controller BBU status Start BBU Capacity test M Cgisisigisisisisisl Clear All Select All Log Out Save to File Apply Changes amp Run Tests Figure 107 Viewing the List of Diagnostics 152 MPC DataFRAME 440 Storage System Software User Manual Reporting 4 Select the diagnostic tests that you want to run The default setting is to run all tests Clear any tests that you do not want to run To clear all selections click Clear All Note Running all of the diagnostic tests will take several minutes To shorten the time required to run tests clear the checkboxes for any tests that you do not need 5 Click Run Tests A progress message
42. w e Snapshots with hard and soft Delete any primary or remote snapshots created thresholds different than volume size by the scripts You can identify these snapshots by viewing the item Created By Script on the Remote copy volumes and snapshots snapshot Details tab Automatic threshold increases Note Turning off the scripting evaluation ensures that no scripts will continue to run the 30 day evaluation clock unintentionally Registering Features and Applications When registering DataFRAMEs for add on features and applications you first submit the appropriate DataFRAME serial number s to purchase the license key s You will then receive the license key s to apply to the DataFRAME s Using License Keys License keys are assigned to individual DataFRAMEs License keys can be added to DataFRAMEs before or after they are in a management group One license key is issued per DataFRAME and that key licenses all the features requested for that DataFRAME Therefore you register each DataFRAME for which you want to use add on features and applications For example if you wanted to configure multiple node clusters in two locations to use with you would license the DataFRAMEs in both the primary location and the remote location for both the Scalability Pak and the Remote Data Protection Pak Note If you remove the DataFRAME from the management group the license key remains with that DataFRAME See the chapter on W
43. which you are restricting permissions 1 In the Edit Volume List window on the volumes tab select the volume or snapshot for which you want to edit the permissions 2 Click Edit Permission Level The Edit Volume in Volume List window opens shown in Figure 190 Edit Yolume in Yolume List Edit Volume in Volume List Volume_List_0 Volume Name Permission Level O No Access Read Access Read and Write Access Fibre Channel LUN Cancel Figure 190 Editing Permissions on a Volume 3 Change the permission level 4 Click OK when you are finished The Edit Volume List window opens 5 Click OK when you are finished editing the volume list The Tab View opens Select the Volume Lists tab shown in Figure 191 to review your changes MPC DataFRAME 440 Storage System Software User Manual 285 Controlling Client Access to Volumes Changing Authentication Groups in a Volume List Removing authentication groups from a volume list Note For Fibre Channel associations you can also change the LUN number if necessary e Prevents the associated application servers from accessing the volumes in that list e Is a prerequisite for deleting the volume list Warning Before editing the volume list stop any applications from accessing volumes for which you are restricting permissions or removing authentication groups 1 In the Edit Volume List window shown in Figure 189 select the Authentication Gr
44. 1 On the Disk Setup tab select the old disk and click Power Off Disk 2 Replace the disk in the DataFRAME 3 On the Disk Setup tab select the new disk and click Power On Disk 4 When disk is powered on select the disk and click Add to RAID Adding Disks to the DataFRAME If the DataFRAME is configured for RAID 1 or RAID 10 you must add an even number of disks to include all the disks in the RAID configuration See Requirements for Configuring RAID on page 66 If the DataFRAME is configured for RAID 5 or RAID 50 you must add disks in complete sets as follows MPC DataFRAME 440 Storage System Software User Manual 75 Storage e MPC DataFRAME 440 Storage System 4 or 8 disks at a time Diagrams of Disk Bays Figure 53 illustrates the placement of the drive bays in the MPC DataFRAME 440 Storage System Figure 53 Diagram of the Drive Bays in the MPC DataFRAME 440 Storage System Adding Disks and DataFRAME Capacity If you are using clustering all DataFRAMEs in a cluster will operate at a capacity equal to that of the smallest capacity DataFRAME Adding capacity to all DataFRAMEs in the cluster will prevent stranded storage Note You must add disks in contiguous disk bays from left to right Note You cannot reduce the capacity of a DataFRAME t
45. 192 168 0 40 Module 13 Storage Time TCP IP Network RAID Setup Disk Setup RAID Status Normal Start RAID Sn DataFRAME Administ Se SNMP Reporting RAID Configuration RAID 0 x Reconfigure RAID The RAID Rebuild Rate is a percentage of RAID card throughput U 20 30 40 50 60 Percent t TR 1 1 10 70 Device Type Subdevices Disk 0 Disk Disk Disk Disk Disk Disk Disk Disk Device Name dewscsi host0 busOitargetO lunO disc dewscsi host0 busOitargett lunOidisc dewscsi host0 busOitarget2 lunOidisc dewscsi host0 busOitarget3 lunDidisc dewscsi host0 busOttarget4 lunO disc dewscsi host0 busOitarget5 unOidisc dewscsi host0 busOitarget6 lunOidisc dewscsi host0 busOitarget8 lunDidisc dewscsi host1 busOstargetO lunO disc 27 Click on Done 28 From the Help menu select Wizards gt Wizard Launch Pad 10 MPC DataFRAME 440 Storage System Software User Manual Getting Started 29 The installed storage system should be listed by host name in the main window area of the Console The Wizard Launch Pad should also be displayed Select Create Volume Wizard to create a volume Wizard Launch Pad E Welcome to the Wizard Launch Pad The Wizard Launch Pad provides wizards for common tasks to get you started Find DataFRAMEs Wizard The Find DataFRAMEs Vizard finds DataFRAMEs on your network Create Volume Wizard The Create Volume Wizard creat
46. 21 Editing TCP Speed Duplex and Frame Size 0 eee eeceeeceeseeeeeeeeeeeeeaeeseeeeeaeeeeeeeeaeeseaeeeaeeseeeeseeseeeeeeenaas 103 Table 22 DataFRAME Network Interface Settings AA 108 Table 23 Data FRAME Route Gettinge AAA 108 Table 24 EBSD Host Network Interface Settings ecceeeeeseteneeeeeeeeeeceeeseeeeeeeeeaeeeaeeseeeesaeseaeeeeeseeeesieeeseeseaeeeaes 109 Table 25 EBSD Host Route SettingSis acccciicsncsccde once aie din tere deveined on tenia headed 109 Table 26 Using Default Administrative Groups 120 Table 27 Administrative Group Name PHeouirements nne 121 Table 28 Descriptions of Group Permissions cccecceeseeeeeeeeeeeeaeeceeeeseeeeeaeeeaeeseaeesaeeseeeeseaeeeaeesaeseaeeeieeseaeeneeeeaaes 123 Table 29 Selected Details of the Passive Report A 140 Table 30 Types of Alerts Available for Active Monitoring 147 Table 31 List of Variables Available for Active Monttoring 149 Table 32 Typical Network e EE 168 Table 33 Disk Space use Reported on Disk Usage Tab 200 Table 34 Setting a Replication Level for a Volume 215 Table 35 Parameters for Volumes Table 36 Progression of Increments in Manual Auto Grow Setting Of 50 MP 223 Table 37 Progression of Increments in Automatic Auto Grow cccccccecccceeeeeeeeseeeeeeaeeeseaeeescaeeeeseeeeeseneeessaeeessaees 224 Table 38 Requirements for Changing Volume PDarameters 229 Table 39 Snapshot Parameters tnnt tnEEAAEEEASEEEEEEEEEEEEEEEEAASEEEEEEEEEEEEEEEEEASE EAE
47. 218 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Table 35 Parameters for Volumes Volume Parameter Configurable in Volume Type What it Means Hard Threshold Primary The amount of physical space allocated for actual data storage Reaching the hard threshold triggers an alert and data can no longer be written to the volume The hard threshold must be less than or equal to the volume size Remote volumes contain no data and do not have a size Therefore you cannot set a hard threshold for a remote volume Soft Threshold Primary The amount of space used on the volume that triggers a warning alert This alert notifies the storage administrator that the volume is approaching the hard threshold The soft threshold must be less than or equal to the hard threshold Because remote volumes have no size and cannot have a hard threshold they also cannot have a soft threshold Auto Grow Any Auto grow automatically increases the hard and soft volume thresholds by a specific amount Automatic auto grow uses a predetermined formula e Manually setting auto grow allows the user to determine the size increment by which the hard and soft thresholds increase Checksum Any Whether to use checksumming to verify data transmission Volume checksumming is in addition to standard IP and ethernet checksumming Enabling checksumming for a volume increases data integr
48. 269 Controlling Client Access to Volumes Client Access and EBSD Clients accessing volumes using EBSD can be restricted via specific subnets and masks or they can be unrestricted When configuring client access using EBSD you create an authentication group that allows EBSD access The New Authentication Group window with the EBSD tab is shown in Figure 172 New Authentication Group New Authentication Group Name CreditTransAG Description Volume List No Volume List iSCSI EBSD Fibre Channel Ji Allow access via EBSD Authentication Hosts with the following subnets and masks Subnet Edit Delete Hosts with all subnets and masks Figure 172 Creating a New Authentication Group for EBSD Access Planning EBSD access requires planning how you want to configure authentication e Only hosts which map to designated subnets and masks e All hosts 270 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Client Access and Fibre Channel MPC DataFRAME 440 Storage System only Storage volumes can also be configured for Fibre Channel access Use the Fibre Channel tab in the New Authentication Group window as shown in Figure 173 New Authentication Group New Authentication Group Name AuthGroup_0 Description Volume List No Volume List iscsi EBSD Fibre Channel
49. 30 35 40 Volumes Alloc 42 56 GB Used 2 8 GB i Full 6 d Snapshots es Snapshots D Free Space Figure 142 Disk Space Allocated and Used for the Cluster and its Volumes and Snapshots iSCSI Tab Displays the virtual IP address if there is one and lists any iSNS servers configured for the cluster Remote Snapshot You must purchase the Remote Data Protection Pak to use remote snapshots beyond the 30 day evaluation period The Remote Snapshot tab lists details about any remote snapshots that have been created Detailed information about remote snapshots is available in the Remote Copy appendix See Viewing a List of Remote Snapshots on page 345 Editing a Cluster When editing a cluster you can change the description add or remove DataFRAMEs and change the hot spare designation of a DataFRAME You can also edit or remove the virtual IP and iSNS servers associated with the cluster MPC DataFRAME 440 Storage System Software User Manual 201 Working with Clusters Prerequisite You must log in to the management group before you can edit any clusters within that group Getting There 1 Select the cluster you want to edit 2 From the Tasks menu on the Details tab select Edit Cluster The Edit Cluster window opens shown in Figure 143 Edit Cluster E ES Edit Cluster General iscsi Cluster Name Description I DataFRAMEs Available DataFRAMEs in Cluster DataFRAME Ip RA
50. 832 bytes VO DFMC Installer Install Cornplate MPC DataFRAME 440 Storage System Software User Manual Getting Started 14 Click on Done once the software completely loads E DFMC Installer The DataFRAME Management Console has been successfully installed to C Program Files MPC DFMC Press Done to quit the installer Caricel Previous i 15 Close out of the browser window of the Drivers CD If you selected Yes to run the Console the Wizard Launch Pad will display Select Find DataFRAMEs Wizard from the Wizard Launch Pad Note Ensure that you have set up an IP address on your storage system before you run the Find DataFRAMEs Wizara Instructions for setting up an IP address and password are available from the MPC Storage System Quick Start User s Guide that shipped with your storage system Wizard Launch Pad 4 xj Welcome to the Wizard Launch Pad The Wizard Launch Pad provides wizards for common tasks to get you started Find DataFRAMEs Wizard The Find DataFRAMEs Wizard finds DataFRAMEs on your network Create Volume Wizard The Create Volume Wizard creates a new volume Gg Access Volume Wizard The Access Volume Wizard creates access privileges for a volume E C Show Launch Dad on startup This Wizard Launch Pad and the wizards are available via the Help menu MPC DataFRAME 440 Storage System Software User Manual 5 Getting Started
51. Across a RAID 5 Set Using Four Disks Parity allows the DataFRAME to use more disk capacity for data storage than RAID 10 allows RAID 50 RAID 50 combines the redundancy of parity within a RAID set with striping across RAID sets Viewing the RAID Setup Report In the Storage category the RAID Setup tab lists the RAID disks in the DataFRAME and provides information about them The RAID Setup Report is shown in Figure 45 The following table describes the information listed in the report edit configuration for DF440 1 192 168 0 40 xj Module JI RAID Setup Disk Setup CT ki RAID Status Normal Start RAID RAID Configuration RAID 0 gt TCP IP Network am DataF RAME Adminis ge SNMP E Reporting Reconfigure RAID The RAID Rebuild Rate is a percentage of RAID card throughput e Percent tepenedve EN et ee 1 10 20 30 40 50 60 70 80 90 100 Apply Device Name Device Type Subdevices dewscsi host0 busOstargetO unOidisc Disk D a dewscsi host0 busOstargett unO disc Disk D deviscsifhost0 busO target2 unOidisc Disk 0 dewscsi host0 busOftarget3 unOidise Disk D v dewscsi host0 busOstarget4 unOidisc Disk D deviscsi hast0 busOitarget5 unOldise Disk 0 dewscsi hast0 busOftarget6 unOidise Disk 0 dewscsiihostO busOtarget8 lun0 disc Disk D Log out dewscsi host busOstargetO unO dise Disk 0 Figure 45 Viewing the RAID Setup Report MPC DataFRAME
52. Add the name of the DNS domain in which the DataFRAME resides 1 On the DNS tab type the DNS domain name 2 Click Apply when you are finished Adding a DNS Server Add up to three DNS servers for use with the DataFRAME 1 Click Add in the DNS Server panel The Add IP Address dialog opens Type the IP address for the DNS server Click OK Repeat steps 1 through 3 to add up to three servers a A O N Use the arrows on the DNS Server panel to order the servers The servers will be accessed in the order they appear in the list 6 Click Apply when you are finished Adding Domain Names to the DNS Suffixes 104 Add up to six domain names to the DNS suffix list also known as the look up zone The DataFRAME searches the suffixes first and then uses the DNS server to resolve host names 1 Click Add in the DNS Suffixes panel MPC DataFRAME 440 Storage System Software User Manual Managing the Network The Add DNS Suffix window opens 2 Type the DNS suffix name Use the domain name format 3 Click OK 4 Repeat steps 1 through 3 to add up to six domain names 5 Click Apply when you are finished Editing a DNS Server Change the IP address for a DNS Server in the list 1 Select the server to edit 2 Click Edit The Edit IP Address window opens 3 Type the new IP address for the DNS server 4 Click OK 5 Click Apply when you are finished Editing a Domain Name in the DNS Suffixes List Change a domain na
53. Block System The Windows file system treats this block device as simply another hard drive that is it is an array of blocks which the file system can use for storing data As the iSCSI initiator or the EBSD driver passes writes from the file system the Storage System Software simply writes those blocks into the volume So when you look at the Console the allocation percentage displayed is based on how many physical blocks have been written for this volume Now when you delete a file typically the file system updates the directory information which removes that file Then the file system notes that the blocks which that file previously occupied are now freed up Subsequently when you query the file system about how much free space is available the space occupied by the deleted files appears as part of the free space since the file system knows it can overwrite that space However the file system does not inform the block device underneath the Storage System Software volume that there is freed up space In fact no mechanism exists to transmit that information There is no SCSI command which says block 198646 can be safely forgotten at the block device level there are only reads and writes So to ensure that our iSCSI and EBSD network block devices work correctly with file systems any time a block is written to that block is forever marked as allocated Then when all blocks are allocated up to the full size of the storage volume
54. Console If you have not created a management group but you have some DataFRAMEs on the network the Console displays those DataFRAMEs as available See Figure 110 Gimec Computers DataFRAME Management Console oj x File Find Tasks View Help Available DataFRAMEs Available DataFRAMEs before creating a manage ment group DF420 1 Details Management Group Information Availability Host Name DF140 RAID Status Normal IP Address 192 168 0 7 RAID Configuration RAID 0 Model DF 100 Series Storage Mode block SerialNumber 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Figure 110 Viewing DataFRAMEs Before Creating a Management Group 2 Log in to one or more of the DataFRAMEs you want to add to the new management group 160 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups 3 Click Done from the DataFRAME Edit Configuration window to return to the main Console window with the DataFRAME tab view shown in Figure 111 Gc computers DataFRAME Management Console File Find Tasks View Help a Available DataFRAMES iol x DF140 DF140 1 DF420 DF 420 1 f Details Management Group ation Availability Host Name DF140 RAID Status IP Address 192 168 0 7 RAID Configuration RAID 0 Model DF 100 Series Storage Mode block Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Figure 111 DataFRAME Tab Adding the
55. DataFRAME Date and Time 11 19 2005 01 41 31 AM 2 Tee Figure 76 Setting the DataFRAME Date and Time 3 Change the date and time to the correct date and time for that time zone Type the time directly in the field Type the date in the field or use the calendar button to select the date 4 Click Apply MPC DataFRAME 440 Storage System Software User Manual 115 Setting the Date and Time Using NTP You can use NTP to manage the time for the DataFRAME NTP updates are fixed at 5 minute intervals You still must set the time zone for the DataFRAME See Setting the DataFRAME Time Zone on page 114 1 Select On in the NTP Mode area The Add NTP Server dialog opens shown in Figure 77 Adan server NTP Server Preferred Not Preferred Figure 77 Adding an NTP Server 2 Type the IP address of the NTP server you want to use 3 Click whether you want the NTP server to be designated preferred or not preferred Note A preferred NTP server is one that is more reliable such as a server that is ona local network An NTP server on a local network would have a reliable and fast connection to the DataFRAME Not preferred designates an NTP server to be used as a back up if a preferred NTP server is not available An NTP server that is not preferred might be located further away and have a less reliable connection 116 MPC DataFRAME 440 Storage System Software User Ma
56. DataFRAME is down Stopping a Virtual Manager When the situation requiring the virtual manager is resolved either the down site recovers or the communication link is restored you must stop the virtual manager Stopping the virtual manager returns the management group to a fault tolerant configuration 1 Select the DataFRAME that is running the virtual manager 2 From the Tasks menu on either the Details tab or the Management Group Information tab select Stop Virtual Manager A confirmation message appears 3 Click OK The virtual manager is stopped However it remains part of the management group and part of the quorum Removing a Virtual Manager You can remove the virtual manager from the management group altogether 1 Select the management group that has the virtual manager 2 From the Tasks menu on the Details tab select Add or Delete Virtual Manager A confirmation message opens 3 Click OK The virtual manager is removed Note The Console will not allow you to delete a manager or virtual manager if that deletion causes a loss of quorum 190 MPC DataFRAME 440 Storage System Software User Manual 11 Working with Clusters Within a management group you create sub groups of DataFRAMEs called clusters A cluster is a grouping of DataFRAMEs from which you create volumes Think of a cluster as a pool of storage You add storage to the pool by adding DataFRAMEs You then carve volumes out of the pool Volumes se
57. Disk D dewscsiihostO busOtarget8 lun0 disc Disk D dewscsithostt busOitargetO lunD disc Disk D Fei Figure 42 Viewing the Storage Configuration Category Storage Configuring and Managing RAID Managing the RAID settings of a DataFRAME includes e Choosing the right RAID configuration for your storage needs e Setting the RAID configuration e Setting the rate for rebuilding RAID e Monitoring the RAID status for the DataFRAME e Starting or reconfiguring RAID when necessary Note The DataFRAME comes configured for RAID 5 Benefits of RAID RAID combines several physical disks into a larger virtual disk This larger virtual disk can be configured to improve both read write performance and data reliability for the module RAID Configurations Defined The RAID configuration you choose depends upon how you plan to use the DataFRAME The DataFRAME can be configured with RAID 0 RAID 1 10 or RAID 5 50 The factory default setting is RAID 5 Number of Disks and RAID The number of disks in the DataFRAME affects the RAID configurations available as illustrated in the following table Number of disks required RAID Level MPC BEE ER Storage RAID 0 from 1 to 16 RAID 1 10 2 4 6 8 up to 16 RAID 5 50 4 8 12 or 16 58 MPC DataFRAME 440 Storage System Software User Manual Storage RAID 0 RAID 0 is available for any number of disks in a DataFRAME RAID 1 10 RAID 1 10 requires pairs of disks
58. Editing an Authentication Group 280 Creating a New Volume Uert 282 Adding a Volume to a Volume List 0 0 eeeeeeeneeeeenteeeeneeeeeeeeeeeeaeeetaaeeeeesaeeeesaaeeseaeeeseaeeeeeneeeseneeeeeneeeenas 282 Connecting Authentication Groups to a Volume List 283 Viewing the New Volume List ii20 ciAs dk inteviid eendi sine EENS ie dened 284 Opening a Volume List to Ed 285 Editing Permissions ona Volume uk EERdeEeE ee SEEELEEEEEREEEEENNEEEC a Na 285 Viewing the Edited Volume List 286 Volume Lists Tabs enges eege geed geed 287 Verifying the Start of the 30 day Evaluation Period 289 Evaluation Period Countdown on Register Tab 290 Evaluation Period Countdown Message 290 Icons Indicating License Status for Features 291 Enabling Scripting Evaluatton atn tt Ent EnntEntEnsenst Ennen ennt 295 Registering Features and Appltcattons A 297 Opening the Feature Registration Window nenen nennen nne 298 Entering License TEE 298 Viewing License KeyS c c c0 vcseaeed deeeisl Rains aati eet edie eed erin a eke 299 Opening the Configuration Interface 302 Enter User Name and Password scented diate caci NEEE E EEGEN 303 Configuration Interface Main Men 303 General Settings Window ssssseeseeseeeeeretesesretttnetnetttttetnntntttnttntttstnttnatusttattattnaenateatenntnnennenntnne nannat 304 Selecting an Interface to Configure 0 ceceeeceeeeeeeeeeteeeeeeeeeaeeeaeeseneeteeeeaaeeesaeseaeeseeeeeaeeeeeseaeesseeeeeeensaes 304 Entering the Host
59. FT 10 0 7 2 Device flash 1 Status Active Value chanued Figure 108 Alert Messages Tab on Console Main Window If the Console is not open these alerts are still logged and you can view them in the Reporting category of the DataFRAME Configuration window the next time you open the Console Note The Alerts tab in the Reporting category displays the most recent alerts up until the alert list reaches 1 MB in size To view alerts older than those displayed on the Alerts tab save the Alerts log on the Log Files tab 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select Reporting from the configuration categories The Reporting window opens 3 Select the Alerts tab The Alerts window opens shown in Figure 109 GBedit configuration for DF140 192 168 0 7 x Module Passive Logfiles Active Email Diagnostics Alerts WW Storage Time TCP IP Network fae ___ Alerts File Alerts log W Sn DataFRAME Administ Nov 18 02 47 32 DF140 hydra Drive 1 Health Status unknown Value does not equal nor ga SNMP Nov 18 02 47 33 DF140 hydra BBU Capacity Test Overdue on Card 1 overdue Hi Reporting Nov 18 02 48 32 DF140 hydra Cache Battery 1 Status weak Value changed Nov 18 02 49 01 DF140 hydra RAID Status normal Value changed Nov 18 02 49 31 DF140 hydra Drive 1 Status On and secured Value changed Nov 19 00 00 00 DF1
60. First DataFRAME to Create a New Management Group 1 Select the first DataFRAME to include in the management group 2 Click the Management Group Information tab in the Tab View shown in Figure 112 Details Management Group Information Management Group NA Manager No Hot Spare NA Virtual Manager No Migrating Data Status RAID Normal Figure 112 Management Group Information Tab MPC DataFRAME 440 Storage System Software User Manual 161 Working with Management Groups 3 Click the Tasks button and select Add to New or Current Management Group The Add to or create a Management Group window opens shown in Figure 113 Add to or Create a Management Group xi Select a management group for DataFRAME DF 140 192 168 0 7 to join or create a new management group New Management Group Group Name Group D Cancel Figure 113 Creating a New Management Group 4 Select New Management Group and type a name for the management group 5 Click Add The Managers IP Addresses window opens shown in Figure 114 e Manager IP Addresses E For each host that is using EBSD enter the following manager IP addresses in the driver dialog box used to create and edit volumes 192 168 0 7 ox Save Manager IP Addresses Figure 114 List of Manager IP Addresses for Management Group 162 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups 6 Click O
61. General iSCSI Cluster Name Description DataFRAMEs Available DataFRAMEs in Cluster DataFRAME E Pap taF RAM IP RAID 192 168 0 9 192 168 0 9 Not Availab agg DF140 1 192 168 RAID 0 DF140 192 168 0 7 RAIDO DF420 192 168 0 3 RAID 0 lt Add lt Add All La Z move gt KS SE Remove All Hot Spare None v Hot Spare Time Out 0 Seconds _Cancel_ Figure 147 Returning the Repaired DataFRAME to the Cluster 9 Remove the ghost DataFRAME 10 Select the repaired DataFRAME and add it to the cluster It will be in the place reserved by the ghost DataFRAME 11 Add any remaining DataFRAMEs to the cluster 12 Click OK The DataFRAMEs are in the cluster in their original order The ghost DataFRAME is removed from the cluster 13 Select the ghost DataFRAME and remove it from the management group A confirmation message opens warning that the DataFRAME cannot be found on the network 14 Click OK to confirm removing DataFRAME from the management group Another confirmation message opens 15 Click OK The ghost DataFRAME disappears from the Console 208 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters Deleting a Cluster Volumes and snapshots must be deleted or moved to a different cluster before you can delete the cluster For more information see Deleting a Volume on page 233 and Deleting a Snapshot on page 253 Prerequisite You must log in to the management
62. Getting R LEE 160 Adding the First DataFRAME to Create a New Management Group ceeceeeseeeeeeeeeeeeeeeeeeeteaeeeaeeeeeeesas 161 Adding Additional DataFRAMEs When Creating a Management Group ceeeeeeeeeseeeeeeeeeseeeeeeneeeees 163 viii MPC DataFRAME 440 Storage System Software User Manual Contents Adding Managers to the Management Group ecceeeceeeeeeeneeeeeeeeeeeeeeeeneeeaeeeeaeesaeeseeeesaeeeaeesaeseaeeseaeesaeeeeeeeeaees 163 Logging In to a Management Group eeeeecceeeeeeeeeeeeeeeneeeeeeesaeeeaeeeaeeseaeeeaeeeeaeeseaeeeaeesaeesaeeseeeeseaeseeeneeesieeeeeenaas 164 Management Group Tab View 170 eine an E AEAN E AE EEEE esa nha 166 Editing a Management Group c ceccceeseseeeeeeeeeeeeeeaeeeseeeeeeeceaeeeaeeeaaeeeaeesaeeecaeeseaeseaeesaeeeaeeseeeeeaeeeneeeeeesieeeeaeenaas 168 Setting or Changing the Local Bandwidth AA 168 Logging Out of a Management Group ceceeeceeeeeeeneeeeeecneeeeaeeeeeseaeeeaaeeeaeeseeeeeaeeeaaeseaeeseeesaeeseaeeeeesiaeseaeeseeeesas 170 Adding a DataFRAME to an Existing Management Group eecceseceeeeeeeneeeeeeeeneeeeeeeseneeeaeeeaeeeeaeeeeeseaeeneaeeaees 170 Adding Manager IP Addresses to Application Servers cecccecceeeseeneeeeneeeeeeeeeeeseeeeeaeeseaeseaeeseeeseeeeeaeeeaets 171 Resetting the Management Group Time 0 ceeeeeceeececeneeeeeeeeeeeeeaeeeaeecsaeecaeeeeaeeseaeeeaeseaaeseaeeseeeeesaeseaeeeaeesieeeeaeenias 171 Starting and Stopping Managers ee
63. Lists the frame size setting for the device Preferred For active backup bonds Indicates whether the device is set as preferred The preferred interface is the interface within an active backup bond that is used for data transfer during normal operation Editing the TCP Speed and Duplex You can change the speed and duplex of the 1000BASE T TCP interfaces Requirements e f you change these settings you must ensure that BOTH sides of the NIC cable are configured in the same manner For example if the DataFRAME is set for Auto Auto the switch must be set the same e f you edit the speed or duplex on a disabled or failed NIC the new setting will not be applied until the NIC is enabled or connectivity is restored Best Practice Change the speed and duplex settings while the DataFRAME is in the Available mode and not in a management group Table 19 Setting DataFRAME Speed and Duplex Settings DataFRAME Setting Switch Setting Speed Duplex Speed Duplex Auto Auto Auto Auto 1000 Full 1000 Full 100 Full 100 Full 100 Half 100 Half 10 Full 10 Full 10 Half 10 Half 1 On the TCP Status tab select the interface you want to edit 2 Click Edit The Edit Speed and Duplex window opens shown in Figure 69 100 MPC DataFRAME 440 Storage System Software User Manual Managing the Network Edit Speed Duplex and Frame Size E Name Motherboard Port0 Speed Duplex Speed Auto Auto Duplex
64. MPC EMA001082 KDA 02 03 2006 Copyright Limited Warranty Contents Contents Chapter i Copyright Limited Warranty ccccssecccseeeeeeeeeseeeeeeeeeneeeeneeeeeeesenneeeeneeeneeeeseeeeees ii Trademark e iii Limitation of BC UE ili Eet iii Life Support and Nuclear Facilities A iii Chapter 1 Getting Started inasscsisicc cents sccscecasccntescsscnceccansdancnactencesenacacnsaesuanccvenecnsneaenascccssnaaseucere 1 IS Glen er 1 Configuration TASKS eccecccesseeeseeseeeeeneeseeeeeaeecseeeeaeescaeeeaeeesaeesaeeseeeeeseeseaeeeaeecaeeeseeescaeesaeseaeseaeeseaeeseeeseaeeseeeeneeeaas 14 VIE DEE 15 Find Data FRAMES Wizard EE 15 Create Volume Wizard 220 EE deed E E AE E EE on eee 15 Access Volume Wizard EE 15 Console Main WiINdOW EE 16 Icons Used in the Storage System Console eeceeeceeeeeeeeeeeeeeeeeeeeeaeeeaeeseaeeeaeeseaeeeaeeeeeseaeesneeesieeseaeeeneeee 17 Finding Storage System Modules on the Network AA 19 Finding by Subnet and Mask cceecceesceeseeeeeeeeeeeseeeeeeeseaeeeaeeseeeeseeeeeaeesaeeseaeeeeeeeeaeeseeeseaeeteeseeeesieeseaeeeneene 21 Finding by Module IP or Host Name eeeseeeeneeeeeneeeeeneeeeeaneeeeaeeeeesaeeeeeaaeeseaeeeseaeeseeeeeeseeeeeeeaeeeseeeeeeeneeees 23 Using the Network View ageseent a eaa aa E TEE ee ce idee eee 26 inter CEET 28 Viewing Storage System Module Details AAA 30 D tails KE 31 Management Group Information Tab eecceecceesceeeseeeeeeeeneeeeeeenaeeeaeeeeae
65. New Remote Snapshot New Remote Snapshot Schedule 234 MPC DataFRAME 440 Storage System Software User Manual 13 Working with Snapshots Snapshots provide a fixed version of a volume for use with backup and other applications Snapshots vs Backups Unlike backups which are typically stored on different physical devices or tapes snapshots are stored on the same cluster as the volume Therefore snapshots protect against data corruption but not device or storage media failure Prerequisites Before you create a snapshot you must have created e a management group e a cluster e a volume Topics covered in this chapter include e Single snapshots and scheduled snapshots e Managing capacity using volume and snapshot thresholds e Creating snapshot schedules Using Snapshots You create snapshots from a volume on the cluster At any time you can roll back to a specific snapshot When you do roll back all the snapshots created after that snapshot are deleted Also using a third party utility you can copy a snapshot to a different server and open the snapshot as a volume on that server Snapshots can be used for e Source volumes for data mining and other data use e Source volumes for creating backups e Data or file system preservation before upgrading software e Protection against data or file system corruption e File level restore without tape or backup software Working with Snapshots Single Snapshots versus Schedule
66. Registration is required for additional software features Current feature registration license keys are displayed below For additional feature registration please contact your storage provider www mpccorp com 1 888 224 4247 and be prepared to provide the serial number s indicated below Feature Registration Scripting Evaluation EJ DF140 i A Serial Number 00 30 48 2E BA 5A DN License Key none set E DF140 1 D Serial Number 00 30 48 2E B9 D2 DN License Key none set elo EM e D Serial Number 423JMMJ00577 D License Key none set J gt D Edit License Key Cancel Figure 199 Opening the Feature Registration Window 4 For each DataFRAME listed in the window that you want to register submit the serial number as instructed in the Feature Registration window Control C copies the serial number so that you can paste it into an application such as Notepad or Word Note Record the host name or IP address of the DataFRAME along with the serial number This record will make it easier to add the license key to the correct DataFRAME when you receive it Entering License Keys When you receive the license keys add them to the DataFRAMEs in the Feature Registration window Select the management group Select the Register tab From the Tasks menu select Feature Registration Select a DataFRAME and click Edit License Key The Edit Feature Registration window opens shown in Figure 200
67. Regular hourly scheduled remote snapshot RemoteSS_0 created in Remote Management Group 1 10 p m Remote copy finishes Copying is complete 1 30 p m Primary volume goes offline OrigPrimaryVol_0 offline 1 33 p m Scripted failover causes remote volume to ActPrimaryVol_0 active in Remote become the acting primary volume Management Group 2 00 p m Original primary volume comes back online OrigPrimaryVol_0 online e Original volume contains data from 1 00 to 1 30 p m e Acting primary volume contains data from 1 33 to 2 00 p m Returning Operations to Original Primary Site Once the original primary site is operational again restore operations to that site The steps to restore operations depend upon the state of the original primary volume e If the primary volume is working synchronize the data between the acting primary volume and the restored primary volume before returning the acting primary volume to its remote volume state e If the primary volume is not available create a new primary volume synchronize the data with the acting primary volume and then return the acting primary volume to a remote volume Synchronizing the Data Between the Acting Primary Volume and the Original Primary Volume 1 Create Snapshots of Data First you create snapshots that contain the data that you need to synchronize The steps to create those snapshots are described Table 63 MPC DataFRAME 440 Storage System Software User Manual 361 Re
68. System Software User Manual Storage DataFRAME 1 72 73 4 7 7 16 disks Mirrored RAID d y y disk pairs Figure 47 RAID 10 on an MPC DataFRAME 440 Storage System Devices Configured in RAID 5 50 DataFRAME T 2 3 4 ue le disks Mirrored RAID d y Y disk pairs If RAID 5 or 50 is configured the physical disks are grouped into sets RAID 5 uses one set of disks RAID 50 uses multiple sets of disks in each DataFRAME RAID 50 in the MPC DataFRAME 440 Storage System RAID 50 in the MPC DataFRAME 440 Storage System consists of multiple RAID 5 sets using either all the disks as in 4 disk or 8 disk sets shown in Figure 48 or n 1 disks so that the single disk acts as a hot spare for the RAID set as shown in Figure 49 The RAID 50 n 7 configurations are 3 disks plus a spare and 7 disks plus a spare AME DataFR all 4 St TBST IS IA HIBS 16 disks RAID 5 i y y y sets Figure 48 MPC DataFRAME 440 Storage System RAID 50 Using 4 Disk Sets o 17 2 S67 he db TI sale HS CT E s uns Y y y Y sets Figure 49 MPC DataFRAME 440 Storage System RAID 50 Using 3 Disks Plus a Hot Spare
69. Time Tcpap TCP Status DNS Routing Communication SZ DataFRAME Adminis Name Description MAC Mode IP Subnet Gateway Type gt SNMP bond Logical F 00 30 48 Static IP 192 168 255 255 192 168 active ba fe Reporting Mothe Intel Cor 00 30 48 Slave 192 168 255 255 192 168 NIC Mothe Intel Cor 00 30 48 Slave 192 168 255 255 192 168 NIC Bonded logical Physical interfaces network interface shown as slaves Loy Out it Delete Bond Figure 65 Viewing a New Active Backup Bond The bond interface shows as bond and has a static IP address The two physical NICs now show up as slaves in the Mode column 14 Optional for active backup bonds To change which interface is the preferred interface in an active backup bond on the TCP Status tab select one of the NICs in the bond and click Set Preferred Viewing the Status of a NIC Bond You can view the status of the interfaces on the TCP Status tab Notice that in the active backup bond one of the NICs is the preferred NIC In the NIC aggregation bond neither physical interface is preferred Figure 66 shows the status of interfaces in an active backup bond Figure 67 shows the status of interfaces in a NIC aggregation bond ER Configuration for DF140 192 168 0 7 x Module d TePap TCP Statu
70. Version Build Number Kernel Version src rpm to install the appropriate RPM package This installs the sources in ust src redhat SOURCES for Redhat Systems usr src packages SOURCES in SuSE distributions 320 MPC DataFRAME 440 Storage System Software User Manual Using the EBSD Driver for Linux Building the Driver for the New Kernel Version 4 Navigate to the SOURCES directory Run mkdir EBSD oN OA Run configure 9 Run make all 10 Run make install 11 Run depmod Navigate to SOURCES EBSD Run tar xvzf ebsd 6 x xx xxxx tar gz The driver is installed in lib modules uname r kernel drivers addon ebsd Configuring the EBSD Driver for Linux Once the EBSD driver is installed on Linux it must be configured and started Creating ebsd conf 1 Copy opt Storage_System Storage_System_Software 6 x Drivers EBSD uname r etc ebsd conf sample etc ebsd conf 2 Modify the etc ebsd conf file Add a device entry in ebsd conf for each volume or snapshot that has been configured on the DataFRAME Required parameters are listed in Table 56 Table 56 Parameters in ebsd conf Parameter What It Is device This is the device section identifier It must be named device where is the device number Corresponding block device is created as dev ebsd disk Valid device numbers 0 to 63 type volume snapshot EBSD device types is a volume or snapsh
71. Volume Length Figure 149 Up to Ten Automatic Increments for Auto Grow Benefits of Automatic Auto Grow Some of the benefits of using the automatic auto grow with its incremental algorithm It allows small volumes to grow as little as possible while still allowing for growth to very large volume sizes without too many auto grow operations e t maintains the ratio of thresholds In the following examples when using the manual setting for auto grow the hard threshold grows to 510 MB and the soft threshold grows to 495 MB The fact that these thresholds are so close may MPC DataFRAME 440 Storage System Software User Manual 221 Working with Volumes allow clients to hit the hard threshold and stall for several seconds before auto grow kicks in In contrast when the automatic algorithm reaches 512 MB for the hard threshold the soft threshold is 384 MB which would trigger the next auto grow well before it is needed 222 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Auto Grow Examples The following examples illustrate the difference between automatic and manual auto grow For this example the volume parameters are e Length 1 GB e Hard Threshold 60 MB e Soft Threshold 45 MB Manual Auto Grow Auto grow is set for Manual to increase the soft and hard thresholds by 50 MB whenever the soft threshold is reached The rate of increase is illustrated belowin Table 36 and Figure 150 Tab
72. Workstations Clustered Set pplication Application Serwer Servers Read Write Read Only Read Write Primary g Primary i Volume Storage Subnet Primary Snapshot Primary Snapshot 1 00 PAS Primary Snapshot k Snapshot 14 00 AM i 4114 00 AM Figure 251 Non destructive Rollback from the Remote Snapshot Best Practices Roll back the primary snapshot and keep the remote snapshots as a backup To ensure that Remote Copy continues to operate roll back the primary volume as follows 1 Preserve the current state of the primary volume that you want to roll back by creating a one time manual remote snapshot of it 2 Roll back the volume Remote snapshots remain intact 3 After the primary volume is rolled back scheduled creation of remote copies will continue Using Remote Copy for Data Migration Remote Copy allows a one time migration of data from one application server to another without interrupting the production application server This capability supports a number of uses such as data mining or content distribution MPC DataFRAME 440 Storage System Software User Manual 381 Sample Remote Copy Configurations Achieving Data Migration You can use Remote Copy to make a complete copy of one or more volumes without interrupting access to the original volumes This type of data migration allows you to copy an entire data set for use by a new application or workgroup To copy data from one location to
73. a remote snapshot The primary snapshot is located on the same cluster as the primary volume Remote Volume The volume that resides in the Remote Copy location where the remote snapshots are created The remote volume contains no data It acts as a pointer to tell the system where to make the copy of the primary snapshot It can be stored on the same cluster or a different cluster than the primary volume Remote Snapshot An identical copy of a primary snapshot The remote snapshot is located on the same cluster as the remote volume Remote Copy Pair The primary volume and its associated remote volume Failover The process by which the user transfers operation of the application server over to the remote volume This can be a manual operation or it can be scripted Acting Primary Volume The remote volume when it assumes the role of the primary volume ina failover scenario Failback After failover the process by which the user restores the primary volume and turns the acting primary back into a remote volume Failover Recovery After failover the process by which the user chooses to fail back to the primary volume or to make the acting primary into a permanent primary volume Synchronize The process of copying the most recent snapshot from the primary volume to a new remote snapshot On failback synchronization is the process of copying the most recent remote snapshot back to the primar
74. capacity Creating larger volumes makes it easy to add additional storage resources to the cluster at a later date How you plan to use the volume is one factor in setting the size Other factors in planning size are calculating the hard threshold and whether you plan to use snapshots Measuring Disk Capacity and Volume Size If you are using Microsoft Windows or Novell NetWare with your Storage System Software you are dealing with two disk space accounting systems the block system and the native file system on Windows this is usually NTFS Block Systems and File Systems Operating systems see hard drives both physical and virtual as abstractions known as block devices arbitrary arrays of storage space which can be read from and written to at will Files on disks are handled by a different abstraction the file system File systems are placed on block devices File systems are given authority over reads and writes to block devices iSCSI and EBSD do not operate at the file system level of abstraction Instead they present the Storage System Software volume to an operating system such as Windows as a block device Typically then a file system is created on top of this block device so that it can be used for storage In contrast an Oracle database uses a Storage System Software volume as a raw block device 212 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Storing File System Data on a
75. core of Remote Copy You use the existing volume and snapshot capabilities along with replication across geographic distances to create remote snapshots 338 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Creating a Remote Snapshot Creating a remote snapshot is the main task in Remote Copy You can create a one time remote snapshot or set up a schedule for recurring remote snapshots Many of the parameters for either case are the same Creating a remote snapshot involves four main steps 1 Log in to the management groups that will contain primary and remote volumes 2 Create a primary snapshot on the primary volume 3 Create a remote volume or select an existing remote volume 4 Specify the settings for the remote snapshot Getting There 1 Log in to the management group that contains the primary volume for which you are creating the remote snapshot 2 Log in to the management group that will contain the remote volume and remote snapshot You can create remote volumes and snapshots within the same management group In that case you only log in to the one management group 3 Right click the primary volume and select Remote Copy gt New Remote Snapshot The New Remote Snapshot window opens shown in Figure 217 New Remote Snapshot New Remote Snapshot Primary Management Group Transaction_Data Volume Name CreditData Snapshot Name Create Primary Snapshot gt New Snapshot Remote
76. eee eeeceeececeeeeeeeeeneeteeeeeaeeeseeeeaeeeaeeseaeeseeeenaeeseeeeaeenaas 191 Using Hot Spares ege Ed eday wit aan ne da ae 191 Requirements for Hot Spares eeeeeceeesceeseeceeeeeeeeeeeeeeaeeeseeeeaeeseeeeaeeeeeesaeeeaeeseeeeeaaeseeeseaeesieeseeeeseeseaeeeaees 192 How a Hot Spare Works AA 192 Cl sters and iSCSI WEE 193 ISCSI Failover and Virtual Po ooo eee eececeeeeeeeeeneeeeeeeeeeeceaeeeeeeeseeeeaeesaeeeeaeesaeeeaeesseeeneeeeaeeseaeeseessaeesieeeeeenaas 193 Greating a e TEE 195 Designating a Hot Spare ccceccesceeeceeseeeseeeeeneeeseeeeeeeesaeseaeeeaeesaaeesaeesaeeesaeeeeenaeeseeeeeaeesaeseaeesnaeesieeeeaeenias 196 Configure Virtual IP and ONG for iSCSI oo eeeeeceeeeeeeeeeeeeeeeneeseeeeeeeesaeeeeaeeeaeeseaeseaeesaeeseeeseaeeeeeseaeesneeees 197 Adding an iSNS Server Optional ccceeccesececeeeeeeeeseeeeeeeeneeeeeeseeeeeaeeesaeseaeeseeeeeaeeeeeeseaeesieeseeeseieeeeeeeenees 197 The Cluster Tab View 217 eee Eeer eege ne dedi 199 Beidler 201 Eur BN NEE 202 Adding a DataFRAME to an Existing Cluster eee ecceeeeeeeeeeneeeeeeeeeeeeaeeeeeeeeaeeseaeeeaeeeeeesaeeseaeeeeeseaeeeenees 202 Changing the Hot Spare Designation eccessseeseeeeeceeeeeceeeeeeesecesseeeaeeesaeseaaeeaeesnaseceeseaeeeeeeeeeeeaeees 203 Changing the Hot Spare Time Out 0 eeceeececeseeeseeeseeeeeeeeeeeseaeeeeaeesaeeeaaeeeaeeseaeseaeesaeeeaeeseeeeeaeeseaeseaeeeaeeees 203 Removing a DataFRAME from a Cluster eee eecceeeeeeseeeeaeeeeeeseeeeee
77. f d Bech Volume i Volume gi i Volume 2 Volume Primary Remote Remote Primary Snapshot Snapshot 1 Snapshot 2 Snapshot Production Site 1 Backup Site Production Site 2 Figure 245 Affordable Disaster Recovery Example Configuration 372 MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations How this Works for Affordable Disaster Recovery If the DataFRAMEs in your primary location fail or volumes become unavailable the off site location contains the most recent remote snapshots e Use the remote snapshots to resume operations as shown in Figure 246 If you created tape backups you can recover data from tape backups as shown in Figure 247 e Only data written to the primary volumes since the last remote snapshot was created will be unavailable e Application servers that were accessing the down volumes will not be available until you reconfigure them to access recovered data To resume operations using the most recent set of remote snapshots 1 In the backup location make the remote volume into a primary volume 2 Configure application servers to access this volume or if network connections are not fast enough to facilitate reading and writing to the off site location copy this volume to a location where application servers can access it more efficiently Application Workstations Ess EEN Ethernet Corporate LAN 2 Production Application Application Servers
78. greater e IP Addresses You must use the same IP addresses for the replacement DataFRAMEs that were assigned to the original DataFRAMEs If you do not have a record of those IP addresses you can retrieve them when performing the restore As part of the restore process the configuration description is displayed and it lists the IP addresses To Restore a Management Group 1 Make sure that the DataFRAMEs you are using to restore your management group are in the Available pool in the Console 2 Right click in the Network view and select Restore Management Group A standard Open window opens shown in Figure 126 178 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Look In EZ Congo Backup ma e n le File Name TransactionDataConfig bin Files of Type Management Group Configuration Files Restore Management Group Cancel Figure 126 Opening the Configuration Binary File 3 Navigate to the location of the configuration binary file 4 Select the file and click Restore Management Group The Verify Management Group Configuration window opens shown in Figure 127 Verify Management Group Configuration Management Group Name Transaction_Data Local Bandwidth 32768 Kb sec Remote Bandwidth 32768 Kb sec Database Version 6 1 virtual Manager None Communication Mode Multicast Off Unicast On Unicast IPs 10 0 33 35 1
79. hours minutes and seconds the copy has been processing N A if not yet available Est Time Remaining Xd Xh Xm Xs X a number and the days hours minutes and seconds estimated to remain in the copy process N A for completed copies or in progress copies not yet calculated 350 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Table 62 Values for Remote Copy Details Window Statistic Values Completion Time MM DD YY HH MM AM PM Date and time copy completed N A for in progress copies Data Data Copied MB GB or TB Amount of data copied so far in smallest unit size Data Remaining MB GB or TB Amount of data remaining to be copied in smallest unit size Current Rate Kb sec Current rate of data being copied in Kb second This rate is recalculated regularly throughout the remote copy process N A If not yet available or completed Avg Rate Kb sec Average rate of copy progress You can leave the Details window open and monitor the progress of the remote copy An example of a Details window with a remote copy in progress is shown in Figure 230 Remote IP Copy Details Source Info Time Primary Mgmt Group Transaction_Data Start Time 6 2105 5 53 PM Primary Snapshot snapshot D Elapsed Time 30m 22s Remote Mgmt Group TransData_Remote Est Time Remaining 59m 56s Remote Snapshot Bkup_0602 Completion Time N A Status Data Manual Scheduled MA
80. notifies you if a remote volume or snapshot is made primary or if the status of the connection between management groups containing primary and remote volumes changes Planning the Remote Snapshot In order to create a remote snapshot e You must be logged in to both the management group that contains the primary volume and the management group containing the target cluster where the remote snapshot will be created e You must designate or create a remote volume in that remote management group e You must have enough space on the target cluster for the remote snapshot Logging in to the Management Group Log in to both management groups before you begin If you are creating the remote volume and remote snapshot in the same management group as the primary volume then you only need to log in to that management group 334 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Designating or Creating the Remote Volume You can create a remote volume by any of the following methods e Make an existing volume into a remote volume e Create a new remote volume during creation of a remote snapshot e Create a new volume from the cluster Details panel and then select the Remote option on the New Volume window For more information about the three methods of creating remote volumes see Creating a Remote Volume on page 342 Using Schedules for Remote Copy Scheduled remote snapshots provide high availability fo
81. of 6 copies of each volume For a 50 GB volume 300 GB of disk capacity is used e In this case data safety and availability are ensured more efficiently by configuring RAID 0 on the DataFRAMEs and then achieving 2 way volume replication on clustered DataFRAMEs For a 50 GB volume 100 GB of disk capacity is used RAID 5 50 uses less disk capacity than RAID 1 10 so it can be combined with replication and still use capacity efficiently One benefit of configuring RAID 5 50 in DataFRAMEs that use replication in a cluster is that if a single disk goes down the data on that DataFRAME can be rebuilt using RAID instead of requiring a complete copy from another DataFRAME in the cluster Rebuilding the disks within a single set is faster and creates less of a performance hit to applications accessing data than copying data from another DataFRAME in the cluster If you are replicating volumes across a cluster Configuring the DataFRAME for RAID 0 allows you to use all of the disk capacity on the module while protecting against failure of individual disks or failure of an entire DataFRAME Configuring the DataFRAME for RAID 5 50 provides redundancy within each DataFRAME while allowing most of the disk capacity to be used for data storage Table 3 summarizes the differences between running RAID 1 or 10 on a stand alone DataFRAME and running RAID 0 or RAID 5 on DataFRAMEs in a cluster MPC DataFRAME 440 Storage System Software User Manual
82. on features and applications are available when you begin using the Storage System Software If you begin using an add on feature or application without first registering a 30 day evaluation period begins Throughout the evaluation period you receive reminders to register and purchase a license for the add on features and applications you want to continue using Evaluating Features Add on features and applications are active and available when you install and configure your system 30 Day Evaluation Period When you use any feature that requires registration a message opens shown in Figure 193 asking you to verify that you want to enter a 30 day evaluation period License Evaluation xj You are about to use an unlicensed feature You may evaluate this feature without a license for up to 30 days If you do not register the feature within that time any volumes or snapshots created using this feature will become unavailable Figure 193 Verifying the Start of the 30 day Evaluation Period Feature Registration During this evaluation period you may configure test and modify any feature At the end of the 30 day evaluation period if you do not register and obtain a license key then all volumes and snapshots associated with the feature or application become unavailable to any clients The data is safe and you can manage the volumes and snapshots in the Console Also the entire configuration can be restored to availa
83. quickly and protecting data however it will slow down user access e Setting the rate low maintains user access to data during the rebuild Setting RAID Rebuild Rate 1 Select the Storage configuration category 2 Click the RAID Setup tab 3 Set the slider for the desired rebuild rate 4 Click Apply The settings are then ready when and if RAID rebuild takes place Starting RAID 68 If RAID has been configured on the DataFRAME and RAID is off it must be started before other RAID tasks can be started Normally once you start RAID you will not have to restart it However in some cases replacing disks requires that you start RAID Example In a DataFRAME two disks were removed and replaced with two new disks However the disks that were removed caused the RAID quorum to break See RAID Quorum on page 69 To prevent losing quorum you replace one of the original disks Then RAID is started Finally the replacement disk is added to RAID MPC DataFRAME 440 Storage System Software User Manual Storage To Start RAID 1 Select the Storage configuration category 2 Click the RAID Setup tab 3 Click Start RAID A confirmation message opens 4 Click OK RAID starts RAID Quorum RAID quorum must be maintained for RAID 1 10 or RAID 5 50 to operate and for data to be preserved Quorum for RAID 1 or RAID 10 For RAID 1 10 quorum requires that at least one disk pair in the DataFRAME and one disk in each
84. remaining pair be intact A DataFRAME configured in RAID 1 contains only one pair of disks so if one of the disks in the pair fails quorum is broken and RAID cannot be rebuilt Data is safe as long as both disks in one of the mirrored pairs are operating normally In order for RAID to rebuild when disks are replaced at least one complete pair of disks must be in the DataFRAME to ensure that data is rebuilt correctly Quorum for RAID 5 or RAID 50 For RAID 5 50 quorum requires that at least n 7 in each RAID set be intact If too many disks fail within one set quorum is broken and RAID cannot be rebuilt Table 4 shows the number of intact disks required to maintain quorum for each configuration of RAID 5 50 sets in the MPC DataFRAME 440 Storage System Table 4 MPC DataFRAME 440 Storage System Disk Requirements for Maintaining RAID Quorum Number of Intact Disks Required to Maintain Quorum RAID Set Configuration 3 plus a spare 4 7 plus a spare 8 NI OJ OIN MPC DataFRAME 440 Storage System Software User Manual 69 Storage Disks are grouped into RAID 5 50 sets from left to right starting with the first disk in the DataFRAME In an MPC DataFRAME 440 Storage System e Disks 1 4 5 8 9 12 and 13 16 or e Disks 1 8 and 9 16 Monitoring RAID Status RAID is critical to the operation of the DataFRAME If RAID has not been configured the DataFRAME cannot be used Monitor the RAID status of a DataFRAME
85. remains valid when you re add that DataFRAME to a management group If you have backed up the configuration as described in Backing Up a Management Group Configuration on page 177 the license key is saved in the binary file used to restore the management group 174 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Removing the DataFRAME 1 Log in to the management group from which you want to remove a DataFRAME 2 Select the DataFRAME to remove 3 Right click and select Remove from Management Group A confirmation message opens 4 Click OK The DataFRAME is removed from the management group and moved to Available status Backing Up a Management Group Configuration Use Backup Configuration of Management Group to save one or both of the following on your local machine e a binary file of the management group configuration from which you can restore the management group e a text file listing the configuration parameters of the management group The binary file enables you to automatically recreate a management group with the same configuration Use the text file for support information or to manually reconstruct the configuration of a management group Note Backing up the management group configuration does not save the configuration information for the individual DataFRAMEs in that management group To back up DataFRAME configurations see Backing Up the DataFRAME Configuration F
86. remote snapshot schedule is maintained java commandline CommandLine jjones saxophone 10 10 45 72 volume_make primary RemVolX Rays 512000 512000 30 java commandline CommandLine jjones trumpet 10 3 11 19 volume_make_remote RemVolX Rays snapshot Convert snapshot from making vol remote 30 Scripting Failover Scripting failover uses a java commandline CommandLine script along with the ebsdvm script for mounting a snapshot Make Remote Volume into Primary Volume java commandline CommandLine lt remote admin name gt lt remote admin password gt lt remote manager ip gt volume_make primary lt remote volume name gt lt soft quota Megabytes gt lt hard quota Megabytes gt lt failure timeout seconds gt Mount New Primary Volume ebsdvm lt remote mgmt group name gt lt remote volume name gt lt auth group name gt lt local ip gt lt number of managers gt lt each managers s ip gt lt lock_mode gt Example Joe s script for failing over to his remote volume would include the following commands to make the remote volume into a primary volume and mount it in the local network to make it available to the backup application servers java commandline CommandLine jjones saxophone 10 10 45 72 volume make primary RemVolX Rays 512000 512000 30 ebsdvm Remote_Images RemVolX Rays adminusers 10 3 11 19 3 10 3 11 27 10 3 11 31 10 3 11 12 ro MPC DataFRAME 440 Storage System Software User Manual 263 Working with Scripting
87. replication through two features e Replication level allows you to choose how many copies of data you want to keep in the cluster e Replication priority allows you to choose whether availability or redundancy is more important in your configuration 214 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Replication Level Note Three replication levels are available depending upon the number of available non hot spare DataFRAMEs in the cluster The level of replication you choose also affects the Replication Priority you can set Table 34 Setting a Replication Level for a Volume With This Number of Select This Available DataFRAMEs Replication For This Number of Copies in Cluster Level One e None e One copy of data in the cluster No replica is created Two not a recommended e None e One copy of data in the cluster no configuration for high e 2 Way replication availability Two copies of data in the cluster One replica is created Three or more e None One copy of data in the cluster no e 2 Way replication e 3 Way Two copies of data in the cluster one replica Three copies of data in the cluster Two replicas are created The system calculates the actual amount of storage resources needed if the replication level is greater than none How Replication Levels Work When you choose 2 way or 3 way replication data is written to either 2 or 3 consecutive Dat
88. same 500 MB hard thresholds and 500 MB soft thresholds The script looks as follows java commandline CommandLine jjones trumpet 10 0 111 212 volume_snapshot X Rays PSS2 xrays 500 500 first primary snapshot 15 java commandline CommandLine jjones saxophone 10 10 45 72 volume_remote_snapshot RemVolX Rays RSS2_ xrays second remote snapshot jjones trumpet 10 0 111 212 PSS2 xrays 15 Creating A Remote Snapshot In The Same Management Group 1 Create the primary snapshot java commandline CommandLine lt primary admin name gt lt primary admin password gt lt primary manager ip gt volume_snapshot lt primary volume name gt lt primary snapshot name gt lt failure timeout seconds gt 2 Create the remote snapshot java commandline CommandLine lt primary admin name gt lt primary admin password gt lt primary manager ip gt volume_remote_ snapshot lt remote volume name gt lt remote snapshot name gt lt remote snapshot description gt lt primary admin name gt lt failure timeout seconds gt Example If Joe Jones was creating his remote snapshot in the same management group the script would look like this java commandline CommandLine jjones trumpet 10 0 111 212 volume_snapshot X Rays PSS2 xrays 500 500 first primary snapshot 30 java commandline CommandLine jjones trumpet 10 0 111 212 volume_remote_snapshot RemVolX Rays RSS2_ xrays second remote snapshot PSS2_xrays 30 Converting a Remote Volume
89. search for modules 1 2 Oona fk Ww Click the Find menu Click By Module IP or Host Name The IP and Host Name List window opens shown in Figure 9 Select the IP Host Name you want to edit Click Edit The Edit IP or Host Name window opens Change the necessary information Click OK to return to the IP and Host Name List window Deleting the IP or Host Name in the Search List Once you enter an IP or host name in the IP and Host Name List that entry is saved Every time you open the Console a search for all the IPs and host names occurs You can delete an IP from the list if you no longer want to search for that DataFRAME 1 2 OOA OQ Click the Find menu Click By Module IP or Host Name The IP and Host Name List window opens shown in Figure 9 Select the IP Host Name to delete Click Delete A confirmation message opens Click OK The IP or host name is removed from the list Click OK MPC DataFRAME 440 Storage System Software User Manual 25 Getting Started Using the Network View The Network View displays DataFRAMEs according to the criteria you set in the Find function The graphics displayed in the Network View provide information about the following e DataFRAMEs e Management groups e Clusters e Volumes Snapshots Remote Copy EE MPC Computers DataFRAME Management Console File Find Tasks View Help Sp Available DataFRAMEs l Group_0 a Clust
90. snapshots to protect against data corruption For more information about hard thresholds and volumes see Planning Hard Thresholds on page 213 Deleting Snapshots One important factor in planning capacity is the fact that when a snapshot is deleted the snapshot s hard and soft thresholds are added to the snapshot or volume directly after it Hard and soft thresholds of the volume or snapshot directly after the deleted snapshot will increase by the hard and soft thresholds of the deleted snapshot up to the size of the volume Adding hard and soft thresholds into the next volume or snapshot insures that all changes to data are accounted for and saved Therefore if you plan a protocol where you routinely delete snapshots you must calculate the effect of adding the hard thresholds back into the volume For a detailed explanation of disk capacity allocation in a cluster and its relationship to disk or volume size in a file system see Measuring Disk Capacity and Volume Size on page 212 Easiest Method for Planning Capacity Make the snapshot hard threshold equal to the volume size and the soft threshold equal to the hard threshold MPC DataFRAME 440 Storage System Software User Manual 237 Working with Snapshots Most Flexible Method for Planning Capacity Make the hard threshold less than the volume size and the soft threshold less than the hard threshold Then increase the volume size hard threshold and soft thres
91. specific World Wide Port Names MPC DataFRAME 440 Storage System only Controlling Client Access to Volumes Client Access and iSCSI Client access using iSCSI can be authenticated via the initiator node name single host or via CHAP Challenge Handshake Authentication Protocol which can support single or multiple hosts Note Our iSCSI terminology is based on the MS iSCSI Initiator terminology Configuring Authentication Groups for iSCSI When configuring client access using iSCSI you create an authentication group that allows iSCSI access The New Authentication Group window with the iSCSI tab is shown in Figure 168 New Authentication Group New Authentication Group Name CreditTransAG Description Volume List No Volume List iscsi EBSD Fibre Channel vj allow access via iSCSI Authentication IS CHAP not required Initiator Node Name CHAP required Figure 168 Creating a New Authentication Group for iSCSI Access Planning iSCSI access requires planning how you want to configure authentication e Single host with or without using CHAP e Multiple hosts with 1 way or 2 way CHAP 266 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Planning iSCSI and CHAP CHAP is a standard authentication protocol The Storage System Software supports no CHAP 1 way CHAP or 2 way CHAP CHAP Glossary e Target secret The t
92. specify the hard threshold and the soft threshold for the primary snapshots 2 Specify the retention policy for the primary snapshots Configuring the Remote Volume and Snapshots 1 Click the Remote Setup tab to bring it to the front janagement Group TransData_Remote Y olume Name Rv _RcDaity v New Volume napshot Name Schedule Name Number Retention for Remote Snapshots 8 E Retain Maximum Of 5 Snapshots Retain Snapshots For Cancel Figure 232 The Remote Setup Tab 2 Select the management group to contain the remote volume and remote snapshots 3 Select an existing volume or click New Volume to create the remote volume See Making a Primary Volume Into a Remote Volume on page 356 4 Specify a retention policy for the remote snapshots MPC DataFRAME 440 Storage System Software User Manual 353 Remote Copy 5 Click OK What the System Does If you created a new volume for the remote volume the system creates a new primary snapshot of the primary volume and a remote snapshot of the remote volume If you selected an existing volume to become the remote volume the system alerts you that all the data on the existing volume will be deleted but that a snapshot of all the existing data will be created first The snapshot that is then created retains all the volume s data 1 Type a name for that snapshot in the alert 2 Click Yes to continue The new snapshot is c
93. square displays underneath the DataFRAME in the Network View MPC DataFRAME 440 Storage System Software User Manual 27 Getting Started Using the Tab View The Tab View displays properties of the item selected in the Network View For example Figure 11 shows the tabs that display when the DataFRAMEs on the network are found Select a tab to perform functions related to the selected item m rc Computers DataFRAME Management Console 101 xj File Find Tasks View Help SS Available DataFRAMEs Tab View DF420 1 Available DataFRAMES Management Groups Alert Messages EBSD Hosts These DataFRAMEs are available for use Name Software Version Model DF 420 1 192 168 0 4 6 3 33 0007 0 SSR316MJ2 DF140 1 192 168 0 9 6 3 41 26080 NSM160 DF140 192 168 0 7 6 3 41 2605 0 NSM160 DF420 192 168 0 3 6 3 33 0007 0 SSR316MJ2 Figure 11 Tab View in Main Window Available DataFRAMEs Tab 28 The Available DataFRAMEs tab shown in Figure 12 lists the DataFRAMEs in the Network View pane that are available that is are not part of a management group Available DataFRAMEs Management Groups Alert Messages EBSD Hosts These DataFRAMEs are available for use Name Software Version Model DF 420 1 192 168 0 4 6 3 33 0007 0 SSR316MJ2 DF140 1 192 168 0 9 6 3 41 2605 0 NSM160 DF140 192 168 0 7 6 3 41 2605 0 NSM160 DF 420 192 168 0 3 6 3 33 0007 0 SSR31
94. the Active Search window closes The Console window opens listing all the DataFRAMEs that were found on the network Click OK to close the Console window MPC DataFRAME 440 Storage System Software User Manual 21 Getting Started 7 Click Done on the List of Subnets to Search window The modules appear in the Network View identified by host name Gmc computers DataFRAME Management Console File Find Tasks View Help SS Available DataFRAMES DETAN Available DataFRAMEs DF140 1 DF420 DF 420 1 Available DataFRAMEs Management Groups Alert Messages EBSD Hosts These DataFRAMEs are available for use g Name IP Software Version Model DF420 1 192 168 0 4 6 3 33 0007 0 SSR316MJ2 DF140 1 192 168 0 9 6 3 41 2605 0 NSM160 DF140 192 168 0 7 6 3 41 2605 0 NSM160 DF420 192 168 0 3 6 3 33 0007 0 SSR316MJ2 Figure 8 DataFRAMEs in the Network View Pane Note The subnet and mask are saved in the list Every time you open the Console the search takes place automatically and all DataFRAMEs on the network are listed in the Network View See Deleting Subnets and Masks on page 23 if you want to disable this search Editing Subnets and Masks Change the subnets and masks used to search for modules Click the Find menu Click By Subnet and Mask The List of Subnets to Search window opens Select the subnet you want to edit Click Edit The subnet an
95. the Configuration Interface if all network connections to the DataFRAME are disabled Use the Configuration Interface to e Add DataFRAME administrators and change passwords e Access and configure network interfaces e Delete a NIC bond e Set the TCP speed and duplex e Edit the frame size e Reset the DataFRAME configuration to factory defaults Connecting to the Configuration Interface Accessing the Configuration Interface is accomplished by attaching a PC ora laptop to the DataFRAME using a null modem cable and connecting to the Con figuration Interface with a terminal emulation program Connecting to the Configuration Interface with Windows On the PC or laptop attached directly to the DataFRAME with a null modem cable open a session with a terminal emulation program such as HyperTerminal or ProComm Plus Use the following settings e Bits per second 19200 e Data bits 8 e Parity None e Stop bits 1 e Flow control None e Backspace key sends Del e Emulation ANSI When the session is established the Configuration Interface window opens shown in Figure 202 Using the Configuration Interface Figure 202 Opening the Configuration Interface Connecting to the Configuration Interface with Linux UNIX If using Linux create the following configuration file You must create the file as root or root must change permissions for dev cua0 in order to create the config file in etc 1 Create the etc minirc
96. the Help merr Maal OS e computers oaa E Eoas and atin 21 Select the storage system in the main window by double clicking on the icon for the system 8 MPC DataFRAME 440 Storage System Software User Manual Getting Started 22 Enter the user name and password for the storage system The Edit Configuration window should display Edit Configuration for DF140 192 168 0 7 ta st Power Off Feature Registration Boot Devices E ge Module Information Backup amp Restore TCP IP Network Sm DataFRAME Administ ge SNMP k Reporting Host Name DF140 Apply IP Address 192 168 0 7 Model DF 100 Series Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Install Software Logged In User admin Change Password 23 Select the storage link in the left hand navigation screen ER Configuration for DF440 1 192 168 0 40 Module RAID Setup Disk Setup Tp Storage P Time RAID Status or Start RAID e TCP IP Network DataF RAME Administ j bs SNMP RAID Configuration Ram 0 v Reconfigure RAID Reporting The RAID Rebuild Rate is a percentage of RAID card throughput Percent Se d RE IS ad IT 70 80 90 100 Apply Device Name Device Type Subdevices dewscsi host0 busOitargetO unO disc Disk 0 deviscsiihostO busOtargeti lun0 disc Disk dewscsi host0 busOitarget2 unO dise Disk dewscsi host0 busOitarget3 lunO disc Disk dewscsi host0 busOitarget4 lunOidise Dis
97. the List of Subnets to Search window e Enter specific IPs or host names to find individual DataFRAMEs See Finding by Module IP or Host Name on page 23 for more information about completing the IP and Host Name List window MPC DataFRAME 440 Storage System Software User Manual 19 Getting Started Note Note 20 Once you have found DataFRAMEs the first time the Find settings are saved Every time you open the Console the search takes place and a window opens as shown in Figure 6 listing which DataFRAMEs have been found The window also lists any EBSD hosts that are found on the network x Summary Found DataFRAMEs are listed below mpc DataFRAME search results Host Name IP Status DF 420 1 192 168 0 4 Found DF140 1 192 168 0 9 Found DF140 192 168 0 7 Found DF420 192 168 0 3 Found To find additional DataF RAMES Repeat this wizard using different search criteria Search again using the Find menu on the DataFRAME Management Console Repeat this Wizard Run the Create Volume Wizard Close Help Figure 6 DataFRAMEs Found Message When you click OK those DataFRAMEs appear in the Network View You can control which DataFRAMEs appear in the Network View by entering only specific IPs or Host Names in the IP and Host Name List window Then when you open the Console only those IPs or Host Names will appear in the Network View Modules Not Found If the network has a lot of traffic or if a mo
98. the Network 4 Select IP Address and complete the fields for IP Address Subnet mask and Default gateway 5 Click OK A confirmation message opens 6 Click OK A message notifying you of an automatic log out opens 7 Click OK The automatic log out occurs Note Wait a few moments for the IP address change to take effect 8 Log in to the newly addressed DataFRAME If you are changing the IP address of a DataFRAME which is a manager ina management group a window opens which displays all the IP addresses of the managers in the management group and a reminder to reconfigure the application servers that are affected by the change Using DHCP A DHCP server becomes a single point of failure in your system configuration If the DHCP server goes down then IP addresses may be lost Warning If you use DHCP be sure to reserve statically assigned IP addresses for all DataFRAMEs on the DHCP server This is required because management groups use unicast communication 1 Select from the list the interface you want to configure for use with DHCP 1 Click Edit The Edit TCP IP Configuration window opens shown in Figure 58 2 Select Obtain an address automatically using the DHCP BOOTP protocol 3 Click OK Configuring NIC Bonding Network interface bonding provides high availability fault tolerance and or bandwidth aggregation for the network interface cards in the DataFRAME Bonds are created by bonding NICs into a sin
99. the New Remote Snapshot Dialog 341 Viewing the Remote Gnapsbot AA 342 Selecting a Cluster for the Remote Volume cece cece tee e tee eeee tect eases eeaeeaeeseeseaetaeseeseaeseeeeaeeeas 343 Creating a New Remote Volume 344 List of Remote Snapshots te sceisiasectaceiedenseie e Amends dentate eaii a 345 Editing a Remote Management Group 346 Editing the Remote Bandwidth AAA 346 Editing a Remote Snapshot AAA 347 Remote Snapshot Details in the Remote Snapshot Tab 349 Remote Snapshot Details for a Completed Remote Copy ececceeeseeeceeseeeteeeeeneeseaeeeaeeseeeeeeeeseeeenas 350 Remote Snapshot Details for a Remote Copy in Progress seseseseeeeesiesreeeeertsirsrtsiesirnrrssrnsrnnrseens 351 Creating a New Remote Snapshot Gchedule 352 The Remote Setup Tab 353 Editing a Remote Snapshot Gchedule AA 355 Volume Changed from Primary to Remote ssseeseeseeeeeeeeeertettetttettetnttntttuinnttntinttnntnntnnntnntnn nenne ne nnt 356 Creating a Snapshot Before Making a Primary Volume into a Remote Volume 357 Finalizing the New Remote VolUMe eeeecceeeeceeeeeeeeeenneeeseneeeeeaeeeeeaaeeeenaeessaeeeseaeeeseneeeeenieeeensneeenaes 358 Making a Remote Volume into a Primary Volume 359 Rolling Back a Primary VOIUMC 2 ccscceedcseccenesdasectasecetsesaasnssadensecenpdinaeatansaeateesscesnenteasatcnetennsesnanceascs 364 Verifying the Primary Volume Roll Back 365 Editing a Management Group 366 High Availability Example Conf
100. the cluster in the new location make the remote volume into a primary volume 4 Configure the application server in the new location to access the new primary volume Figure 253 shows migration of data by making a remote volume into a primary volume Application Workstations Design Design Team 2 Application eam a CAD Server for CAD Server Next Generation for Current Product Product Servers Read Write Read Write Primary Primary Volume H Y Volume 2 Storage Subnet Primary Remote Primary Snapshot Snapshot Snapshot 2 Production Site 1 Production Site 2 Figure 253 Configuration after Data Migration MPC DataFRAME 440 Storage System Software User Manual 383 Sample Remote Copy Configurations 384 MPC DataFRAME 440 Storage System Software User Manual
101. the correct management group from the list of existing management groups 4 Click Add 5 Optional If you want the DataFRAME to run a manager select the DataFRAME in the management group right click and select Start Manager 170 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Adding Manager IP Addresses to Application Servers When you add a manager to a management group a window opens which displays all the IP addresses of the managers in that management group and a reminder to reconfigure the application servers that are affected by the change e Click Save Manager IP Addresses to save this list You can print the list and use it as a reference when reconfiguring application servers Resetting the Management Group Time Any time you change the time setting of a DataFRAME that is running a manager you must reset the time of the management group as well If the manager DataFRAME time is different from the management group time then e File creation times on volumes and snapshots might be affected e Scheduled snapshots might not kick off at the intended time Note Use NTP to ensure closely synchronized times on the DataFRAMEs in the management group When resetting the management group time first verify the time settings of the DataFRAMEs running managers If necessary change time settings to ensure all the manager DataFRAMEs have the same time For information about setting the time
102. the health of the DataFRAME Active monitoring allows you to set up notification through emails alerts in the Console and SNMP traps You can choose which variables to monitor and choose the notification methods for alerts related to the monitored variables For a detailed list of monitored variables see List of Monitored Variables on page 149 Note Critical variables such as the CPU temperature and motherboard temperature have thresholds that trigger a shutdown of the DataFRAME 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select Reporting from the configuration categories The Reporting Window opens 3 Select the Active tab The Active Reporting window opens shown in Figure 101 144 MPC DataFRAME 440 Storage System Software User Manual Reporting E Edit Configuration for DF140 192 168 0 7 xi Module N Passive LogFiles Active Email Diagnostics Alerts G Storage Time e D TCP IP Network d r DataFRAME Adminis g SNMP P i E Reporting Monitored Variables Summary Frequency 15 seconds CPU Temperature Thresholds 2 Power Supply Status If the value exceeds 65 alert CPU Utilization If the value exceeds 70 alert shut down Memory Utilization RAID Status Network Interface Status Save Logfiles Drive Status _Save Log Files D Add Edit Remove Figure 101 Set
103. the relationship between the authentication group fields and the MS iSCSI Initiator see Planning iSCSI and CHAP on page 267 1 In the Authentication box select CHAP not required 2 Copy the Initiator node name into the initiator node name field Authenticate with CHAP Required 1 In the Authentication box select CHAP required 2 Complete the fields necessary for the type of CHAP you intend to configure as shown in Table 49 Table 49 Entering CHAP Information in a New Authentication Group For this CHAP Mode Complete these Fields 1 way CHAP e CHAP name e Target Secret minimum of 12 characters MPC DataFRAME 440 Storage System Software User Manual 275 Controlling Client Access to Volumes Table 49 Entering CHAP Information in a New Authentication Group For this CHAP Mode Complete these Fields 2 way CHAP CHAP name e Target Secret minimum of 12 characters Initiator Secret minimum of 12 characters must be alphanumeric Best Practice Keep a separate record of the iSCSI initiator CHAP information and the corresponding authentication group information Finishing iSCSI Configuration Click OK if you are finished configuring the authentication group Configuring EBSD 1 On the EBSD tab shown in Figure 179 select the check box to allow access via EBSD New Authentication Group Name CreditTransAG Description Volume List No Volume List i
104. the time required to complete the second remote snapshot This is the minimum amount of time that you should allow between scheduled copies Be sure to check the remote bandwidth setting for the management group containing the remote volume since that setting affects the time required to copy a remote snapshot Thresholds Retention Does the cluster that contains the remote snapshots have sufficient space to accommodate scheduled snapshots See Chapter 13 Working with Snapshots for information about managing capacity using volume and snapshot thresholds See Managing Capacity Using Volume and Snapshot Thresholds on page 237 If the cluster does not have sufficient space available the remote snapshot will appear in the Console and it will flash red On the Details tab of the remote snapshot the status says Read only not enough space in cluster to start copy Policies How long do you want to retain the primary snapshots The remote snapshots You can set different retention policies for the primary and remote snapshots For example you can choose to retain two primary snapshots and five remote snap shots The number of snapshots retained refers to completed snapshots Parameters for Remote Snapshot Schedule Retention Policies 336 The system will never delete the last fully synchronized remote snapshot Under some circumstances such as unpredictable network speeds or varying snapshot size a remote snapshot sc
105. to ensure that operation is normal Data Transfer and RAID Status RAID status of Normal Rebuild or Degraded all allow data transfer The only time data cannot be transferred to the DataFRAME is if the RAID status shows Off Data Redundancy and RAID Status In a RAID 1 10 or RAID 5 50 configuration when RAID is degraded there is not full data redundancy Therefore data is at risk if there is a disk failure when RAID is degraded Warning In a degraded RAID 1 10 configuration loss of a second disk within a pair will result in data loss In a degraded RAID 5 50 configuration loss of a second disk will result in data loss The RAID Status is located at the top of the RAID Setup tab in Storage RAID status also displays in the Details Tab on the main Console window when a DataFRAME is selected in the Network view 70 MPC DataFRAME 440 Storage System Software User Manual Storage EMPe computers DataFRAME Management Console ll x File Find Tasks View Help Available DataFRAMEs DF140 DF140 1 DF420 RAID Status DF 420 1 Details Management Group Information Availability Host Name DF140 RAID Status Normal IP Address 192 168 0 7 RAID Configuration RAID 0 Model DF 100 Series Storage Mode block Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Figure 50 Monitoring RAID Status on the Main Console Window The status displays one of four RAID states e Normal RAID i
106. updates are replicated to the Des Moines backup facility using Remote Copy This chapter provides instructions for registering configuring and using Remote Copy for business continuance backup and recovery and failover Registering Remote Copy Remote Copy is a feature upgrade You must purchase a Remote Data Protection Pak license to use Remote Copy beyond the 30 day evaluation period For infor mation about registering Remote Copy licenses see Chapter 16 Feature Regis tration Number of Remote Copy Licenses Required Register Remote Copy on each management group that contains DataFRAMEs that will participate in Remote Copy If there are DataFRAMEs in a management group that will not contain Remote Copy primary or remote volumes you do not need to purchase licenses for those modules For example if your management group contains a cluster of two DataFRAMEs that will contain a remote volume and another cluster of three DataFRAMEs that will not use Remote Copy you only need two Remote Copy licenses Remote Copy Glossary for Remote Copy 330 The following terminology is used in describing the components and processes involved in Remote Copy Term Definition Primary Volume The volume which is being accessed by the application server The primary volume is the volume that is backed up with Remote Copy Primary Snapshot A snapshot of the primary volume which is created in the process of creating
107. 0 0 33 18 10 0 33 13 Authentication Groups Name AuthGroup_0 Description 4 Restore Management Group Figure 127 Verifying the Management Group Configuration MPC DataFRAME 440 Storage System Software User Manual 5 After you have reviewed the configuration parameters click Restore Management Group The management group is restored 179 Working with Management Groups Deleting a Management Group Delete a management group when you are completely reconfiguring your network storage Warning When a management group is deleted all data stored on DataFRAMEs in that management group are lost Prerequisites e Logging in to each DataFRAME in the management group e Stopping the virtual manager and managers on the individual DataFRAMEs 1 Log in to the management group in the Network View The management group Tab View opens From the Tasks menu on the Details tab select Delete Management Group Aconfirmation message opens Click OK When the management group is deleted the DataFRAMEs return to available status in the Network View a A WwW N Setting the Management Group Version If instructed by customer support you can set the management group version back to a previous version of the software Setting the version back requires using a special command line option before opening the Console Customer support will instruct you on using the command line option Selecting a Management Group from
108. 16 Select the By Subnet and Mask global broadcast or By Module IP or Host Name individual search option to search for your storage system If you are using a fixed IP address select the By Module IP or Host Name individual search option Press Next to continue xi Search for DataFRAMEs either globally or individually Select subnetmask or IP hostname mpc How would you like to search for your DataFRAMEs By Subnet and Mask global broadcast By Module IP or Host Name individual search To continue click Next 17 If you selected the By Module IP or Host Name individual search option a Search by DataFRAME IP or Hostname list will display Click on the Add button Find DataFRAMEs Wizard xj Search by DataFRAME IP or hostname Enter IP address or hostname mpc Click Add to enter an IP address or host name Use Edit or Delete to refine your list IPorHostName Status Add em Delete To find your DataFRAMEs on the network click Finish 6 MPC DataFRAME 440 Storage System Software User Manual Getting Started 18 At the Add IP or Host Name screen enter the static IP address of your storage system Click on OK to continue Add IPorHostName IP or Host Name 169 254 1 1 19 Click on Finish to search for the specified IP address The storage system will display a Found under Status if the un
109. 227 Edit Snapshot Snapshot Name Description Created Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Figure 227 Editing a Remote Snapshot 3 Change the desired information and click OK Deleting a Remote Snapshot 1 Log in to the management group that contains the remote snapshot 2 Right click the remote snapshot and select Delete Snapshot from the menu A confirmation message opens 3 Click OK MPC DataFRAME 440 Storage System Software User Manual 347 Remote Copy Monitoring Remote Snapshots Information for monitoring remote snapshots is available from multiple sources Active monitoring features provide you the capability to configure alerts that you view in the Console as well as receiving alerts as emails and through SNMP traps The Console tab view also provides monitoring information for remote snapshots Configuring Active Monitoring Alerts for Remote Copy There are four variables for remote snapshots for which you can configure alerts Notification for these variables automatically come as alert messages in the Console You can also configure Active Monitoring to receive email notification or for SNMP traps The Remote Copy variables that are monitored include e Remote Copy status an alert is generated if the copy fails e Remote Copy complete an alert is generated when the remote cop
110. 377 Achieving Non Destructive Rollback AAA 378 Configuration for Non Destructive Rollback 378 How This Configuration Works for Non Destructive Rollback AA 379 Best Practices siii seastasevccudheneeseenayeeecsuagey aE A aa aaa a O a A Ea tectsbeaediasaccspesaeeanides Using Remote Copy for Data Migration Achieving Data Migration E Configuration for Data Migration How This Configuration Works for Data Migration 0 eccceeceeeeceseeeeeeeeeeeeeaeeeeeeeaeeeaeeseaeseaeeeseeeseeseeeeeaaes 382 xiv MPC DataFRAME 440 Storage System Software User Manual List of Figures List of Figures Figure 1 Viewing the Wizard Launch Pad ccecececeeesceeceeeeeeeecee encase ee eeeeneseeeeeseeeeenseeeseseaeeseeeeesseneeseaaeeseeneesenesens 15 Figure 2 Features of the Console Main Window cceeccseeeeeeeeeeneeeeeeteeeeeeeeeeaeeseeseeeesaeeseaeeeaeseaeseeeeeeeeseeeeaeeeneee 16 Figure 3 Graphical Legend Items from the Help Men 17 Figure 4 Graphical Legend States Tab from the Help Men 18 Figure 5 Graphical Legend Hardware Tab from the Help Men 19 Figure 6 DataFRAMEs Found Message 20 Figure 7 Using Subnet and Mask to Gearch A 21 Figure 8 DataFRAMEs in the Network View Panen 22 Figure 9 Using IP or Host Name to Search AA 24 Figure 10 Network View Features c ccescceeceseeeeseneeeeeeeeeteeeeneeaeeesneesessgeeeesaaeeesseeeesageesesaaeeseseseeseeseneneaeesseneenseneees 26 Figure 11 Tab View in Main Window
111. 40 logger MARK Nov 19 02 47 33 DF140 hydra BBU Capacity Test Overdue on Card 1 overdue Figure 109 Viewing Alerts 4 To refresh the list of alerts click Refresh MPC DataFRAME 440 Storage System Software User Manual 155 Reporting 5 Optional To save the list of alerts click Save to File Then select a location for the file 156 MPC DataFRAME 440 Storage System Software User Manual 9 Working with Management Groups A management group is a collection of one or more DataFRAMEs It is the container within which you cluster DataFRAMEs and create volumes for storage Creating a management group is the first step towards maximizing the clustering capacity of the DataFRAME Management groups serve several purposes To organize your DataFRAMEs into different groups for different functional areas of your organization For example you might create a management group for your Oracle applications and a separate management group for user file share storage To ensure added administrative security For example you could give one storage administrator access to the DataFRAMEs in one management group but not in another management group To prevent some storage resources from being used unintentionally If a DataFRAME is not in a management group the management group cannot use that DataFRAME as a storage resource For example all of the DataFRAMEs in management group 1 can be pooled together for use by volumes i
112. 440 Storage System Software User Manual 61 Storage This Item Describes This Device Name The disks pairs of disks or sets used in RAID For RAID 0 an entry for each disk in the DataFRAME For RAID 1 and RAID 10 one entry for each disk pair For RAID 5 and RAID 50 one entry for each set Device Type The RAID level of the device For RAID 0 the device type is Disk For RAID 1 and RAID 10 the device type for each disk pair is RAID 1 For RAID 5 and RAID 50 the device type for each disk set is RAID 5 If the device is not functioning properly the RAID Level reads failed and the level For example failed 5 Subdevices The number of disks included in the device For RAID 0 there are no subdevices because each disk is listed separately in the report For RAID 1 and RAID 10 there are 2 subdevices per device MPC DataFRAME 440 Storage System For RAID 5 and RAID 50 there are either 4 or 8 subdevices per set Devices Configured in RAID 0 If RAID 0 is configured each physical disk operates as a separate RAID 0 disk as shown below RAID 0 disks T Z 3 4 10 Figure 46 RAID 0 on an MPC DataFRAME 440 Storage System Devices Configured in RAID 1 10 If RAID 1 or 10 is configured the physical disks are combined into mirrored pairs of disks RAID 1 uses only one pair of disks RAID 10 uses up to 8 pairs of disks 62 MPC DataFRAME 440 Storage
113. 60 72 os ae a e m Hard Threshold 60 64 80 96 128 192 256 512 1024 Auto Grow Increments Figure 151 Example Automatic Auto Grow Chart 224 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Creating a Volume A volume resides on the DataFRAME s contained in a cluster 1 Log in to the management group for which you want to create a volume The management group Tab View opens 2 Select the cluster on which you want to create a volume The cluster Tab View opens 3 Click the Volumes tab The Volumes tab opens shown in Figure 152 mre Computers DataFRAME Management Console File Find Tasks View Help a7 Available DataFRAMEs Group_0 aaa Cluster_0 DF 420 1 lt m a DF140 DF140 1 olumes ta Details Volumes DataFRAMEs DiskUsage Usage Graphs iSCSI Remote Snapshot Name Replication Level Size Hard Threshald Soft Threshold Figure 152 Viewing the Volumes Tab 4 From the Tasks menu click New Volume The New Volume window opens shown in Figure 153 See Table 35 on page 218 for detailed information about setting volume parameters See Requirements for Volumes on page 218 for a detailed description of each item MPC DataFRAME 440 Storage System Software User Manual 225 Working with Volumes New volume amp ix New Volume Type Primary Remote Volume Nam
114. 61 5 MB 2 596 Kbisec COMPLETE Transaction Fus Op TransData_ Bkup0601 38 3 495 60 5 MB 10 114 Kbisec COPYING Figure 228 Remote Snapshot Details in the Remote Snapshot Tab The remote snapshot details displayed include e Primary Management Group containing the primary volume from which remote snapshots are created e Primary Snapshot from which the remote snapshot is copied e Remote Management Group containing the remote volume to which the remote snapshot is attached e Remote Snapshot target for the copied primary snapshot e Complete the incremental progress of the remote copy operation e Elapsed Time incremental time of the copy operation e Data Copied incremental quantity of data copied e Rate rate at which data is being copied or when the remote snapshot is complete the average rate for the total operation e State status of the operation Viewing Status in the Remote Copy Details Window The Remote Copy Details window displays additional details about a remote snapshot 1 From the Remote Snapshot tab select the remote snapshot for which you want to view details 2 Right click and select View Remote Snapshot Details or double click the snapshot The Remote Copy Details window opens as shown in Figure 229 MPC DataFRAME 440 Storage System Software User Manual 349 Remote Copy Remote IP Copy Details Source Info Time Primary Mgmt Group Transaction_Data Start T
115. 65 Storage Table 3 Data Availability and Safety in RAID 1 10 Configuration and in a Clustered RAID 0 or RAID 5 50 Configuration Data l pate Availability If Configu Safety and Avail ability If Network Hot Spare To ration ability During Disk Entire Connection Replace Failed Failure DataFRAM to Hardware E Fails DataFRAME Lost Stand alone Yes In any No No No hot spare DataFRAMEs configuration 1 disk disk within the per mirrored pair can DataFRAME RAID 1 10 fail but there is no redundancy in pairs with a failed disk Replicated Yes However if any Yes Yes Yes configure a volumes on disk in the DataFRAME hot spare clustered fails the entire DataFRAME DataFRAMEs DataFRAME must be within a cluster copied from another RAID 0 DataFRAME in the cluster Replicated Yes 1 disk per RAID Yes Yes Yes select a volumes on set can fail without hot spare disk clustered copying from another RAID DataFRAMEs DataFRAME in the configuration RAID 5 50 cluster Planning RAID for Capacity Growth If you plan to add more DataFRAMEs to your network as your storage needs grow remember that all DataFRAMEs in a cluster must have the same RAID configuration For example if you configure RAID 10 now and later decide to replicate volumes through clustering then any new DataFRAMEs must also be configured for RAID 10 Alternately you can remove all data from your existing DataFRAMEs configure RAID 0
116. 6MJ2 Figure 12 DataFRAMEs Tab with DataFRAMEs Listed MPC DataFRAME 440 Storage System Software User Manual Getting Started Management Groups Tab The Management Groups tab shown in Figure 13 lists all the management groups currently created with the DataFRAMEs that are displayed in the Network View pane For information on management groups see Working with Management Groups on page 157 Available DataFRAMEs Management Groups Alert Messages EBSD Hosts Name Logged In Transaction_Data Yes Figure 13 Management Groups Tab from Network Tab View Alert Messages Tab Use the Alert Messages tab to review any alert messages Figure 14 shows the area in which alert messages display These messages include alerts from the monitoring parameters you set in Reporting for individual DataFRAMEs See Using Active Monitoring on page 144 for detailed information about setting reporting parameters Anola Dog aftnAbt 3 Management Gres Meri Messajyes BESD Hosts 7 807 2005 1252 58 PAE EST Alert bom Wie TT 210 074 RAD States arma Vaise changed mp 07 2005 1242 17 PH MSI Are Groen Showy VIAN 11 17 CPU Fon Speed lt D Vater m betow 7100 mp 607 2005 12521 PM MSI Mort tom Mighty FT 7 19 8 7 1 Dewice fash 1 Status Actie Vake Changed zt 0507 2006 125211 PM MST Mort from bmastors i001 108 CPU Fan Speed D Vater is betow 3900 P3 0567 2005 1251461 PA MES Anni tom Mi
117. 7 MPC DataFRAME 440 Storage System Software User Manual 273 Controlling Client Access to Volumes New Authentication Group New Authentication Group Name AuthGroup_o Description Volume List No Volume List iSCSI EBSD Fibre Channel 0 Allow access via iSCSI Authentication Figure 177 Creating a New Authentication Group 4 Type a name and description for the authentication group The authentication group name is case sensitive 5 Select the volume list if appropriate The volume list can be added later 6 Select the tab for the appropriate type of host access Configuring iSCSI 1 On the Authentication Group iSCSI tab shown in Figure 178 select the check box to allow access via iSCSI 2 Select the authentication method Warning Allowing more than one iSCSI application server to connect to a volume could result in data corruption 274 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes New Authentication Group New Authentication Group Name CreditTransAG Description Volume List No Volume List iscsi EBSD Fibre Channel vj allow access via iSCSI Authentication CHAP not required Initiator Node Name CHAP required Figure 178 Creating iSCSI Access in New Authentication Group Authenticate with CHAP Not Required For detailed illustrations of
118. 7 Adding an Administrative User 127 Figure 88 Adding a Group to an Administrative User 128 Figure 89 Sorting Administrative Users EE 128 Figure 90 Editing an Administrative Leer 129 Figure 91Using SNMP EE 131 Figure 92 Enabling the SNMP Agent 132 Figure 93 Adding an SNMP Client 133 Figure 94 Editing a Host in the Access Control Uert 134 Figure 95 Editing SNMP Client from the Access Control Let 134 Figure 96 Enabling SNMP Trape ennnen nnt Figure 97 Adding an SNMP Trap Pecipienmt A Figure 98 Viewing the Reporting Wimdow A Figure 99 Saving Log Files to a Local Machine Figure 100 Adding a Remote Log siiccs nsstihertes tetas nee a ated ee eee i ee Figure 101 Setting Active Monitoring Variables AAA Figure102 Adding a WETTER Figure 103 Adding a Variable Step 3 Figure 104 Setting Alerts for Monitored Variables Figure 105 Viewing the Monitoring Variable Summary on the Active Window ssseeseeseeseeesresrrenrsrerirsrsrrerresseee 148 xvi MPC DataFRAME 440 Storage System Software User Manual Figure 106 Figure 107 Figure 108 Figure 109 Figure 110 Figure 111 Figure 112 Figure 113 Figure 114 Figure 115 Figure 116 Figure 117 Figure 118 Figure 119 Figure 120 Figure 121 Figure 122 Figure 123 Figure 124 Figure 125 Figure 126 Figure 127 Figure 128 Figure 129 Figure 130 Figure 131 Figure 132 Figure 133 Figure 134 Figure 135 Figure 136 Figure 137 Figure 138 Fig
119. 92 Evaluating the Configurable Snapshot Pak Starting the License Evaluation Period Backing Out of the License Evaluation Period AA 292 Evaluating the Remote Data Protection Pak 0 0 eeeceeseeeeeeneeeceeeeeeeeeeseaeesesaeeeseneeeeseeeeseeeeesaeeeennaeeeesneeeseaeees 293 Starting the License Evaluation Period A 293 Backing Out of the License Evaluation Period AA 293 te RETTEN 294 Turn On Scripting Evaluation ceeccesceesceceeeeeseeeeseeceeeseenecseeseaeneneesaaesaeeseeeeseessaaeeeesseeeeaeeseaeseneeenaeees 294 Turn Off Scripting Evaluation cistsicieeieced tintin dannii nina ciel 295 Registering Features and Applications cc eeeseeeeseeeeeneeeeeneeeeeeaeeeeeneeeeseaeessaeeeeeaeeeseneeeseneeeseneeeeeaeeeeeeeeseeees 296 Using He 296 Registering Available DataFRAMEs for License Keys ccesceeeeeeeeeeeneeeseeeeeeneeeeesaaeeseaeeeesneeeneneetenes 296 Registering DataFRAMEs in a Management Group ceeeceeeceeeeeeeeeeeeeeeeeeeeeseeeeaeeseeeeeaeeeeeeseneeseeeeaeeseaee 297 Appendix A Using the Configuration Interface ccccessseeeeeeeeeeseeeeteeeeeeeeees 301 Connecting to the Configuration Interface oo eeceeeceeeseeeneeeeeeeeeeeceaeeeeeeaeeeaeeseeeeeaeeseaeeeaeesaeesieeeeaeestaeeeaeeeneeeeas 301 Connecting to the Configuration Interface with Windows c cecceeseeeeeeeeeeeteeeeeeteeeetaeeseaeesaeteaeeeeeesas 301 Connecting to the Configuration Interface with LINUX UNIX eceecceeece
120. AME Administration Click Users and find the admin user on the list 1 Tab to Login and press Enter The Configuration Interface main menu opens shown in Figure 204 Figure 204 Configuration Interface Main Menu Configuring Administrative Users Use the Configuration Interface to add new administrative users or to change administrative passwords You can only change the password for the administrative user that you used to log in to the Configuration Interface 1 On the Configuration Interface main menu tab to General Settings and press Enter The General window opens shown in Figure 205 MPC DataFRAME 440 Storage System Software User Manual 303 Using the Configuration Interface Figure 205 General Settings Window 2 To add an administrative user tab to Add Administrator and press Enter Then enter the new user s name and password Confirm password tab to Ok and press Enter 3 To change the password for the user that you are currently logged in as tab to Change Password and press Enter Then enter the new password Confirm password tab to Ok and press Enter 4 On the General window tab to Done and press Enter Configuring a Network Connection The DataFRAME has two 1000BASE T Gigabit Ethernet NICs in its motherboard These interfaces are named Motherboard Port0 and Motherboard Port1 In addition the DataFRAME can include multiple add on PCI cards each with up to 4 interfaces These add on interfaces are
121. AME before you add the DataFRAME to a management group Allocate a static IP address for the logical bond interface You cannot use DHCP for the bond IP MPC DataFRAME 440 Storage System Software User Manual 93 Managing the Network Warning To ensure that the bond works correctly you should configure it as follows e Create the bond on the DataFRAME before you add it to a management group e Verify that the bond is created If you create the bond on the DataFRAME after it is in a management group and if it does not work correctly you might e lose the DataFRAME from the network e lose quorum in the management group for a while Creating the Bond 1 Ensure that the DataFRAME is not in a management group 2 Log in to the DataFRAME 3 On the TCP IP tab shown in Figure 63 select 2 or 4 NICs to bond The NICs that you select do not have to be consecutive NICs in the list Gedit configuration for DF140 192 168 0 7 x e Module A Tcpap TCP Status DNS Routing Communication W Storage Time EE Alen Ge wel al ele S E wl Sex SNMP K Reporting 00 30 48 Disabled 0 0 0 0 0000 0000 NIC BR Log Out New Bond it Delete Bond Figure 63 Selecting Motherboard Port0 and Slot1 Port0 for a New Bond 4 Click New Bond The Create Bond Configuration window opens shown in Figure 64 94 MPC DataFRAME 440 Storage System Software User Manual Managing the Network
122. Administrative User to a Group 2 Select one or more users you want to add to the group 3 Click Add Adding Administrative Group Permissions Administrative groups can have e Different levels of access to the DataFRAME such as read write e Access to different management capabilities for the DataFRAME such as creating volumes When you are creating a group you also set the management capabilities available to members of a group The default setting for a new group is Read Only for each category 1 From the Create Administrative Group window click the Group Permission tab to bring it to the front shown in Figure 83 122 MPC DataFRAME 440 Storage System Software User Manual Administrative Users and Groups Create Administrative Group Xx General Group Permission Network TCPAP SNMP DNS Time Management Groups RAID Drive Hot Swap System and Disk Report CH Change Password DataFRAME Administration Boot Upgrade e Read Only Read Modify Full Kg Figure 83 Adding Permissions to Administrative Groups 2 Click the permission level for each function for the group you are creating 3 Click the General tab and complete the rest of the information if you have not already done so 4 Click OK to finish adding the group The DataFRAME Administration window opens with the Groups tab in front The new group is added to the list Description of Administrative Group Permissions Table 28 De
123. BSD Configuration 1 Click OK if you are finished configuring the authentication group or Select the iSCSI tab if you want to add iSCSI configuration to this authentication group MPC DataFRAME 440 Storage System Software User Manual 277 Controlling Client Access to Volumes Configuring Fibre Channel MPC DataFRAME 440 Storage System only 1 On the Fibre Channel tab shown in Figure 181 select the check box to allow access via Fibre Channel New Authentication Group EZ New Authentication Group Name AuthGroup_O Description Volume List No Volume List iscsi EBSD Fibre Channel vi Allow access via Fibre Channel Authenticate hosts with the following WWWPNs Name Optional WWPN oae Cancel Figure 181 Configuring Fibre Channel for New Authentication Group 2 Click Add The Add Name and WWPN window opens shown in Figure 182 Add Name and WWPN xj Name Optional WWPN must be 16 hex characters Cancel Figure 182 Adding a Name and WWPN for Fibre Channel Authentication 3 Add a name optional and a WWPN 4 Click OK 5 Repeat for additional host WWPNs 278 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Finishing Up the New Authentication Group 1 Click OK on the New Authentication Group window when you are finished The Network View opens Go to the Authe
124. Chapter 16 Feature Registration Requirements for Hot Spares To designate a hot spare DataFRAME for a cluster the following requirements apply e A cluster must contain at least 3 DataFRAMEs to have one DataFRAME designated as a hot spare e At most one hot spare DataFRAME can be designated per cluster However a cluster does not require a hot spare The hot spare DataFRAME should be equal to or greater in size than the other DataFRAMEs in the cluster How a Hot Spare Works If a DataFRAME in a cluster goes down and a hot spare is designated for that cluster then the spare is automatically activated and data from replicated volumes start to migrate to the new DataFRAME At this point the cluster no longer contains a hot spare When the down DataFRAME comes back up it becomes the hot spare When a hot spare is activated it is not configured as a manager If you want to designate the activated hot spare as a manager you must start the manager See Starting and Stopping Managers on page 171 Setting the Hot Spare Time Out The hot spare time out designates the amount of time before a hot spare is activated in the cluster When a hot spare is activated the system will migrate data onto the new DataFRAME This data migration may take some time Setting the hot spare time out allows you to control for situations in which you don t want the hot spare activated for example if your network has high latency The time
125. D DataFRAMG Ip RAD DF140 1 192 168 RAID 0 DF140 192168 RAID 0 DF420 192168 RAID 0 lt Add Add all BI Remove gt Remove All gt Hot Spare None s Hot Spare Time Out 0 Seconds Cancel Figure 143 Editing a Cluster Adding a DataFRAME to an Existing Cluster Add a DataFRAME to an exiting cluster to expand the storage for that cluster or to designate the DataFRAME as a hot spare Prerequisites e Configure the DataFRAME to match the DataFRAMEs already in the cluster See DataFRAME Configuration Window on page 34 for information about what features must be configured e Add the DataFRAME to the management group that contains the cluster Note If you mix DataFRAMEs with different capacities in a cluster all DataFRAMEs in the cluster will operate at a capacity equal to that of the smallest capacity DataFRAME 1 Select a DataFRAME from the DataFRAMEs Available list 202 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters 2 Click Add The DataFRAME moves to the DataFRAMEs in Cluster list or Click Add All to move all the DataFRAMEs from the DataFRAMEs Available list to the DataFRAMEs in Cluster list 3 Click OK when you are finished Changing the Hot Spare Designation You can add a hot spare or remove a hot spare from a cluster as long as the volumes in that cluster have a replication priority of avai
126. Data Snapshot Name Schedule Name Number Thresholds for Primary Snapshots Size 9 0 GB Hard Threshold 1 0 GB v Soft Threshold 1 0 GB e Retention for Primary Snapshots Retain Maximum of 10 Snapshots Retain Snapshots For Cancel Figure 233 Editing a Remote Snapshot Schedule Deleting a Remote Snapshot Schedule 1 Select the volume for which you want to delete the remote snapshot schedule The volume tab view opens 2 Click the Remote Snapshot Schedule tab to bring it to the front 3 Select the schedule you want to delete 4 From the Tasks menu select Delete Schedule A confirmation message opens 5 Click OK Changing the Roles of Primary and Remote Volumes Changing the roles of primary and remote volumes comes into play during failover recovery You use these procedures when you are resynchronizing data between MPC DataFRAME 440 Storage System Software User Manual 355 Remote Copy the acting primary volume and the recovered or newly configured production site primary volume Making a Primary Volume Into a Remote Volume You can make any primary volume into a remote volume First the system takes a snapshot of the volume to preserve the existing data that is on the volume The data can then be accessed on that snapshot Next the volume is converted to a remote volume The remote volume is a placeholder for the remote snapshots and does not contain data itself So the size hard thr
127. Data Availability and Safety in RAID 1 10 Configuration and in a Clustered RAID 0 or RAID 5 50 Configura le 66 Table 4 MPC DataFRAME 440 Storage System Disk Requirements for Maintaining RAID Quorum eseseeeeen 69 Table 5 Relationship of Software Disk Display Numbering to Hardware Drive Bay Numbering in the MPC DataFRAME 440 Storage Syste AEN 73 Table 6 Description of items on the disk report 74 Table 7 Memory Requirements for Fully Populated Data FRAME AAA 77 Table 8 Network interfaces Displayed on the TCP IP Tab 80 Table 9 Identifying the NICs in the Motberboar AAA 80 Table 10 Identifying Add on NICS EEN 81 Table 11 Comparison of Active Backup and NIC Aggregation Bondimg 85 Table 12 Physical and Logical Interfaces in a Bond 86 Table 13 Description of NIC Status in an Active Backup Configuration cccecceeseeeeeeeeeeeeeeeeneeeeeeeeeeeeeeeeeeeeeneen 86 Table 14 DataFRAME Active Backup Failover Scenario and Corresponding NIC Gtatus cc eeeeeeeeeeeeeeeeneees 87 Table 15 NIC Status During Failover with Active Backup 88 Table 16 DataFRAME NIC Aggregation Failover Scenario and Corresponding NIC Gras 91 Table 17 NIC Status During Failover with NIC Aooregatton 91 Table 18 Status of Information About Network Interfaces ccceeeeeeeeseeeeeeteeeeeeeeteeeeeaeeeaas Table 19 Setting DataFRAME Speed and Duplex Settings Table 20 Setting Corresponding Frame Sizes on DataFRAMEs and Windows or Linux Clients 102 Table
128. DataFRAME 440 Storage System Software User Manual Disaster Recovery Using A Virtual Manager virtual manager added 2 regular 2 regular managers managers Normal Operation 2 regular 2 regular managers managers 1 virtual manager started Scenario 1 Communication link is lost Start virtual manager on only one site 2 managers 2 regular managers lt 1 virtual manager started Scenario 2 Site A fails Start virtual manager on Site B 2 regular managers 1 virtual manager started Scenario 3 Site B fails Start virtual manager on Site A Examples of two site failure scenarios where a virtual manager is started to regain quorum In all the failure scenarios only one site becomes primary with a virtual manager started Figure 128 Correct Two site Failure Scenarios Using Virtual Managers MPC DataFRAME 440 Storage System Software User Manual 183 Disaster Recovery Using A Virtual Manager Site B virtual manager added 2 regular 2 regular managers managers Normal Operation 2 regular managers 2 regular managers 1 virtual manager started 1 virtual manager started Scenario 1 Communication link is lost Virtual manager started on both sites 2 regular managers 2 virtual managers started Scenario 2 Site A goes down Two virtual managers started on Site B Warning Examples of what NOT to do when using a virtual manager to regain quorum Figure 129 Incorrect Uses of Virtual Ma
129. DataFRAME Management Console O xj File Find Tasks View Help a Cluster _0 d DF 420 1 am d DEI A0 DF420 a DF140 1 New Volume List New Authentication Group New Cluster Groups Volume Lists Register Times Remote Snapshot Log In to Management Group got Local Bandwidth 32768 Kb sec Log Out of Management Group Edit Management Group Delete Management Group Back up Configuration of Management Group Add or Delete Virtual Manager Feature Registration Figure 133 Starting a Virtual Manager The virtual manager starts on that DataFRAME and the black triangle the graphic indicator of the virtual manager appears under the DataFRAME shown in Figure 134 See Icons Used in the Storage System Console on page 17 for a key to all the graphic indicators a a Transaction_Data a Card_Payments Cd Virtual manager g Sr indicator y Denvert Bouldert Figure 134 Indicator of the Virtual Manager MPC DataFRAME 440 Storage System Software User Manual 189 Disaster Recovery Using A Virtual Manager Note If you attempt to start a virtual manager on a DataFRAME that appears to be up in the Console and you receive a message that the DataFRAME is down start the virtual manager on a different DataFRAME This situation can occur when quorum is lost because the Console displays the DataFRAME in a normal state even though the
130. DataFRAMEs Available list Note The DataFRAMEs in the list are all those included in the management group that are not already in a cluster 7 Click Add The selected DataFRAMEs move to the DataFRAMEs in Cluster list or Click Add All to move all the DataFRAMEs from the Available list to the DataFRAMEs in Cluster list Designating a Hot Spare You must purchase the Scalability Pak to use the hot spare feature beyond the 30 day evaluation period Detailed information about using hot spares is in Using Hot Spares on page 191 1 Optional Click the Hot Spare drop down list to designate a hot spare Only DataFRAMEs in the cluster are displayed in the Hot Spare list Hot spares cannot be used for storage that is you cannot create volumes on them See Using Hot Spares on page 191 for detailed information about hot spares 196 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters 2 Optional If you designate a hot spare you can set the hot spare time out See Setting the Hot Spare Time Out on page 192 Configure Virtual IP and iSNS for iSCSI Optional To configure iSCSI failover for initiators that do not support multiple addresses per target add a virtual IP for the cluster 1 Click the iSCSI tab to bring it to the front shown in Figure 137 x New Cluster General iscsi Virtual IP A cluster of multiple DataF RAMES requires a virtual IP to configure fai
131. E EEEEE EEEE EnEn 236 Table 40 Space used by Snapshots when Hard Threshold is set to the Original Volume Gize 238 Table 41 Space used by Snapshots when Hard Threshold is Reduced cccsccccescceessneeeseeeeeeeeeeeessueeesseeessaeess 238 Table 42 Data Requirements for Editing a Snapshot AA 245 Table 43 Requirements for Scheduling Snapshots AAA 247 Table 44 Requirements for Rolling Back a Volume 251 Table 45 Setting the Environment for Using Scripting Tools 255 Table 46 Parameters for commandline COMmmMandLine ccsecceeceeseesseeeeeeeeeeeeeeeeeeneaseeenseeseeeseeeensaeseeenseneneeeas 256 Table 47 Parameters for ebedvm ccssseeessennseneesseeeessceeesseneeeaseeeeaseeesaseeeeeseeesessueeeesneeeaseneesesoesaseneeeaseeeesseneeaseaees 258 Table 48 Configuring iSCSI CHAP A 268 MPC DataFRAME 440 Storage System Software User Manual xxi List of Tables Table 49 Table 50 Table 51 Table 52 Table 53 Table 54 Table 55 Table 56 Table 57 Table 58 Table 59 Table 60 Table 61 Table 62 Table 63 Table 64 Xxii Entering CHAP Information in a New Authentication Group 275 Choosing the Level of Access for Hosts using the EBSD Driver 277 Characteristics of Permission Levels A 283 Safely Backing out of Scalability Pak Evaluatton 292 Safely Backing out of Configurable Snapshot Pak Evaluation cceeceeeceeseeeeeeeeeeteeeeeeeeeeeeteeeeneeenaes 293 Safely Backing Out of Remote Data Protection Pak E
132. ERRE AEE octaecorseniseacieu ETELE Oni Entire 105 Editing a Domain Name in the DNS Suffixes List cc eeceeeeeeeeeeeeeeneeseaeeeaeeeeeeeaeeeaeeeeeseeeesaeeseeeeeeeeaaes 105 Removing a DNS Server AAA 105 Removing a Domain Name from the DNS Suffixes List A 106 PROUT Be EE 106 Adding Routing Information eeesceeeseeeeesneeeeeeeeeeeeeeeeeaeeeseaaeeeesaeeeseaeeeeeeaeeeseaeeeseeeeseaaeeessaeeeseaeeenenneeneaes 106 Editing Gieifutetlyiteiu ue 107 Deleting Routing Information 108 Configuring a Direct Connection Between the DataFRAME and an EBSD Host A 108 Configuring DataFRAME Communication eccccecceeeeeeeeeeeneeeeeeeeeeeceaeeeaeeeeeseaeeseeeecieeseaeeeeesiaeseaeeseaeeeaeeeeeeenaeee 109 Selecting the Interface Used by the Storage System Software eeeceesseseeeeeeeeeeeeeeeeseeeseaeeeneeseaeeeaeetee 110 Updating the List of Manager IP Addresses AAA 110 Chapter 5 Setting the Date and Time cccssseeeeeesseeeeeeesseeeeesseneeeeeeeseeeeenseeneeeenseeneeeeees 113 Reset Management Group Time EE 113 Getting TINGS aasien r e E Eaa esi R gees vesasiaensigedycesesennaec a E E 113 Setting the DataFRAME Time Zone 0 cecceecceeeceeeeeeeeeeeeaeeeeeeeeeeeeeeeaeeeaeeeeeeesaeeseaeseaeesaeseaeesaeeesaeeeeaeeeeeseaeeeneeees 114 Setting DataFRAME Date and Time eeeeeecceceeeneeeeeeeeeeteeeeeeeeeseeeaeeeaeeeeeeesaeeseaeeeaeesaeseaeesaeeseaeeeeaeseeeseaeeeneeees 114 Setting the Date and Time ooo cece eeeceeeceeseeenee
133. Es in the management group Information available on the Times tab includes e Current time setting of the management group MPC DataFRAME 440 Storage System Software User Manual 167 Working with Management Groups e Current time setting of each DataFRAME in the management group Remote Snapshot Tab The Remote Snapshot tab lists details for the remote snapshot including e Primary and remote management groups and snapshots e Copy rate percent complete and status information Editing a Management Group When editing a management group you can change the local bandwidth The local bandwidth setting controls the copy rate within the local management group Therefore it sets the data restripe rate for the management group If you use Remote Copy between two clusters within one management group local bandwidth will also control the remote copy rate For more information about setting the bandwidth for Remote Copy see the Remote IP Copy User Manual Note When Remote Copy is used to copy a snapshot from one management group to another the remote bandwidth setting of the management group containing the remote volume determines the rate per second that the manager will devote to copying data Setting or Changing the Local Bandwidth After a management group has been created you can edit the management group to change the local bandwidth This is the maximum rate per second that a manager will devote to non application processing such as
134. Ges Sege Stes ete ened tesa ited ede acne 85 Physical and Logical Interfaces AAA 86 How Active Backup Note use aenid nc Wierik idied ieee ie vee 86 How NIC Aggregation Works A 90 Creating a NIC Bond ENEE 93 Viewing the Status of a NIC Bond c eeceecceeeceseeeeeeeeeeeeeaeecaeeeeaeesaaeseaeeeaeeeeaeeeaeeseaeseaeeeeeesaeeseeeeeaeesieeenaeesaes 96 Deleting a NIC BONA gedeckt deed Eege eatae ea a esaii e abide deed 97 Disabling a Ee lee 98 Disabling a Network Interface AA 98 Configuring a Disabled UE 98 TOP ET 99 The TGP Status KC EE 99 Editing the TCP Speed and Duplex ceecceeseeseeeeeeeeeneeeeaeeeaeeteeeeeaeeeaeeeaeesaeeseaeesaeseaeesseeseaeeseeeteaeeseaeeeaees 100 Se Il 100 Beet Plate egereegtereerersberee geegent degeE Seefe 100 Editing the NIC Frame Size oo eeeeceecceceseeeeeeeeeeeeeeeeeeeseaeeeaeseaeeseeeeseeeesaeseaeeseeseaeesaeseaeeesaeseaeeeneeesaeeseaeeeaees 101 Beer PRaCtiCe sesachegiicdveasdecacd habs eae A aE cave a rarei On E aaa EN eee E Eaa sue iagesespeagearanee 102 Using a DNS SOrve nr eier freie eege anaa a eaa oaa aaia i ae aa aeaaaee aaa enia eede 103 DNS and DHCP EE 103 DNS and Static IP Addresses AA 103 Adding the DNS Domain Name ccccesceeeceeseeeeeeeeeeeteaeeeeeeeeeseeeeeaeeeseeseaeeeaeeseeeeseaeeeaeeeaeesiaeesieetiaeeneeeeenees 104 Adding a DNS Server EE 104 Adding Domain Names to the DNS Suffixes AA 104 EGitingia DNS Server 2s ecctedscecescvescs eaor ironiese oni Eea SER SEn EON
135. Group Name your management group and select DataFRAMES mpc Management Group Name Group This name cannot be changed after the management group is created Available DataFRAMEs Host Name IP Address Version Logged In You may have to login to an DataFRAME to continue To continue click Next 12 MPC DataFRAME 440 Storage System Software User Manual Getting Started 33 Enter a cluster name in the Cluster Name field Ensure that the I do not want to use a virtual IP address for this cluster checkbox is checked Click on Next to continue See Working with Clusters on page 191 for additional information on clusters IT x Create Cluster m Name your cluster and select DataF RAMES Weg Cluster Name cluster This name cannot be changed after the cluster is created r Virtual IP IP Address A cluster of multiple DataFRAMEs i r requires a virtual IP to configure Sune Mask failover for iSCSI initiators that do Default Gateway rc not support multiple IP addresses per target C I do not want to use a virtual IP address for this cluster Available DataFRAMEs Host Name IP Address Version Logged In To continue click Next 34 Enter the Volume Name Description Size and Replication Level for your storage system Click on Finish to continue x Create Volume m Name your volume and choose a size appropriate for its
136. IC Frame Size on page 101 1 On the Configuration Interface main menu tab to Network TCP Status and press Enter The Available Network Devices window opens shown in Figure 210 i if C C Figure 210 Available Network Devices Window 2 Tab to select the network interface for which you want to set the TCP speed and duplex and press Enter The Network TCP Status window opens shown in Figure 211 Figure 211 Setting the Speed Duplex and Frame Size 3 To change the speed and duplex of an interface tab to a setting in the Speed Duplex list 4 To change the frame size select Set To in the Frame Size list Then tab to the field to the right of Set To and type a frame size The frame size value must be between 1500 bytes and 9000 bytes 5 On the Network TCP Status window tab to OK and press Enter 6 On the Available Network Devices window tab to Back and press Enter MPC DataFRAME 440 Storage System Software User Manual 307 Using the Configuration Interface Removing a DataFRAME from a Management Group Removing a DataFRAME from a management group deletes all data from the DataFRAME clears all information about the management group from the DataFRAME and will reboot the DataFRAME Warning Removing a DataFRAME from a management group deletes all data on the DataFRAME 1 On the Configuration Interface main menu tab to Config Management and press Enter The Configuration Management window opens shown i
137. K The DataFRAME joins the management group and starts the manager The Console displays the newly created management group shown in Figure 115 E MPC computers DataFRAME Management Console File Find Tasks View Help a Available DataFRAMEs Group_0 DF420 DF420 1 Available DataFRAMEs Management Groups Alert Messages EBSD Hosts These DataFRAMEs are available for use Name I IP Software Version Model DF420 1 192 168 0 4 6 3 33 0007 0 SSR316MJ2 DF420 192 168 0 3 6 3 33 0007 0 SSR316MJ2 Figure 115 New Management Group with One DataFRAME Adding Additional DataFRAMEs When Creating a Management Group 1 Select the next DataFRAME that you want to add 2 Click the Tasks button on the Management Group Information tab and select Add to New or Current Management Group The Add to or create a Management Group window opens with the existing management group selected 3 Click Add The DataFRAME is added to the specified management group 4 Repeat steps 1 through 3 to add additional DataFRAMEs Adding Managers to the Management Group After adding the DataFRAMEs to the management group you can start managers on the additional DataFRAMEs in the management group The number of managers you start depends upon the overall design of your storage system See Managers on page 158 for more information about how many managers to add 1 Select a DataFRAME in the management grou
138. KI intended use Volume Name Volume This name cannot be changed after the volume is created Description Size 200 c5 we Replication Level None e Replication Priority To create the volume click Finish 35 Select the IP address for the DataFRAME and click on OK if asked for the Manager IP address MPC DataFRAME 440 Storage System Software User Manual 13 Getting Started 36 A summary screen displays showing you that the volume has been successfully created Click on Close to continue Create Volume Wizard EI Create Volume Summary The new volume has been successfully created mpc You have just completed the Create Volume Wizard Please review the volume details below Management Group Group Cluster Cluster Volume Name Volume Description None Size 200 GB Hard Threshold 200 GB Soft Threshold 200 GB Replication Level None Replication Priority Availability You can edit the volume settings from the DataFRAME Management Console Repeat this Wizard Run the Access Volume Wizard 37 Click on OK at the Wizard Launch Pad and you should see the volume listed in the main window This means that it has been successfully created Refer to the remainder of this manual for detailed information on managing and configuring your storage system Configuration Tasks Complete the following tasks to configure DataFRAMEs and create clusters and vo
139. ME from the cluster Repair or replace the DataFRAME Optional Add the new or repaired DataFRAME to the cluster Wait for the restripe of the volume to finish oOo N OO F amp F WwW DY Edit the volume 9 Change the data priority from data availability to data redundancy 10 Restore the clients access to the volume See Editing a Volume on page 229 for detailed requirements Deleting a Volume Delete a volume to remove that volume s data from the DataFRAME and make that space available When deleting volumes you must delete all snapshots of that volume before you can delete the volume itself Warning Deleting a volume permanently removes that volume s data from the DataFRAME Prerequisites e Delete all snapshots of the volume that you want to delete e Stop applications from accessing the volume e Disable the drives on the host Use these steps to delete the volume 1 Select the volume you want to delete The volume Tab View opens 2 Click Delete Volume A confirmation window opens 3 Click OK The volume is removed from the cluster MPC DataFRAME 440 Storage System Software User Manual 233 Working with Volumes Selecting a Volume or Snapshot from the List The Select Volume list window opens when you select the Tasks menu and then one of the following choices e Tasks gt Volume e Tasks gt Snapshot or Snapshot Schedules New Snapshot New Snapshot Schedule e Tasks gt Remote Copy
140. Means Plan for capacity Scheduling snapshots should be planned with careful consideration for management capacity management as described in Managing Capacity Using Volume and Snapshot Thresholds on page 237 Pay attention to how you want to retain snapshots and the capacity in the cluster If you want to retain lt n gt snapshots the cluster should have space for lt n 1 gt It is possible for the new snapshot and the one to be deleted to coexist in the cluster for some period of time If there is not sufficient room in the cluster for both snapshots the scheduled snapshot will not be created and the schedule will not continue until an existing snapshot is deleted Synchronize The time setting on the DataFRAMEs running managers and the time DataFRAME times with setting of the management group must be synchronized If they are not management group time synchronized then the snapshot schedule might run incorrectly Be sure to configure the correct time on the DataFRAMEs and then reset the management group time See Chapter 5 Setting the Date and Time Also see Resetting the Management Group Time on page 171 and Creating Snapshot Schedules You can create one or more snapshot schedules for a volume For example one schedule could be for daily snapshots intended for backup and recovery A second schedule could be for weekly snapshots used for data mining 1 Select the volume for which you want to sched
141. NUAL Data Copied 291 MB State Complete COPYING Data Remaining 1 15 GB 37 6 Current Rate 1 445 Kbisec Avg Rate 1 308 Kb sec KS Figure 230 Remote Snapshot Details for a Remote Copy in Progress Scheduling Remote Snapshots Scheduled remote snapshots provide high availability for business continuance disaster recovery and provide a consistent predictable update of data for remote backup and recovery The first step in scheduling remote snapshots is planning for creating and deleting primary and remote snapshots Issues that require planning include e Recurrence frequency MPC DataFRAME 440 Storage System Software User Manual 351 Remote Copy e Snapshot thresholds e Retention policies For detailed information about these issues see Planning for Remote Copy on page 334 Once you have defined your plan you are ready to create the remote snapshot schedule These are the basic steps you will follow 1 Create the schedule 2 Configure the primary volume and snapshot 3 Create the remote volume and configure remote snapshots Detailed instructions are provided in the following sections Creating the Schedule 1 Right click the volume for which you want to create the remote snapshot schedule and then select Remote Copy gt New Remote Snapshot Schedule The New Remote Snapshot Schedule window opens shown in Figure 231 2 Type a name for the schedule 3 Optional Type a description for the
142. Name Must be from 1 to 127 characters Names are case sensitive Hard Threshold Hard threshold size must be equal to or less than the size of the volume Soft Threshold Soft threshold size must be equal to or less than the hard threshold size 7 Click OK The Roll Back Volume confirmation message opens 364 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Storage System Console x A Rolling the volume back to this snapshot will delete the volume Before proceeding you must manually delete all snapshots of the volume that are newer than this snapshot To preserve the volume and snapshots that you need to delete create a copy of the volume and snapshots before rolling the volume back Are you sure you want to roll the volume back to this snapshot Cancel Figure 239 Verifying the Primary Volume Roll Back 8 Click OK The primary snapshot version of the primary volume is restored as a read write volume 9 Reconfigure application servers to access the new volume Rolling Back a Remote Volume A remote volume cannot be rolled back In order to roll back a remote volume you must make the remote volume into a primary volume Using Remote Snapshots for Data Migration and Data Mining Use remote snapshots to create split mirrors for data mining and data migration A split mirror is a one time remote snapshot created from the volume containing the data you want to use or move Split
143. Name and Settings for an Interface s sssesesnesesseerereererenrerrnrrernsrnrnernrrnrrnrnernrnsrnnns 305 Selecting a Bonded Interface in the Available Network Devices Window c sseesceeeeeeeteteneeteees 306 Deleting a NIC Bonn EE 306 Available Network Devices Wimdow AAA 307 Setting the Speed Duplex and Frame Glze 307 Removing the DataFRAME from a Management Group 308 Resetting to Factory Defaults 2 2 0 0 eceeceseceesseeeeeneeeeceeeeeeeeeeeeseeeeseesensaaseeeseneeseeeeeseseeeeeseeeeeeeneeseneaeeees 308 Basic Flow of Remote Copy ssn acted ace deeg ined dec A ees E O Eoi Ea S 331 Icons Depicting the Primary Snapshot Copying to the Remote Snapshot A 332 Icons for Remote Copy as Displayed in the Graphical Legends Window ssssssssssnssneneennnerneerrrsenn 332 Creating a New Remote Gnapehot AA 339 MPC DataFRAME 440 Storage System Software User Manual Figure 218 Figure 219 Figure 220 Figure 221 Figure 222 Figure 223 Figure 224 Figure 225 Figure 226 Figure 227 Figure 228 Figure 229 Figure 230 Figure 231 Figure 232 Figure 233 Figure 234 Figure 235 Figure 236 Figure 237 Figure 238 Figure 239 Figure 240 Figure 241 Figure 242 Figure 243 Figure 244 Figure 245 Figure 246 Figure 247 Figure 248 Figure 249 Figure 250 Figure 251 Figure 252 Figure 253 List of Figures Creating a New Primary Gnapehot naene nnt 340 New Primary Snapshot Created 341 Completing
144. P Figure 171 illustrates the configuration for a single host authentication with 2 way CHAP required New Authentication Group x New Authentication Group iSCSI Initiator Properties xj General Discovery Targets Persistent Targets Bound Volumes Devices Name AuthGroup_0O The iSCSI protocol uses the following information to uniquely identify this initiator and authenticate targets Description Initiator Node Name ign 1991 05 com micrasoft kwegner wxp Volume List No Volume List iSCSI EBSD Fibre Channel To rename the initiator node click Change Change Ki Allow access via iSCSI To authenticate targets using CHAP click Secret to Authentication specify a CHAP secret Secret To configure IPSec Tunnel Mode addresses click FE I CHAP Secret Setup x The secret allows the initiator to authenticate targets when performing mutual CHAP Targets must also be configured with this initiator secret CHAP not required CHAP required Enter a secure secret CHAP Name iqn 1991 05 com microsoft l kwegner wxg eooooooocoooooo Target Secret preneren Initiator Secret a Cancel Figure 171 Adding an Initiator Secret for 2 way CHAP Shown in the MS iSCSI Initiator Warning Allowing more than one iSCSI application server to connect to a volume could result in data corruption MPC DataFRAME 440 Storage System Software User Manual
145. P IP interfaces DNS servers and the routing table Working with DataFRAMEs e DataFRAME Administration The DataFRAME comes configured with 2 default groups and 2 default users All administrative users and groups are added and managed locally e SNMP Monitor the DataFRAME using an SNMP Agent You can also enable SNMP traps e Reporting The DataFRAME offers multiple reporting capabilities including real time statistical information active monitoring of selected variables and diagnostics Logging In to the DataFRAME After finding all the DataFRAMEs on the network you must log in to each DataFRAME individually to configure modify or monitor the functions of that DataFRAME 1 On the Network view double click the DataFRAME that you want to log in to The Log In window opens shown in Figure 28 x Log in to DataFRAME DF 140 192 168 0 7 User Name Password Cancel Figure 28 Logging in to a DataFRAME 2 Type the User Name and Password 3 Click Log In When you are successfully logged in the DataFRAME configuration window opens to the Module configuration category shown in Figure 27 and Figure 28 Logging In to Additional DataFRAMEs Once you are logged in to a DataFRAME you can log in automatically to additional DataFRAMEs configured with the same user name and password by double clicking those DataFRAMEs in the Network view If you try to log in to a DataFRAME that uses a different user
146. PAP Network Sn DataFRAME Adminisi ge SNMP IS Reporting Reboot Power Off in 5 minutes Reboot Figure 37 Shutting Down or Rebooting the DataFRAME 3 Select Reboot 4 In the minutes field type the number of minutes before the reboot should begin You can enter any whole number greater than or equal to 0 If you enter 0 the DataFRAME will reboot as soon as you complete step 6 5 Click Reboot A confirmation message appears 6 Click OK The DataFRAME will reboot in the specified number of minutes When reboot actually begins the DataFRAME disappears from the Network View The reboot takes 3 to 4 minutes 7 Search for the DataFRAME to reconnect the Console to the module once it has finished rebooting See Finding by Subnet and Mask on page 21 or Finding by Module IP or Host Name on page 23 MPC DataFRAME 440 Storage System Software User Manual 49 Working with DataFRAMEs Canceling a Reboot 1 When you click OK the Reboot button changes to Cancel as shown in Figure 38 Edit Configuration for DF140 192 168 0 7 2 V Power Off Feature Registration Boot Devices W Storage Time ro TCP IP Network B Module Information Backup amp Restore DataF RAME Adminis gt SNMP Zi Reporting Figure 38 Canceling the DataFRAME Reboot 2 Click Cancel to stop the reboot A confirmation message opens 3 Click OK The Power Off window returns with
147. Read Only Read wWrite Read Mitte SE SE Remote i De Volume Storage L SS Subnet Remote Snapshot Backup Site Production Site Figure 246 Restoring from a Remote Volume MPC DataFRAME 440 Storage System Software User Manual 373 Sample Remote Copy Configurations Application Workstations Production i Production Application Application Server Application Servers Primary 3 Remote fai Remote Primary Ha 3 Volume iVolume 1 i Volume 2 Volume Storage Subnet Primary Remote Remote Primary Snapshot Snapshot 1 Snapshot 2 Snapshot Production Site 1 Backup Site Production Site 2 Figure 247 Restoring from Tape Backup Best Practices Select a recurrence schedule for remote snapshots that minimizes the potential for data loss Any data written to the primary volume since the most recent remote snapshot was created will be unavailable if the primary volume is unavailable Consider how much data you are willing to lose in the event of an emergency and set the recurrence for creating remote snapshots accordingly If you do not want a large number of remote snapshots to accumulate on your remote volume you can use more than one remote snapshot schedule each with different retention policies For example suppose you want to create remote snapshots every 4 hours to ensure that no more than 4 hours worth of data is lost in an emergency In addition you want to retain 1 week s worth of rem
148. Remote WARNING Changing this primary volume to a remote volume can result in data loss if any hosts are accessing this volume Disconnect any hosts before completing this change Volume CreditData05 in management group Transaction_Data is currently a primary volume When it is changed to a remote volume a snapshot will be created with the name below to save any data currently on the volume Do you want to make this volume a remote volume Snapshot Snapshot Description Snapshot_1 Cancel Figure 235 Creating a Snapshot Before Making a Primary Volume into a Remote Volume 5 Type a name for the snapshot that will be created This snapshot preserves any existing data on the volume 6 Optional Type a description for the snapshot 7 Click OK The snapshot is created and the volume becomes a remote volume The Edit Remote Volume window opens again with the editable fields enabled as shown in Figure 236 MPC DataFRAME 440 Storage System Software User Manual 357 Remote Copy Edit Remote Yolume Edit Remote Volume Type Primary Volume Name Description Created Cluster TransData_BkUp Replication Level None Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Figure 236 Finalizing the New Remote Volume 8 Optional Make any necessary changes to the new remote volume Making a Remote Volu
149. Remote Snapshots on page 348 Remote Snapshot Schedules Target Information Volume List Memberships Details Snapshots i Snapshot Schedules Hosts f Remote Snapshot Primary Man _ Primary Sna Remote m Man Remote Sna _ Complete Fass Time Data Copied Rate L State Transaction Snapshot_0 Transaction Snapshot_1 100 13 0 25 MB 157 Kbisec Complete Figure 224 List of Remote Snapshots Setting the Remote Bandwidth The remote bandwidth sets the maximum rate for data transfer between manage ment groups The remote bandwidth setting is the upper limit of the range of data transfer that is the copy rate will be equal to or less than the rate set The remote bandwidth specifies the speed at which data is received from another management group This means that to control the maximum rate of data transfer to a remote snapshot set the remote bandwidth on the management group that contains the remote snapshot 1 Right click the remote management group and select Edit Management Group The Edit Management Group window opens shown in Figure 225 MPC DataFRAME 440 Storage System Software User Manual 345 Remote Copy Edit Management Group Edit Management Group Group Name Local Bandwidth 32768 Remote Management Groups Management Gr Loggedin Remote Bandwi Connected Transaction_D Yes 32768 Kbisec Yes Disassociate Edit Bandwidth
150. SCSI EBSD Fibre Channel Zi Allow access via EBSD Authentication Hosts with the following subnets and masks Subnet Add Edit Delete Hosts with all subnets and masks Figure 179 Configuring EBSD for New Authentication Group Authenticate Hosts with Specific Subnets and Masks 1 In the Authentication box select Hosts with the following subnets and masks 2 Click Add The Add Subnet and Mask window opens shown in Figure 180 276 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Add Subnet and Mask i xj Subnet Mask 255 255 255 0 K Ess Figure 180 Adding a Subnet and Mask for EBSD Host Authentication 3 Type in the subnet 4 Select the appropriate mask 5 Click OK 6 Optional Repeat for additional subnets and masks Authenticate Hosts with All Subnets and Masks 1 In the Authentication box select Hosts with all subnets and masks Table 50 Choosing the Level of Access for Hosts using the EBSD Driver Authentication Method What Happens No hosts No application server gains access Hosts with the following subnets and masks Only hosts on the designated subnet and mask gain access If selecting this method 1 Click Add 2 Enter a subnet and mask 3 Click OK All hosts on the network All hosts gain access Finishing E
151. SD device If an application attempts to write to a disconnected or disabled EBSD device raw or file the application will hang Use these steps to unmount the EBSD disk 1 Execute the command umount For example umount mnt ebsdo 2 Run ebsdvm remove where the device you want to disconnect Remember device numbers can be from 0 63 This disconnects the DataFRAME volume from the host device entry Disabling an EBSD Device 326 Disabling the EBSD device keeps the device from being accessed takes it offline saves the data on the device and maintains the device information in ebsd conf 1 Modify the etc ebsd conf file Set the enabled field for that device to false as shown in the sample entry below Sample Device Entry as Disabled Device HHH HHH HHH HHH RN Sample device entries deviceOd type volume client name myclient ip bind 10 0 1 63 management group my_mgtgroup auth group public volume name my_volume_0 access mode rw use_unicast true unicast_list 10 0 0 12 10 0 0 13 enabled false use multicast false MPC DataFRAME 440 Storage System Software User Manual Using the EBSD Driver for Linux Deleting an EBSD Device Deleting the EBSD device erases the volume s data from the DataFRAME Warning Make sure the EBSD disks you plan to delete are NOT in use Deleting EBSD disks from the host and deleting volumes from the cluster removes all data stored in thos
152. Saving the Report to a File 1 On the Passive Tab click Save To File to download a text file of the reported statistics The Save dialog opens 2 Choose the location and name for the report 3 Click Save The report is saved with a doc extension It is a text file and will open with Word in Windows or any text editor in Linux or UNIX Passive Reporting Detail 140 This list details selected information available on the Passive Reporting window Not all items are listed here Table 29 Selected Details of the Passive Report This Term Means This Passive Report Date and time report created Host Name Host name of the DataFRAME IP Number IP address of the DataFRAME DataFRAME Software Full version number for DataFRAME software GUI Software Full version number for the Console Support Key Support Key is used by a Technical Support representative to log in to the DataFRAME Boot Devices flash 0 flash 1 Status information about the compact flash card s used to boot the DataFRAME MPC DataFRAME 440 Storage System Software User Manual Reporting Table 29 Selected Details of the Passive Report This Term Means This NIC Data Information about NICs in the DataFRAME Card Indicates which NIC in the list is being described Description Card name manufacturer and capable speed of the NIC MAC Address Physical address of the NIC Each card has a un
153. See Measuring Disk Capacity and Volume Size on page 212 for a detailed explanation of disk capacity allocation in a cluster J J Details Volumes DataFRAMEs Disk Usage Usage Graphs iSCSI Remote Snapshot Type Name Space Max Used Max Full Misc Cluster Cluster_0 2 61 TB 0 25 MB 0 0 Allocated DataFRAME DF140 1 889 65 GB OMB 0 DataFRAME DF140 889 65 GB OMB 0 DataFRAME DF420 3 48 TB 0 25 MB 0 PREY Mex Used and Max Full can temporarily exceed 100 during data migration asks wax Full for snapshots is Max Used Writable Max Used Hard Threshold Wititable Hard Threshold Figure 141 Statistics for a Cluster Note Data migration referred at the bottom of the Disk Usage tab occurs during a snapshot deletion when snapshot data returns to top level or when moving a DataFRAME into or out of an active cluster The usage table provides the following information Table 33 Disk Space use Reported on Disk Usage Tab Column Heading Information Reported Space Space is block level raw space e Cluster e Total raw space in the cluster e DataFRAME Total raw space on the DataFRAME Volume e Hard threshold of the volume e Snapshot e Hard threshold of the snapshot and of the writable snapshot Max Used Maximum block level space ever written to e Cluster e Total raw space ever used in the cluster e DataFRAME Total raw space ever
154. Setting the DataFRAME Time Zone You must set the time zone whether or not you use NTP Set the time zone for the physical location of the DataFRAME HTTP files display the time stamp according to this local time zone 1 Choose the time zone for the location of the DataFRAME 2 Click Apply Setting DataFRAME Date and Time If using NTP the NTP server controls the date and time for the DataFRAME See Using NTP on page 116 Note Even if you are using an NTP server you can set the date and time manually If the difference between the date and time on the DataFRAME and the date and time on the NTP server is too large the NTP server will not change the date and time on the DataFRAME To ensure that the NTP server is able to control the DataFRAME date and time first set the date and time manually 114 MPC DataFRAME 440 Storage System Software User Manual Setting the Date and Time Setting the Date and Time 1 If you are not using an NTP server make sure NTP mode is set to Off 2 Click the DataFRAME Date and Time tab to bring it to the front shown in Figure 76 GBeait Configuration for DF140 192 168 0 7 Module T Storage E NTP Mode TOPE NEWOIK NTP is off To turn NTP on add an NTP server to the NTP tab below Sm DataFRAME Administ gi SNMP fa Reporting Time Zone Date and Time NTP Enter the current date and time for the DataFRAME time zone DataFRAME Time Zone Greenwich Mean Time
155. Soft Threshold Auto Grow Off auto Manual Checksum Data Yes No Figure 223 Creating a New Remote Volume 4 Type a name for the volume A volume name must be from 1 to 127 characters and is case sensitive 5 Optional Type a description of the volume 6 Select the replication level You can set different replication levels for the remote volume and the primary volume Note You cannot set the size or thresholds for the remote volume Those values are 0 since the remote volume is a placeholder for data 7 Select a replication priority If you select a replication level of None you cannot set a replication priority See Chapter 12 Working with Volumes for detailed information about creating volumes 8 Select the Auto Grow and Checksum Data options 9 Click OK to return to the New Remote Snapshot window The new remote volume has been created at this point 344 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Viewing a List of Remote Snapshots You can view a list of remote snapshots associated with management groups clusters volumes or snapshots 1 Click the item for which you want to view the list of remote snapshots 2 Click the Remote Snapshot tab The tab view opens shown in Figure 224 The report on the tab lists both management groups and all the snapshots The other columns report status information about the remote snapshots as described in detail in Monitoring
156. Status Passive Ready Data Transfer Yes Data Transfer No Motherboard Pom Fails Preferred Yes Preferred No Data Transfer Fails Over to Status Passive Failed Status Active piel rome Data Transfer No Data Transfer Yes Motherboard Port Preferred Yes Preferred No Restored Status Active Status Passive Ready Data Transfer Yes Data Transfer No Example Network Configurations with Active Backup Two simple network configurations using active backup in high availability environments are illustrated 88 MPC DataFRAME 440 Storage System Software User Manual Managing the Network Active Dow ca Storage Module Cluster Figure 59 Active Backup in a Two switch Topology with Server Failover The two switch scenario in Figure 59 is a basic yet effective method for ensuring high availability If either switch failed or a cable or NIC on one of the DataFRAMEs failed the active backup bond would cause the secondary connection to become active and take over MPC DataFRAME 440 Storage System Software User Manual 89 Managing the Network SeverA SererB Active et Passive Storage Module Cluster Figure 60 Active Backup Failover in a Four switch Topology Figure 60 illustrates the active backup configuration in a four switch topology How NIC Aggregation Works NIC aggregation allows the DataFRAME to use both interfaces simultaneously for data transfe
157. Therefore a DataFRAME must contain an even number of disks to configure RAID 1 10 RAID 5 50 RAID 5 50 requires sets of disks to configure e MPC DataFRAME 440 Storage System requires sets of 4 or 8 disks up to a total of 16 disks RAID Set Size The RAID set size is limited to 2 TB This means that the combined capacity of the disks participating in the RAID set cannot exceed 2 TB RAID 5 50 in the DataFRAME 440 Storage System The following table illustrates RAID 5 50 capacity calculations for three different disk capacities in the DataFRAME 440 Storage System Since 2 TB equals 2048 GB the RAID 5 50 configuration available for the 400 GB disks is 5 plus a spare RAID 5 50 is not supported for 500 GB disks For DataFRAME a i 440 Storage Using Disk Capacity of System bar iat 250 GB 400 GB 500 GB ize 5 plus a spare 1250 GB 2000 GB 2500 GB 6 disks 1500 GB 2400 GB 3000 GB 1 1 Parentheses indicate RAID set size greater than 2 TB RAID 0 RAID 0 creates a striped disk set Data will be stored across all disks in the RAID which increases performance However RAID 0 does not provide fault tolerance If one disk in the set is powered down or fails all data on the set will be lost MPC DataFRAME 440 Storage System Software User Manual 59 Storage DataFRAME capacity in RAID 0 is equal to the total capacity of all disks in the module RAID 1 and RAID 10 RAID 1 RAID 1 provides data redundancy by
158. To complete the request for email notification that you configured for monitored variables 1 In the Reporting category select the Email tab The Email window opens shown in Figure 106 edit Configuration for DF140 192 168 0 7 Ze Module d Passive Logfiles Active Email Diagnostics Alerts i Storage Time i TCP IP Network Sn DataFRAME Adminis g SNMP Si SMTP Settings Server IP Host name Port 25 Test Email Sender Address if Log Out Figure 106 Configuring Email Settings for Email Alert Notifications 2 Enter the IP or host name of the email server 3 Enter the email port MPC DataFRAME 440 Storage System Software User Manual 151 Reporting The standard port is 25 4 Optional If your email server is selective about valid sender addresses on incoming emails enter a sender address for example username company com If you do not enter a sender address the From field of email notifications will display root hostname where hostname is the name of the DataFRAME 5 Click Apply Note Notification of undeliverable email messages are sent to the sender address Note If you are requesting email notification be sure to set up the email notification in Active monitoring Running Diagnostics Use diagnostics to check the health of the DataFRAME hardware Note Running diagnostics can help you to monitor the health of the DataFRAME or to
159. aFRAMEs in the cluster For example 2 Way Replication A cluster with three DataFRAMEs configured for 2 way replication There have been five writes to the cluster Figure 148The figure illustrates the write patterns on the three DataFRAMEs MPC DataFRAME 440 Storage System Software User Manual 215 Working with Volumes Store 1 Store 2 Store 3 Figure 148 Write Patterns in 2 way Replication Replication Priority Set the replication priority according to whether data availability or data redundancy is the goal for the volume e Availability Priority If data availability is your priority you can set any replication level e Redundancy Priority If redundancy is the priority you must select either 2 way or 3 way replication Note When you have volumes with a priority of redundancy you cannot use a hot spare in the cluster Volume is available to a client with a replication level of anda priority None 2 way 3 way setting of Availability AllDataFRAMEs One of every two One of every three must be up consecutive consecutive DataFRAMEs must be DataFRAMEs must be up up Redundancy N A All DataFRAMEs must Two of every three be up consecutive DataFRAMEs must be up Warning A management group with three DataFRAMEs is the minimum configuration for fault tolerant operation Although the system allows you to configure 2 way replicati
160. aan eee ii a Aedes 352 Configuring the Primary Volume and Snapshots ecceeeceeeseeeeeeeeeeeeeeeeeeeeeeeseaeeeeesneeeseeeseaeeseaeseaeereetes 353 MPC DataFRAME 440 Storage System Software User Manual xiii Contents Configuring the Remote Volume and Snapshots eeeeeseeeeeeeeeeeeeeeeeeeeeeseaeeeeeeeeaeesaeeeeeteaeeeeeeeeaeesas 353 What the System Does oo ceecececceeseeseeeeneeeeeeeeaeeseaeeeaeeeaeeseaeeeseeseaeeeaeeeseeesaeeesaeseaeeseeseaeeseeseeescaeeeaeeeneeesas 354 Editing a Remote Snapshot Schedule AAA 354 Deleting a Remote Snapshot Schedule o eeeeeeeeceeeeeeeeeeeeeceeeeeeeeeeeseaeeeaeesaeeeseeeseaeeseaeseaeesaeeseeeeseeeteeeeieeeneeenaes 355 Changing the Roles of Primary and Remote Volumes cecceesceseeeeeeeeeeeeeeeeeeeeeesaeeeaeeeeeeenaeesaeeseaeescaeesneeeneeesas 355 Making a Primary Volume Into a Remote Volume eesceeeeseeeeneeeeeneeeeeeeeeeneeeesaeeeeeaaeeseeaeeeseneeeeseneereaes 356 Making a Remote Volume Into a Primary Volume 00 0 ee eseeeeeceeeeeneeeeeeeeeeeneeetenaeeeesaeeeesaneeseaeeeeeneeeeseneereaas 358 Designating Size and Threshold Values for the Converted Volume e ecseeceeeeee eee teeter teeeeeeteateees 359 Configuring ST EE 360 Planning allover a cccecsscccussiceeteicsgaazacesdszayestencdendeesdguadk cexddasasiens hag adedass secu usd cgutecevesasncassusseenssecuevenessvaceteesdes 360 Using Scripting for Failover snes eee ececceeeeeeeseeeeeeeeeeeeeseeeeeeeeeseeseaeeeaeeseeee
161. ab For iSCSI volumes displays the virtual IP address if configured and lists any iSNS servers For Fibre Channel volumes displays the Fibre Channel Port World Wide Name WWPN for that volume Volume List Memberships Tab Lists the volume lists and which volumes and authentication groups they are connected to 228 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Editing a Volume When editing a volume you can change the description replication level replication priority size hard and soft thresholds the cluster that contains the volume whether the volume is configured for auto grow and whether checksumming is enabled Table 38 Requirements for Changing Volume Parameters Item Requirements for Changing Description Must be from 0 to 127 characters Cluster The target cluster must Reside in the same management group e Have sufficient unallocated space for the hard threshold and replication level of the volume being moved When moving a volume to a different cluster that volume will temporarily exist on both clusters Replication Level The cluster must have sufficient DataFRAMEs and unallocated space to support the new replication level Replication Priority To change the replication priority the replication level must support the change You can always go from Redundancy to Availability However you cannot go from Availability to Redundancy unless a suff
162. aeeeaeeseaeeeeeeeeaeeseeeeseesieeseaeesieeees 180 Chapter 10 Disaster Recovery Using A Virtual Manager cccesseecceseseeeeeeseeeeeeeeeees 181 When to Use a Virtual Manager sisson ioeina E E EO EEE 181 Benefits of a Virtual Manager AAA 182 Requirements for Using a Virtual Manager AA 182 Configuring a Cluster for Disaster Recovery cecceecceeeneceneeeeeeeeeeeeaeeeeeeaeesaeeeeaeeeseeseaeeeaeesaeeseeesnaeeeieeeeaeeseeens 185 Best Practice coana a Gees tice hel a de de Sec ee 185 Configuration StepS oo eeceeccececeseeeeeneeeeeeeeeeseeeeeeeeeeaeeeneesaaeeeaeeseaeeeeeeseaeesaeeseeeeseeseaeeeeesiaeesaeessaeeeieeeeeeenaeees 185 Configuring a Virtual Manager ou eeceeeceeeeeeeeeeeeeeseeeeseeeeeaeeeneeeeeeeeaeecaeseaeesaaeseaeeseeeesieeseaeeeaeeseaeeeaeeseaeeeieeeeeeeeaeess 187 Adding a Virtual Manager EE 187 Starting a Virtual Manager to Regain Ouorum ou eeeeeceeeeeeeneeeeneeeeeeeeaeeseeeeeaeetaeeeeaeeeaeeeeaaeeeeeeeaeeseeeeeaeesiaeseaeeseeens 188 Starting a Virtual Manager ecceeecceeeeeeeeeeeeeeceaeeeeeeeeeesaeeeeeeecaaeseaeeeaeeeaaeeseeeseaeeeseeseeeeeaeeseeeseaeessaeseaeesneeees 189 Stopping a Virtual Manager 190 Removing a Virtual Manager iis asaccs cece ccccceatscetcches scene cdcavact as dcenten dacs aa Eia aa eaa siadeeedascrs Eienaars 190 Chapter 11 Working with Clusters ccccessseceeseeeeeeeneeeeeeeeseneeeeeseeeeeeesseeeeeseeeeeeeeensenes 191 Mixing DataFRAMEs of Different Capacities in Clusters oo
163. age System Software User Manual 203 Working with Clusters e Cluster with unavailable DataFRAME reducing the time out value will activate the hot spare earlier For example a DataFRAME is not available and the hot spare time out is configured for 6 hours After 3 hours you reduce the time out to 1 hour thinking that will activate the hot spare in 60 minutes However the hot spare activates immediately This is because the clock that is tracking the time out started when the DataFRAME became unavailable and it considers the 1 hour interval to have passed already e Cluster with unavailable DataFRAME increasing the time out value will activate the hot spare later For example a DataFRAME is not available and the hot spare time out is configured for 6 hours Four hours have passed You increase the time out to 8 hours adding an additional 2 hours to the time out interval before the hot spare activates To change the hot spare time out 1 Change the value for the hot spare time out 2 Click OK Removing a DataFRAME from a Cluster You can remove a DataFRAME from a cluster only if the cluster contains sufficient storage modules to maintain the existing volumes and replication level See Chapter 12 Working with Volumes on page 211 for details about editing volumes 1 Select a DataFRAME from the DataFRAMEs in Cluster list 2 Click Remove The DataFRAME moves to the DataFRAMEs Available list 3 Click OK when you are fin
164. ailure timeout seconds gt MPC DataFRAME 440 Storage System Software User Manual 259 Working with Scripting Mounting a Snapshot Below is an example of mounting the snapshot using ebsdvm ebsdvm lt mgmt group name gt lt snapshot name gt lt auth group name gt lt local ip gt lt number of managers gt lt each managers s ip gt lt lock_mode gt Example Joe Jones plans to mount his XRayReview snapshot and mount it on another server where the group named adminusers the administrators of the orthopedic section can access the images for filing the patient database ebsdvm Images XRayReview adminusers 10 0 20 212 3 10 0 13 79 10 0 33 47 10 0 33 87 ro Increasing Volume Hard and Soft Thresholds You can create a script that will automatically increase the hard and soft volume thresholds by a specific amount The operation is triggered when a soft threshold is reached It then raises both the soft and hard thresholds by the amount you specify in the script The thresholds will only increase e When there is sufficient room in the cluster to accommodate the increases or e To the point where the hard threshold equals the volume length whichever of these conditions occur first To increase space in the cluster by adding more DataFRAMEs or to increase the volume length follow instructions as described in Chapter 11 Working with Clusters or Chapter 12 Working with Volumes Scripting Automatic Threshold Increases
165. all data volumes on a single DataFRAME use RAID 1 10 or 5 50 to replicate data within the DataFRAME Using Volume Replication in a Cluster A cluster is a group of DataFRAMEs across which data can be replicated Volume replication across a cluster of DataFRAMEs protects against disk failures within a DataFRAME and failure of an entire DataFRAME For example if a single disk or an entire DataFRAME in a cluster goes down data reads and writes can continue because an identical copy of the volume exists on other DataFRAMEs in the cluster Clustering is part of the Scalability Pak feature upgrade See Working with Clusters on page 191 for more information Note If you plan to create data volumes that span two or more DataFRAMEs use replication in a cluster to ensure data safety and availability 64 MPC DataFRAME 440 Storage System Software User Manual Storage Using RAID with Replication in a Cluster Note If you use replication in a cluster to replicate volumes across DataFRAMEs then the redundancy provided by RAID 10 uses excess capacity and may not be necessary For example e Using replication up to 3 copies of a volume can be created on a cluster of 3 DataFRAMEs The replicated configuration ensures that 2 of the 3 DataFRAMEs can go down and the volume will still be accessible e Configuring RAID 10 on these DataFRAMEs means that each of these 3 copies of the volume is stored on 2 disks within the DataFRAME for a total
166. aluating the Configurable Snapshot Pak The Configurable Snapshot Pak includes programmable snapshots Features included are e Scheduled snapshots e Scripting for snapshots Starting the License Evaluation Period The Configurable Snapshot Pak 30 day evaluation period begins if you create a snapshot schedule Backing Out of the License Evaluation Period If you decide not to purchase the Configurable Snapshot Pak you must delete any snapshot schedules that you have configured 292 MPC DataFRAME 440 Storage System Software User Manual Feature Registration 1 Back up any volumes you plan to retain Table 53 describes how to safely back out of the Configurable Snapshot Pak evaluation Table 53 Safely Backing out of Configurable Snapshot Pak Evaluation Feature Being Evaluated Steps to Back Out Scheduled snapshots Delete the snapshot schedule Evaluating the Remote Data Protection Pak The Remote Data Protection Pak includes Remote Copy Features included are e Remote volumes e Remote snapshots e Remote snapshot schedules e Scripting for remote copy Starting the License Evaluation Period The Remote Data Protection Pak 30 day evaluation period begins if you create a remote volume by e Making an existing primary volume into a remote volume e Creating a remote volume in the process of creating a remote snapshot e Creating a new volume and selecting the Remote radio button on the New Volume dialog When a rem
167. amlessly span the DataFRAMEs in the cluster This chapter discusses e Mixing DataFRAMEs in a cluster e Hot spares e iSCSI and clusters e Creating and managing clusters e Repairing a DataFRAME Mixing DataFRAMEs of Different Capacities in Clusters Clusters can contain DataFRAMEs with different capacities However all DataFRAMEs in a cluster will operate at a capacity equal to that of the smallest capacity DataFRAME Prerequisites e All the DataFRAMEs in a cluster must be configured alike See DataFRAME Configuration Window on page 34 e Before you create a cluster you must have created a management group See Creating a Management Group on page 159 Using Hot Spares A cluster of DataFRAMEs can contain a hot spare DataFRAME A hot spare is a DataFRAME that is not used for data storage but stands by in case a DataFRAME in the cluster goes down at which time the hot spare is activated When the hot spare is activated in the cluster replicated volumes restripe onto the hot spare DataFRAME A hot spare DataFRAME is designated in the Console by the icon show below s Working with Clusters Note Hot spares do not provide protection for non replicated volumes There must be some copy of data to be restriped onto the new DataFRAME and non replicated volumes do not have copies In order to have more than one DataFRAME in a cluster beyond the 30 day evaluation period you must purchase the Scalability Pak See
168. anced Server 3 0 SMP the kernel version is 2 4 21 4 ELsmp The corresponding RPM is ebsd 6 x xx xXxxx 2 4 21 4 ELsmp i386 rpm 3 Configure the driver See Configuring the EBSD Driver for Linux on page 321 Querying An Existing RPM Package To determine whether an RPM package is installed run rpm q ebsd For example rpm q ebsd would return ebsd 6 x xx XXXX 1 To list the files in an RPM package run rpm ql ebsd For example rpm ql would return opt Storage System Storage System_Software 6 1 Drivers EBSD 2 4 21 4 ELsmp opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp etc opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp etc ebsd conf sample opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp init d opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp init d ebsd opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp license opt Storage System Storage System_Software 6 1 Drivers EBSD 2 4 21 4 ELsmp modules opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp modules ebsd o opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp readme opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp sbin opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp sbin ebsdvm opt Storage System Storage Sy
169. and Open PCI Slots on the Back of the MPC DataFRAME 440 Storage System Adding Interfaces to PCI Slots You can add interface cards to the PCI slots located to the right of the motherboard NIC ports on the back of the DataFRAME These interface cards can contain Ethernet or Fibre Channel ports The MPC DataFRAME 440 Storage System contains one open 64 bit 66 MHz PCI slot and two open 32 bit 33 MHz PCI slots The other three covered slots are occupied by Serial ATA cards MPC DataFRAME 440 Storage System Software User Manual 81 Managing the Network e The 64 bit PCI slot can hold a quad 4 port card e The 32 bit slots can hold dual 2 port cards To distribute bandwidth and to ensure fault tolerance connect to ports across more than one PCI slot For example connect to the first port in the first 64 bit PCI slot Then connect to the next port in the second 32 bit slot and connect to the third port in the third 32 bit slot Connect to the fourth port in the first slot and so on The following figure shows the optimal configuration of add on ports Add port to 64 bit slot first i 64 bit 32 bit 32 bit 66MHz 33 MHz 33 MHz Figure 56 Distributing Bandwidth and Ensuring Fault Tolerance of Add on Ports Across PCI slots Note When adding more than one port to the DataFRAME you can distribute bandwiath and ensure fault tolerance by distributing the ports across more than one PCI slot S
170. andwidth Figure 120 Editing a Management Group 3 Change the local bandwidth 4 Click OK The new rate displays on the Details tab in the management group Tab View MPC DataFRAME 440 Storage System Software User Manual 169 Working with Management Groups Logging Out of a Management Group Logging out of a management group prevents unauthorized access to that management group and the DataFRAMEs in that group 1 Select the management group to log out of 2 Right click and select Log Out of Management Group Adding a DataFRAME to an Existing Management Group DataFRAMEs can be added to management groups at any time Adding a DataFRAME to a management group increases the storage space available to the group The newly added DataFRAME can also be used as a hot spare for a cluster within the management group Note All DataFRAMEs in a cluster must be configured alike See DataFRAME Configuration Window on page 34 1 Select the DataFRAME that you want to add to a management group 2 Right click and select Add to New or Current Management Group The Add to or Create a Management Group window opens shown in Figure 121 Add to or Create a Management Group xj Select a management group for DataFRAME DF420 192 168 0 3 to join or create a new management group Existing Management Group Group Name Group 0 v O New Management Group Figure 121 Adding a DataFRAME to Existing Management Group 3 Select
171. another simply create a one time remote snapshot of the volume To make the remote snapshot a read write volume make it into a primary volume Configuration for Data Migration To make a copy of a volume in a remote location configure a cluster of DataFRAMEs in the remote location with enough space to accommodate the volume See the example illustrated in Figure 252 Configuration Diagram Application Workstations Log See Design Team 2 Application CAD Server CAD Server for Servers tor Currant Next Generation Product Product Read Write Read Only Primary pt Remote Volume i Volume Storage Primary Snapshot Production Site 1 Production Site 2 Figure 252 Data Migration Example Configuration How This Configuration Works for Data Migration 382 Suppose you want to create a complete copy of a volume for an application to use in different location 1 Configure a cluster of DataFRAMEs in the new location to contain the copied volume 2 Create either a one time remote snapshot of the volume onto the cluster in the new location MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations If your application server uses multiple volumes that must be in sync use a script to quiesce the application before creating remote snapshots Optional You can create regular one time snapshots and use remote copy to move the snapshots to the remote cluster at your convenience 3 On
172. application features will be installed This option is recommended for users that use SNMP Typical The most common application features will be installed This option is recommended for most users 2 Install Folda Minimal CH Only required application features will be installed This option is recommended only for users with limited disk space Custom Choose this option to customize the features to be installed 2 DFMC Installer Where Would You Like to Install C Program Files MPC DFMC Restore Default Folder Choose MPC DataFRAME 440 Storage System Software User Manual Getting Started 10 Leave Desktop checked if you want the install routine to automatically create a shortcut to the Console on your desktop Click on Next to continue SE DFMC Installer Select Shortcut locations I Desktop SD DFMC Installer Would you like to start the DataFRAME Management Console after the installation completes Yes C No MPC DataFRAME 440 Storage System Software User Manual Getting Started 12 Review the Pre Installation Summary and click on Install SE DFMC Installer Please Review the Following Before Continuing Product Name DataFRAME Management Console Install Folder C Program Files MPC DFMC Product Components SNMP Support Application Documentation Disk Space Information for Installation Target Required 128 112 653 bytes Available 56 139 128
173. are User Manual Working with Snapshots Edit Snapshot Edit Snapshot Snapshot Name Description Created Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Figure 162 Editing a Snapshot 4 Navigate to the field you want to change and change the information Table 42 Data Requirements for Editing a Snapshot Item Requirements for Changing Description Must be from 0 to 127 characters Hard Threshold Hard threshold size must be equal to or less than the size of the volume and available storage in the cluster You cannot decrease the hard threshold Soft Threshold Soft threshold size must be equal to or less than the hard threshold size 5 Click OK when you are finished The snapshot Tab View opens shown in Figure 160 on page 242 MPC DataFRAME 440 Storage System Software User Manual 245 Working with Snapshots Manually Copying a Volume from a Snapshot Once you have mounted the snapshot on a host you can do the following e Copy the snapshot to a read write volume e Back up the data Mounting the snapshot on a host 1 Create an authentication group for the client that you want to mount the snapshot on See Creating an Authentication Group on page 273 2 Create a volume list for the snapshot and configure the snapshot for read write access See Creating a Volume List on pa
174. arget secret is required It is used in both 1 way and 2 way CHAP when the target volume challenges the iSCSI initiator e Initiator secret The initiator secret is optional It is used in 2 way CHAP when the iSCSI initiator challenges the target volume How CHAP Works e No CHAP Authorized initiators can log in to the volume without proving their identity The target does not challenge the client e 1 way CHAP Initiators must log in with a target secret to access the volume This secret proves the identity of the initiator to the target e 2 way CHAP Initiators must log in with a target secret to access the volume as in 1 way CHAP In addition the target must prove its identity to the initiator using the initiator secret This second step prevents target spoofing 1 Challenging for the target secret nitiator Target volume ra 1 way CHAP 4 Challenging for the target secret 5 Challenging for the initiator secret initiator g Marget volume 2 way CHAP CHAP is optional However if you configure 1 way or 2 way CHAP you must remember to configure both the authentication group and the iSCSI initiator with the appropriate parameters Table 48 The table lists the requirements for configuring CHAP MPC DataFRAME 440 Storage System Software User Manual 267 Controlling Client Access to Volumes Requirements for Configuring CHAP Table 48 Configuring iSCSI CHAP
175. asking you to confirm the changes 4 Press Enter 5 On the Available Network Devices window tab to Back and press Enter Setting the TCP Speed Duplex and Frame Size You can use the Configuration Interface to set the TCP speed duplex and frame size of a network interface e TCP speed and duplex You can change the speed and duplex of a 10 100 1000 interface If you change these settings you must ensure that BOTH sides of the NIC cable are configured in the same manner For example if the DataFRAME is set for Auto Auto the switch must be set the same For more information about TCP speed and duplex settings see Editing the TCP Speed and Duplex on page 100 e Frame size The frame size specifies the size of data packets that are transferred over the network The default Ethernet standard frame size is 1500 bytes The maximum allowed frame size is 9000 bytes 306 MPC DataFRAME 440 Storage System Software User Manual Using the Configuration Interface Increasing the frame size improves data transfer speed by allowing larger packets to be transferred over the network and by decreasing the CPU processing time required to transfer data However increasing the frame size requires that routers switches and other devices on your network support that frame size For more information about setting a frame size that corresponds to the frame size used by routers switches and other devices on your network see Editing the N
176. atus column Note The disks are labeled 1 through 16 MPC DataFRAME 440 Storage System in the Disk Setup window and correspond with the disk drives from left to right starting on the left when you are looking at the DataFRAME 72 MPC DataFRAME 440 Storage System Software User Manual Storage Disk Setup Tab for the MPC DataFRAME 440 Storage System For the MPC DataFRAME 440 Storage System the drives are labeled 1 through 16 in the Disk Setup window and correspond with the disk drives from left to right when you are looking at the front of the MPC DataFRAME 440 Storage System as shown in Figure 52 The physical drive bays are numbered from 0 through 15 Table 5 Relationship of Software Disk Display Numbering to Hardware Drive Bay Numbering in the MPC DataFRAME 440 Storage System Disk Numbering in Disk Setup Tab Drive Numbering on Physical System Disk 1 Drive 0 Disk 2 Drive 1 Disk 3 Drive 2 Disk 4 Drive 3 l l Disk 16 Drive 15 G edit configuration for DF440 1 192 168 0 40 d RAID Setup Disk Setup RAID Status Norma TCP IP Network Sn DataFRAME Adminis Z SNMP Reporting Se ki Start RAID RAID Configuration RAID 0 The RAID Rebuild Rate is a percentage of RAID card throughput Percent 1 10 20 30 40 50 60 70 80 90 100 Anny v Reconfigure RAID Device Name Device Type Subdevices dewscsi host0 busOstargetO unO disc dewsc
177. b select the information you want to change 2 Click Edit The Edit Routing Information dialog opens shown in Figure 73 nda routing nro E Device Motherboard Port1 Y Net Netmask 255 255 255 255 v Cancel Gateway Figure 73 Editing Routing Information 3 Change the relevant information 4 Click OK 5 Click Apply MPC DataFRAME 440 Storage System Software User Manual 107 Managing the Network Deleting Routing Information You can only delete optional routes you have added 1 On the Routing tab select the information you want to delete 2 Click Delete A confirmation message opens 3 Click OK 4 Click Apply when you are finished Configuring a Direct Connection Between the DataFRAME and an EBSD Host If you want to configure a direct point to point connection between the DataFRAME and the EBSD host computer you must specify the route to be used for communication between the DataFRAME and the EBSD host 1 On the TCP IP tab edit the Ethernet port to be used for communication as shown below Table 22 DataFRAME Network Interface Settings Ee Value IP Address The IP address of the DataFRAME Subnet Mask The same subnet as the EBSD host Default Gateway The IP address of the DataFRAME 2 On the Routing tab add a route for communication with the EBSD host Table 23 DataFRAME Route Settings oe Value Device The network interface you configure
178. bility when a license key is obtained and applied to the DataFRAMEs in the management group containing the configured features Note If you know you are not going to purchase the feature plan to remove any volumes and snapshots created by using the feature before the end of the 30 day evaluation period Tracking the Time Remaining in the Evaluation Period Track the time left on your 30 day evaluation period by using either the management group Register tab shown in Figure 194 or the reminder notices that open periodically as shown in Figure 195 Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot D You have 29 days 21 hours and 22 minutes left in your evaluation period After the evaluation period volumes and snapshots attached to clusters containing license violations will become unavailable D Management group version is 6 3 e D Communications protocol version is 3 E DataF RAME DF140 F D Provisioning server version is 6 3 41 2605 0 Figure 194 Evaluation Period Countdown on Register Tab License iolations for Group_0 The following license agreements are being violated in Group_0 DN You have 29 days 23 hours and 3 minutes left in your evaluation period After the evaluation period volumes and snapshots attached to clusters containing license violations will become unavailable E DataF RAME DF140 B Not licensed for Multi N
179. ce Volume 2 Marketing Volume 1 Marketing Authentication wall Group Marketing Volume 2 Figure 174 Example Configuration for Assigning LUN Numbers Prohibited Host with Duplicate Numbered LUN Figure 175 illustrates a prohibited LUN numbering configuration Ge Engineering Volume 1 Engineering r No D Engineering Volume 2 Figure 175 LUN Numbering Configuration that is NOT Allowed Authentication 272 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Possible Hosts with a Shared LUN It is possible to associate multiple hosts to one LUN in which case you should make one host association read write and the other host associations read only as shown below in Figure 176 Engineeri i Authentication el F LUN A Engineering Volume Group A LUNO riw wll Liun 1 Finance Volume L 4 N LUN 0 r o Marketing Authentication wall LUN 1 Marketing Volume Group ed LUN 0 r o Shared Volume Used by All Groups Via Lun 0 Figure 176 LUN Numbering Configuration with one LUN Shared Among Three Hosts Creating an Authentication Group 1 Log into the management group and select that management group in the Network View The management group tab view opens 2 Click the Authentication Groups tab to bring it to the front 3 From the Tasks menu select New Authentication Group The New Authentication Group window opens shown in Figure 17
180. ce the ext2 filesystem is created the disk can be mounted For example 1 Make a mount point for the disk For example mkdir mnt ebsdo 2 Mount the EBSD disk For example mount dev ebsd disk0O mnt ebsdo At this point you can treat the mounted disk like any other OS file directory You can copy files add and delete files and perform other file functions there Adding an EBSD Disk at Runtime 1 Modify ebsd conf 2 Run ebsdvm add all The new disk is added to dev ebsd Use it as a raw or block device See Verifying EBSD Devices on page 323 Starting the EBSD Service If enabled for the current run level the EBSD service is started when the operating system is booted You can also start the EBSD service manually using the following command s service ebsd start 324 MPC DataFRAME 440 Storage System Software User Manual Using the EBSD Driver for Linux or etc init d ebsd start The EBSD service reads the file etc sysconfig ebsd to initialize the EBSD volume manager tool ebsdvm and the EBSD configuration file ebsd conf The file contains the following default information ebsdtool usr local sbin ebsdvm Note To prevent file system checking of ebsd disks set the environment variable fsck_check 0 in etc sysconfig ebsd You can modify this file for different directories and names The EBSD start service first checks for the EBSD configuration file ebsdconf environment variable and the EBSD volu
181. ceaeeeaeeeeeeeaeeeseeseaeeseaeesaeeeaaeeeaeessaeseseeeaaeseaeeseeeseaeesiaeseaeeseeees 115 USING CH EE 116 Editing NTP Servers scence coccetecenaneeband sdunes E ae E E sauinsecessanmsugegueenedignsdeeteasbandee 117 Chapter 6 Administrative Users and Group ssssssssssssnnnerrrnnnnnnnnnnnnnnnnnnnnnnnn nnmnnn nnmnnn 119 Getting EE 119 Managing Administrative Groups cceeceeeeeeeeeeeeeeeeeeeseeeeeeeseaeeeeeeeaeeeeaeesaeeesaeeseaeeeaeesaeesaeeseaeeseeeeneeenaeesieeeeaeesias 120 Default Administrative Groups 4 ccdsnncciciech nie Aeneid inci A R EES 120 Adding Administrative Groups ceccceeceeeeeeeseeceeeeeeeeceaeeeaeesaeeseeeeeaeeesaeseaeeeaeeseeeesaeeeaeesaeseaeeseeeseaeeneaeeenees 120 Adding Administrative Group Permissions ccecceecceeeeeeeeeeeeeeeeeeseaeeeaeeeneeesaeeseaeenaeesaeseaeeseeeeseeeeeeesiaes 122 Description of Administrative Group Permissions c eeceeceeeeeeceeeeteeeeeeeeeeaeeeaeeseeeesaeeesaeeeeeteaeeeaeeeneeeaas 123 MPC DataFRAME 440 Storage System Software User Manual vii Contents Editing Administrative Groups cecceeseeeeeseeeceneeeeeeeseeseaeeeaeesaaeeeaeeseeeeseaeeaaeceaeeseaeeeaeeseaeseaeessaeesieeeeaeesaaes 124 Deleting Administrative Groups c ccceceeeceseneeeneeeeeeeseeeeeaeeeseeseaeeeaeeseeeeseaeeaeecsaeseaeesieeeeaeeseeeeeaeesieeeeeeesaes 126 Managing Administrative USers sipiin e E A E A 126 Adding Elei CH UTC 126 Editing Administrative Users es Gg EES Mi
182. ceeecceeeeeeeeeeeeeeeeeseaeeeaeeeeaeeseeeeseeseeeeeaeesaeeeeaeeeaeeseaeseeeenaeeseeeseaeeenaeseaeeeseeess 171 Stopping aA MAMAGEN 02 cessecenteeeadee ce sen EES e a aaae aana a iaeia a a eioi eN iranis 173 Removing a DataFRAME from a Management Group cceceeeeeeeeeeeneeeeeeeeaeeseaeeeaeesaeeeaeeseeeeeaaeeeeeeeeesnaeeeaeeeaes 174 Removing a DataFRAME With a License Key ssesesesriesrissrrissrireerirerirnsiinnsitntritnnttnnntinnstinnnnnnnnnnnnnn nenn 174 Removing the Data e 175 Backing Up a Management Group Configuration cccceeeceeeeeeeeeeeeeeeeeeeeeeeeeaeeseaeeeaeeeaeeeeaeeseeeeseeeeeeseaeesieeeeeesaas 175 Backing Up a Management Group Configuration ecceeeceesceeseeeeeeseeeeeeeesseeteaeesaeeesaeeseaeseeeseaeseneeeeatenas 177 Saving the Management Group Configuration Description ceeeceseeeeeeeeeeeeeeeeeeeeeeneeeseeeeeaeestaeeeaeeenetens 178 Restoringia Management Group seisis Ed si e A A EE EN eenaa adii ae aai 178 Requirements for Restoring a Management Group eeeeeeceeeeeeeeeeeneeteaeeeeeeeeeseeeeeaeesaeseaeessaeesaeeesaeeeaes 178 Deleting a Management Group ecceeseeeeeceneeeeeeeeaeeeaeeeeeeeteaeeeaeeeaaeeeaeesaeeesieeseaeeeaeesaeesaeeseaeeseaeseeeeeaeesieeeeaeenias 180 Setting the Management Group Version ceecceeseeeeeeeeneeeneeeeeeeeeeeeaeeseeeeeaeecaeeeeaeeeeaeeeeaeeeeenaeeseeeetaeesieeseaeeeeeees 180 Selecting a Management Group from the List ceceeeseeeeeseeeeeeeeeeeeeeaeeeeeeesaeeee
183. cify the TCP IP settings of the DataFRAME manage DNS information manage the routing table and update the communication mode information if the DataFRAME is running a manager See Chapter 4 Managing the Network on page 79 DataFRAME Administration Add edit and delete administrative users and groups See Chapter 6 Administrative Users and Groups on page 119 SNMP Enable SNMP and enable SNMP traps See Chapter 7 Using SNMP on page 131 Reporting View real time statistical information about the DataFRAME run diagnostic tests and configure selected variables for active monitoring See Chapter 8 Reporting on page 139 Configuring Multiple DataFRAMEs Note When planning the configuration of your DataFRAMEs note that all of the DataFRAMEs in a cluster must be configured the same way MPC DataFRAME 440 Storage System Software User Manual 35 Getting Started You can copy the reporting and monitoring configuration of one DataFRAME to multiple DataFRAMEs Copying these configurations makes it easy to ensure that those DataFRAMEs have exactly the same configuration 1 On the Network View select the DataFRAME that has the configuration that you want to copy 2 Right click and select Copy Configuration The Copy Configuration window opens as shown in Figure 26 x Copy Configuration from DataFRAME DF140 1 192 168 0 9 Configuration Settings C SNMP Setti
184. cing Disks and RAID saisuc cscctecsbecteicceceSi asia e ecb cietes cece ndedeededed Tai r E eE E Sune EE ie See 71 Managing EE 72 Getting CEET 72 Using the Disk RepOlt EE 74 Vening DISK TEE 74 Be Elei 75 SC EI CET 75 Replacing Disken RAID O egecueseu ne Zeg deed deed EE deen ih eet 75 Adding Disks to th Data wd LCE 75 Neil 76 Adding Disks and DataFRAME Capacity ecceecceeeceseeeceeeeceeeeeeeeeeeeseaeeseaeeeaeeseeeeeaeeseeseaeeseeseaeeseaeeneesaas 76 Memory Requirements for Adding DISKS AAA 77 Adding HR LEE 77 Powering Drives On or Off EE 78 Powering Drives EE 78 P wer ing Drives ON cciscc eseceesiecess sorties deceit Egeter dv ed ee tere 78 Chapter 4 Managing the Network eccssseeeessseeeeeeessnneeenseeneesesseeneeseseeeeeeenseneeeeeeseeaeeese 79 Managing the Network Overview EE 79 Getting T TEE 79 The TEPP Tab EE 80 Identifying the Network Interfaces AAA 80 vi MPC DataFRAME 440 Storage System Software User Manual Contents Adding Interfaces to PCI Slots ceeceeeceeeceeeneeeeeeeeaeeeeeeeeaeesaeeeeaeesaaeseaeeeaeeseaeesaeesaeeseeesaeeseeseeeenaeesieeeneeenaes 81 Configuring the IP Address Manually ccccessceeeeeeeeeeeeneeeneeeeeeeeeeeeeseeseaeeeaeecaaeesaeeeeaeessaeseaeseaeeseeeenaeeeneeenieeeneeenaes 83 Using e E ler GEET 84 Configuring NIC Bonding ccceesceeecceeeeeeeeeeneeceeeeeaeeesaeeeaeeeaeeeeaeesaeeseaeessaeseaeesaeeseaeeeaeeecaeeeaeeseaeeseeeeseeesneesieeeneeenaes 84 Best Practi
185. ck OK Deleting an iSNS Server 1 Select the iSNS server to delete 2 Click Delete A confirmation message opens 3 Click OK Finishing Up 1 Click OK when you are finished changing or removing an iSNS server 2 Reconfigure the iSCSI initiator with the changes 3 Reconnect to the volumes 4 Restart the applications that use the volumes MPC DataFRAME 440 Storage System Software User Manual 205 Working with Clusters Swapping in a Hot Spare You can manually swap in a hot spare if a DataFRAME in the cluster is not available and is blinking red in the Console 1 Right click on the DataFRAME designated as the hot spare for the cluster The hot spare DataFRAME has the hot spare icon next to it 5 A confirmation message opens 2 Click OK The DataFRAME begins the process of data migration See Swap in Hot Spare on page 193 for more information Repairing a DataFRAME Repairing a DataFRAME allows you to replace a failed disk in a DataFRAME that is in a cluster configured for 2 way or 3 way replication and only trigger one resync of the data stored on DataFRAMEs in that cluster rather than restriping Resyncing the data is a shorter operation than a restripe Prerequisites for Using Repair DataFRAME e Volume must have 2 way or 3 way replication e DataFRAME must be blinking red in the Console e If the DataFRAME is running a manager stopping that manager must not break quorum How Repair DataFRAME Works
186. cklist in Table 61 to help plan scheduled remote snapshots MPC DataFRAME 440 Storage System Software User Manual 337 Remote Copy Scheduled Remote Copy Planning Checklist Table 61 Remote Copy Planning Checklist Configuration Category Parameters Snapshot Schedule Start Time Date and time for the schedule to begin e Start date mm dd yyyy Start time mm hh ss Recurrence e Recurrence v Recurrence is a yes no choice You can schedule a remote snapshot to occur one time in the future and not have it recur e Frequency minutes hours days or weeks Primary Setup Hard Threshold Set the hard threshold and soft threshold for the primary snapshot Soft Threshold Retention Retain either e Maximum number of snapshots Set period of time minutes hours days or weeks Remote Setup Management Group The management group to contain the remote snapshot Volume The remote volume for the remote snapshots Retention Retain one of the following e Maximum number of snapshots This number equals completed snapshots only In progress snapshots take additional space on the cluster while they are being copied Also the system will not delete the last fully synchronized snapshot For space calculations figure N 2 with N maximum number of snapshots Set period of time minutes hours days or weeks Working with Remote Snapshots Remote snapshots are the
187. coiii a oe Lee RA eh Ri ee O AEE aaa 143 Using Active MOmmtOrirag EE 144 Setting Notification Methods for Monitored Variables AAA 145 Removing a Variable from Active Monitoring AA 146 Adding Variables to Monitor AA 146 Downloading a Variable Log File A 148 Viewing the Variable Summary ecceeeeeeeeeeneeeeeeeeeeeeeeeeaeeteeeeeaeeeaeesaeesaeeseaeeeaaeseaeeeaeeseaeeseeteaeseaeeeneeesas 148 Listof Monitored Variables isrann nnan E E E E eee ete 149 Setting Email Notification e ce ecccceeceeeceeeceeeeeeeneeeneeeeeeeeeeseneeeeceeeeeeseeeeeaeesaeeseaeesaeeeeaeeseaeseaeesaeseaeesieeeseeseeeseneeeas 151 RUNNING Elei Ge EE 152 Viewing the Diagnostic Report 00 ee eeeeesceeeeseeeeeneeeeeeeeeeeeneeeceaeeeeeaaeeseaaeeeeeaeeeseaaeeseeeeeeieeeeeeneeeseeeeeneeeee 153 List Of Diagnostic RE 153 KEN E 155 Chapter 9 Working with Management Groupe cccseseeeeeeeeeeeeeseeeeeeesneeneeeeseeeeeeneeenens 157 Requirements for Creating Management Groups cceecceseeeeeneeeeeeeeeeeeaeeseeeeseaeeaaeeeseesaeeeaeeesaeeeaeeeaeetieeeneeensaes 157 ENN CEET 158 FUNCHIONS Of Managers sisirin oean araia ar o aD ai eaoin inariana 158 Managers and QUOTUM sisii ainiai ai NA ER ee ee i ai 158 COMMUNICATION d EE 159 Unicast Communication enges d et ES ia aint nase ee ec ee 159 Adding or Removing Managers AA 159 Creating a Management Group ceeceeeceeseceneeeeeeeceeeeeaeeeseeeaeeseeeeseesaeeaeesaaeseaeessaeseaaeeaeeeaeeseeeeeaeescaeseeeeieeesas 159
188. configured to access the remote now primary volume operation of backup application server begins The application continues to operate after the failover to the backup application servers Failback to the production configuration When the production server and volumes become available again you have two failback options MPC DataFRAME 440 Storage System Software User Manual 369 Sample Remote Copy Configurations e Resume operations using the original production server and return the backup volumes to their original remote status as illustrated in Figure 243 This will require migration back onto the production volumes of data that was written to the backup volumes since the failure e Continue operating on the backup application server When the production server and volumes become available configure the production server to be the backup server role reversal Merging data for failback In the failover scenarios described above there are probably two snapshots with different data As part of failback users must make a decision whether to merge the data from the two snapshots and the most effective method for doing so See Synchronizing the Data Between the Acting Primary Volume and the Original Primary Volume on page 361 Application Workstations Clustered Production St Backup Production Application Application Application Application Servers Serer Server Read Only Read Write Primary Remote Volume Vol
189. cpConnTable MIB Tree udp ipv 6 UdpTable MIB Tree egp MIB Tree transmission MIB Tree snmp snmpSilentDrops snmp snmpProxyDrops rmon MIB Tree application MIB Tree mta MIB Tree ipv6 MIB MIB Tree schedMIB MIB Tree scriptMIB MIB Tree agentxMIB MIB Tree ifInvertedStackMIB MIB Tree Host Resources MIB host hrDevice hrDeviceTable hrDeviceEntry hr DeviceStatus host hrDevice hrDeviceTable hrDeviceEntry hr DeviceErrors host hrDevice hrProcessorTable hr ProcessorEntry hrProcessorLoad host hrDevice hrPrinterTable MIB Tree host hrSWRun hrSwoSIndex host hrSWInstalled MIB Tree host hrMIBAdminiInfo MIB Tree UCD Extensions MIB ucdavis processes MIB Tree ucdavis prTable MIB Tree ucdavis extensible MIB Tree ucdavis memory memTotalSwapTXxT ucdavis memory memAvailSwapTXT ucdavis memory memTotalRealTXxT ucdavis memory memAvailRealTXT ucdavis disk MIB Tree ucdavis loadaves MIB Tree ucdavis extTable MIB Tree ucdavis dskTable MIB Tree ucdavis systemStats ssCpuRawWait ucdavis systemStats ssCpuRawKernel ucdavis systemStats ssCpuRawInterrupt ucdavis systemStats ssIORawSent ucdavis systemStats ssIORawReceived ucdavis systemStats ssRawInterrupts ucdavis systemStats ssRawContexts ucdavis ucdExperimental MIB Tree ucdavis fileTable MIB Tree 312 MPC DataFRAME 440 Storage System Software User Manual SNMPv3 MIB snmpModules snmpModules snmpModules snmpModules snm
190. ct an IP Netmask from the list Select Single Host if the SNMP Client is a single computer 4 Click OK The IP address and netmask appear in the Access Control list Note You can either enter a specific IP address and the IP Netmask as None to allow a specific host to access SNMP or you can specify the Network Address with its netmask value so that all hosts that match that IP and netmask combination can access SNMP By Name 1 Click By Name 2 Type a host name That host name must exist in DNS and the DataFRAME must be configured with DNS for the client to be recognized by the host name See Using a DNS Server on page 103 3 Click OK The host name appears in the Access Control list MPC DataFRAME 440 Storage System Software User Manual 133 Using SNMP Editing Access Control Entries You can change the information for the hosts granted access 1 Select a host listed in the Access Control list shown in Figure 94 GBedit configuration for DF140 192 168 0 7 xj Module d SNMP General SNMP Traps WW Storage Time Disable SNMP Agent TCP IP Network Sm DataFRAME Administ 1 Enable SNMP Agent E d Reporting Community String public Access Control IP Host Name Mask Add System Location System Contact Log Out Apply Figure 94 Editing a Host in the Access Control List 2 Click Edit The Edit SNMP Client window opens shown in Figure 95 Keier x
191. d This is normal behavior controlled by the way networking is configured The DataFRAME receives the broadcast and replies through eh regardless of which NIC received the broadcast The Console picks up the address from the packet that was sent through ethO and displays it as representative of the DataFRAME MPC DataFRAME 440 Storage System Software User Manual 23 Getting Started To Find by IP or Host Name 1 If this is the first time you have opened the Console select By Module IP or Host Name at the dialog box then click OK or Click the Find menu and click By Module IP or Host Name The IP and Host Name List window opens as shown in Figure 9 IP and Host Name List List of IPs or host names to search for IP or Host Name Status 10 0 33 25 10 0 33 35 10 0 7 11 Found Found Found add En delete Figure 9 Using IP or Host Name to Search Adding IPs or Host Names Use the following steps to add specific IP addresses or host names to the list Click Add The Add IP or Host Name window opens Type in the IP or Host Name for the module Repeat steps 1 through 3 for each module you want to find 1 2 3 Click OK 4 5 Click Find all 24 MPC DataFRAME 440 Storage System Software User Manual Getting Started Editing the IP or Host Name in the Search List Use the following steps to change the IP or Host Name of a DataFRAME in the list used to
192. d acts as the master interface As the master interface bondN controls and monitors the two physical slave interfaces Table 12 Physical and Logical Interfaces in a Bond Interface Name Description bondo Logical Interface acting as master Motherboard Por Physical interface in the motherboard to This interface acts as a slave Slot1 PortO Physical interface in a PCI slot This interface acts as a slave How Active Backup Works 86 Bonding NICs for active backup allows you to specify a preferred interface that will be used for data transfer This is the active interface The other interface acts as a backup and its status is Passive Ready The logical master bond interface monitors each physical slave interface to determine if its link to the device to which it is connected such as a router switch or repeater is up As long as the interface link remains up the interface status is preserved Table 13 Description of NIC Status in an Active Backup Configuration If the NIC Status is The NIC is Active Currently enabled and in use Passive Ready Slave to a bond and available for failover Passive Failed i to a bond and no longer has alin If the active NIC fails or if its link is broken due to a cable failure or a failure in a local device to which the NIC cable is connected then the status of the NIC becomes Passive Failed and the other NIC in the bond if it has a statu
193. d Groups Users U Storage Sorting indicator 1 Group Contains Administrators Description full_administrator admin Manage all Functions view_only_administrator View only all Functions Some k Reporting Figure 84 Sorting Administrative Groups Editing Administrative Groups Change information about administrative groups Administrative groups are listed on the DataFRAME Administration window on the Groups tab 1 Select DataFRAME Administration from the configuration categories 2 Select the group you want to edit 3 Click Edit The Edit Administrative Group window opens shown in Figure 85 124 MPC DataFRAME 440 Storage System Software User Manual Administrative Users and Groups Edit Administrative Group General Group Permission Group Name jfull_administrator 3 40 characters 0 9 a z _ Must begin with a lower case letter Description Manage all Functions 3 40 characters 0 9 a z A Z _ space Must not begin with a space Users kee Add Remove Figure 85 Editing an Administrative Group 4 Change the name and description as necessary Adding or Removing Administrative Users in an Existing Group Adding New Users to the Group 1 Click Add in the Users section The Add Users window opens with a list of administrative users 2 Select one or more users to add to the group 3 Click Add The users are added to the list
194. d Launch Pad Find DataFRAMEs Wizard The Find DataFRAMEs Wizard guides you through the process for finding DataFRAMEs on your network See also Finding Storage System Modules on the Network on page 19 Create Volume Wizard The Create Volume Wizard guides you through the process for creating a volume See also Chapter 12 Working with Volumes Access Volume Wizard The Access Volume Wizard guides you through the process for configuring client access to your volumes See also Chapter 15 Controlling Client Access to Volumes MPC DataFRAME 440 Storage System Software User Manual 15 Getting Started Console Main Window The Console main window presents a two pane view as shown in Figure 2 e Network View The top pane displays the DataFRAMEs on the network The graphic display indicates the configuration of management groups clusters and volumes e Tab View The bottom pane presents information about and functions associated with the selected item in the Network View Other features of the Console include the following Menu Bar The menu bar provides access to the following menus Find Use to find modules on the network Tasks Access all available storage configuration tasks tasks are also accessible through right click menus and from the Tasks button on the Tab View pane View Change the Network View in the Console Help Access online help and other information about th
195. d Snapshots Some snapshot scenarios call for creating a single snapshot and then deleting it when it is no longer needed Other scenarios call for creating a series of snapshots up to a specified number or for a specified time period after which the earliest snapshot is deleted when the new one is created Scheduled snapshots For example you plan to keep a series of daily snapshots up to four After creating the fifth snapshot the earliest snapshot is deleted thereby keeping the number of snapshots on the cluster at four Scheduled snapshots are an add on feature You must purchase the Configurable Snapshot Pak to use snapshot schedules beyond the 30 day evaluation period Requirements for Snapshots Review Planning Volumes on page 211 to ensure that you configure snapshots correctly When creating a snapshot you define the following parameters Table 39 Snapshot Parameters Snapshot Parameter What it Means Snapshot Name The name of the snapshot that is displayed in the Console A snapshot name must be from 1 to 127 characters and is case sensitive Description Optional A description of the snapshot Hard Threshold This becomes the hard threshold of the writable volume and defines the amount of space allocated for changes to the original volume When reached the hard threshold triggers an alert and data can no longer be written to the volume The hard threshold must be less than or equal to the volume
196. d delete volumes from this tab MPC DataFRAME 440 Storage System Software User Manual 227 Working with Volumes Snapshots Tab Displays information about the existing snapshots You can also create edit or delete snapshots from this tab See Chapter 13 Working with Snapshots for more information about snapshots Snapshot Schedules Tab Displays the name of the snapshot schedule the hard and soft thresholds set for the snapshots the actual schedule and any error that prevented a scheduled snapshot from taking place Hosts Tab Lists all EBSD hosts that have accessed the volume For EBSD hosts to retrieve the correct information for the volume enter the IP address of the host in Find By Module IP or Host Name No host information is displayed for iSCSI and Fibre Channel volumes Fibre Channel volumes may display FC Internal or the display may be blank iSCSI volumes display the DataFRAME name that the iSCSI initiator is logged into Remote Snapshots Tab Displays the names of the primary and remote snapshots the management groups they reside in and the status of the copying from primary to remote Create a remote snapshot here or cancel one that is in progress Remote Snapshot Schedules Tab Displays the name of the remote snapshot schedule the hard and soft thresholds set for the remote snapshots the actual schedule and any error that prevented a scheduled remote snapshot from taking place Target Information T
197. d in step 1 Net The IP address of the EBSD host Gateway The IP address of the DataFRAME Netmask 255 255 255 255 108 MPC DataFRAME 440 Storage System Software User Manual Managing the Network 3 On the EBSD host computer use the command line to configure the host computer s Ethernet interface as follows Table 24 EBSD Host Network Interface Settings oS ES IP Address The IP address of the EBSD host Subnet Mask The same subnet as the DataFRAME Default Gateway The IP address of the DataFRAME 4 On the EBSD host computer use the command line to add a route to communicate with the DataFRAME Table 25 EBSD Host Route Settings DEER Value Device The network interface you configured in step 3 Net The IP address of the DataFRAME Gateway The IP address of the DataFRAME Netmask 0 0 0 0 Note If the network interfaces on the DataFRAME and EBSD host are both 10 100 NICs then you must use crossover connection cables Configuring DataFRAME Communication Use the Communication tab to configure the network interface used by the DataFRAME to communicate with other DataFRAMEs on the network and to update the list of managers that the DataFRAME can communicate with 140 192 168 0 7 7 WE TCP Status DNS Routing Communication ME Adrreres Figure 74 Selecting the Storage System Software Network Interface and Updating the List of Managers MPC DataFRAME 440 Sto
198. d mask window opens Change information as necessary Click OK oor Wn 22 MPC DataFRAME 440 Storage System Software User Manual Getting Started Deleting Subnets and Masks You can delete a subnet and mask from the search list if you remove modules from that network or if you do not want to view those modules in the Network View Click the Find menu Click By Subnet and Mask The List of Subnets to Search window opens Select the subnet and mask to delete Click Delete A confirmation message opens Click OK Click Done Oona hr WN Finding by Module IP or Host Name Identify DataFRAMEs by listing module IP or host names and searching for those DataFRAMEs You can connect to one specific IP or host name or find all the DataFRAMEs in the list Network Configuration and Find by IP or Host Name The way your network is configured may affect the results of finding DataFRAMEs by IP address An example of the effect of network configuration is detailed below e You configure both NICs in a DataFRAME ethO and eth1 e The NICs are on separate subnets e You open the Console on a system on the same subnet as the ethO NIC on the DataFRAME e The Console Find function is set to Module IP or Host Name using only the IP address of the eth1 NIC The DataFRAME is discovered and appears in the Console However the IP address returned to the Console is that of the ethO NIC The eth1 IP address is not discovere
199. displays When the tests complete the results of each test display in the Result column 6 Optional When the tests complete if you want to view a report of test results click Save to File Then select a location for the diagnostic report file and click Save The diagnostic report is saved as a doc file in the designated location It is a text file and will open with Word in Windows or any text editor in Linux or UNIX 7 Optional Click Save Configuration to save the list of diagnostics that you selected so that next time you open the Diagnostics window it will be preconfigured with your selections Viewing the Diagnostic Report The results of diagnostic tests are written to a report file For each diagnostic test the report lists whether the test was run and whether the test passed failed or issued a warning Note If any of the diagnostics show a result of Failed call your Technical Support representative To view the report file 1 After the diagnostic tests complete save the report to a file 2 Browse to the location where you saved the diagnostics report doc file 3 Open the report file List of Diagnostic Tests The following table shows the diagnostic tests that are available for the DataFRAME For each test the table lists the following information e A description of the test e Pass fail criteria MPC DataFRAME 440 Storage System Software User Manual 153 Reporting
200. dit olume Edit Volume Type Volume Name Description Created Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Figure 156 Editing a Volume Primary TransData05 2 Way 8 0 8 0 8 0 On Changing the Volume Description Availability Redund ancy GB v GB v GB v 1 In the Description field change the description 2 Click OK when you are finished Working with Volumes See Requirements for Changing Volume Parameters on page 229 for detailed requirements Changing the Cluster 1 In the cluster list select the cluster to which you want to move the volume 2 Click OK when you are finished The volume will reside on both clusters until all of the data is moved to the new cluster See Requirements for Changing Volume Parameters on page 229 for detailed requirements MPC DataFRAME 440 Storage System Software User Manual 231 Working with Volumes Changing the Replication Level 1 In the Replication Level drop down select the level of replication you want 2 Click OK when you are finished See Requirements for Changing Volume Parameters on page 229 for detailed requirements Changing the Replication Priority 1 Select the replication priority you want 2 Click OK when you are finished See Requirements for Changing Vo
201. ds AAA 260 Scripted Commands for Remote Copy 261 Creating A Remote Snapshot In A Different Management Group eeceeeeeeeeeeeeeeeseeeeeeeeeeeeteaeeeeeteatenes 261 Creating A Remote Snapshot In The Same Management Group sssssssseessessrssrssrrrirsirsitsrtsrnsirssinsrnsrsene 262 Converting a Remote Volume to a Primary Volume and Back to a Remote Volume sssssssssnnssenerernerennn 262 Scripting SU VE 263 Make Remote Volume into Primary Volume 0 eeeseeeeseeeeeeneeeeeaeeeteneeeeeeneeeeeaaeeseaeeeeseeeeeeeneeeseneeeeeneeee 263 Mount New Primary Volume 0 0 0 cccesccceecesenecsececseesecaeesenscssenecsecsacsecsecsscsecaecsscsecsecssnssesensenscneseceenecnes 263 Chapter 15 Controlling Client Access to Volumes cscseeeceesseeeeeeeeeeeeeeneeeneeeesneenneeeees 265 Creating Access to VOIUMES iiion esd need cen aa a a a E ages sewesueceuandideneadencleesstesteneueceeg 265 TYPOS Of Client ACCESS insisi eios narse eheann a Jabacendcei dacs ata onus aated saved daussar EE Ne aE EIEE E NaRa 265 Client ACCESS ANG iSCSI iu ccvstaiscesdescedndesatbersededededecnsutvnedbecslebhcedececedentade Ji0Ubs dueseancdusaeiueddceaesbbidayentbeecdivacdeicecdoentan 266 Configuring Authentication Groups for IC 266 Planning iSCSI and CHAP oeenn naene eaaa naaa aa EAA ANEA EE AENEA aaar aa aaia aasa 267 Client Access opgi EBS Daira e a r Eed E E E Paneer aed N NENE 270 Client Access and Fibre Channel MPC DataFRAME 440 Storage System only sssssss
202. ds are SanDisk 256 Mb Compact Flash Kingston 256 Mb Compact Flash P725228X1 Check with your supplier for an updated list of approved flash cards A flash card from one DataFRAME cannot be used in a different DataFRAME Ifa card fails replace it with a fresh flash card MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs Removing a Boot Flash Card Before you remove one of the boot flash cards from the DataFRAME deactivate the device in the Storage System Console 1 On the Boot Devices window select the flash card that you want to remove 2 Click Deactivate The flash card status changes to Inactive It is now safe to remove the card from the DataFRAME 3 Power off the DataFRAME 4 Remove the flash card from the front of the DataF RAME Replacing and Activating a New Boot Flash Card If you replace a boot flash card in the DataFRAME you must activate the card before it can be used Activating the card erases any existing data on the card and then synchronizes it with the other card in the DataFRAME 1 Insert the new flash card in the front of the DataFRAME 2 Power on the DataFRAME 3 Log in to the DataFRAME 4 On the Boot Devices window select the new flash card 5 Click Activate The flash card begins synchronizing with the other card When synchronization is complete Active displays in the Status column MPC DataFRAME 440 Storage System Software User Manual 55 Worki
203. dule is busy reading or writing data it may not be found when a search is performed If a module is not found try the following steps to find it 1 Search again using the Find menu 2 If you have searched by Subnet and Mask try using the Find by IP or Host Name search 3 If searching again does not work try the following Check the physical connection of the module Wait a few minutes and try the search again If activity to the module was high the module might not have responded to the search Other problems can occur that prevent connection such as a bad cable connection MPC DataFRAME 440 Storage System Software User Manual Getting Started Finding by Subnet and Mask Find all the DataFRAMEs on the network by searching subnets with masks To do this Click the Find menu and click By Subnet and Mask The List of Subnets to Search window opens shown in Figure 7 List of Subnets to Search List of subnets with masks to search Subnet Mask Add Edit Delete Find all Done Figure 7 Using Subnet and Mask to Search Adding Subnets and Masks 1 a A OO N Click Add to enter a subnet and mask The Add Subnet and Mask window opens Type in the Subnet Select the appropriate mask from the list Click OK to close the Add Subnet and Mask window Click Find all The Active Search window opens tracking the search process When the search is complete
204. e volume_delete Use this value as written This is verbatim Deletes a volume volume_remote_snapshot Use this value as written This is verbatim Makes a remote snapshot of a volume volume_make_primary Use this value as written This is verbatim Makes a remote volume into a primary volume volume_make_remote Use this value as written This is verbatim Makes a primary volume into a remote volume volume_autogrow_set Use this value as written This is verbatim Sets the value by which to increase a volume threshold volume_autogrow_get Use this value as written This is verbatim Returns the value currently in the volume_autogrow_set command FC_LUN_set Use this value as written This is verbatim Sets the LUN number for a volume FC_LUN_get Use this value as written This is verbatim Returns the LUN number currently set for a volume Note You must provide either all three of the noted items or none of them For example you cannot provide only a soft threshold value MPC DataFRAME 440 Storage System Software User Manual 257 Working with Scripting ebsdvm ebsdvm is the program that mounts the snapshot or volume Parameters for ebsdvm are shown below Table 47 lists the parameters available in ebsdvm Getting Help 1 Type ebsdvm help and press Enter Table 47 Parameters for ebsdvm Parameter What It Is mgmt group name Value
205. e administrative users and groups All administrative users and groups must be added and managed locally Note The user who is created during DataFRAME configuration using the Configuration Interface becomes a member of the Full Administrator group by default Getting There 1 On the Network View double click the DataFRAME and log in if necessary The Module Configuration window opens 2 Select DataFRAME Administration from the configuration categories The Groups tab opens as shown in Figure 80 G edit Configuration for DF140 192 168 0 7 Module T Storage Time D TCP IP Network Group Contains Administrators Description t DataFRAME Administ full_administrator admin Manage all Functions ge SNMP view_only_administrator View only all Functions Reporting Groups Users if Log Out Figure 80 DataFRAME Administration Groups Tab Administrative Users and Groups Managing Administrative Groups The DataFRAME comes configured with a set of default administrative groups You can use these groups or create new ones Default Administrative Groups If you assign an administrative user to one of the following groups that user will have the privileges associated with the group Table 26 Using Default Administrative Groups Name of Group Management Capabilities Assigned to Group Full_Administrator Manage all functions read write access to all functions View_On
206. e Console and Storage System Software m rc Computers DataFRAME Management Console File Find Tasks View Help SS Available DataFRAMEs DF140 1 Network View DF 420 1 if Available DataFRAMEs Management Groups Alert Messages EBSD Host These DataFRAMEs are available for use Name Software Version Model DF 420 1 192 168 0 4 6 3 33 0007 0 R316MJ2 DF140 1 192 168 0 9 6 3 41 2605 0 aD VIEW ndmeo DF140 192 168 0 7 6 3 41 2605 0 Nd M160 192 168 0 3 6 3 33 0007 0 R316MJ2 Figure 2 Features of the Console Main Window MPC DataFRAME 440 Storage System Software User Manual Getting Started Icons Used in the Storage System Console A description is available of all the icons used in the Console To see them 1 Click Help on the menu bar 2 Select Graphical Legend from the menu The icon display window opens The Graphical Legend has three tabs e The Items tab shown in Figure 3 displays the icons used to represent virtual items displayed in the Console For example management groups and clusters are virtual items Graphical Legend xj items States Hardware Big Icons Small icons Primary Volume i Remote Volume UD Snapshot S Si Eo Item is selected Item is reconfiguring or unavailable Item has mouse over it o E Figure 3 Graphical Legend Items from the Help Menu MPC DataFRAME 440 Storage Sys
207. e Creating a remote snapshot e Making a primary volume into a remote volume e Failing over to a remote snapshot Creating A Remote Snapshot In A Different Management Group 1 Create the primary snapshot java commandline CommandLine lt primary admin name gt lt primary admin password gt lt primary manager ip gt volume_snapshot lt primary volume name gt lt primary snapshot name gt lt soft threshold Megabytes gt lt hard threshold Megabytes gt lt description gt lt failure timeout seconds gt 2 Create the remote snapshot java commandline CommandLine lt remote admin name gt lt remote admin password gt lt remote manager ip gt volume_remote_snapshot lt remote volume name gt lt remote snapshot name gt lt remote snapshot description gt lt primary admin name gt lt primary admin password gt lt primary manager ip gt lt primary snapshot name gt lt failure timeout seconds gt Example Joe Jones plans to create a remote snapshot of his X Rays volume in the backup management group in the corporate backup site He is naming this new remote snapshot RSS2_xrays and the new primary snapshot PSS2_ xrays He created MPC DataFRAME 440 Storage System Software User Manual 261 Working with Scripting his remote volume RemVolX_Rays using the Console and named his first primary snapshot PSS1_ xrays and his first remote snapshot RSS1_xrays The size of the thresholds for the new primary and remote snapshots are the
208. e DataFRAME 2 On the Module Information tab click Change Password The Change Password window opens Type in the User Name and Old Password Type in the New Password Retype the New Password for confirmation Click OK Oona fk Ww Change any other user s password in the DataFRAME Administration configuration category MPC DataFRAME 440 Storage System Software User Manual 41 Working with DataFRAMEs Upgrading the Storage System Software When you upgrade the Storage System Software the version number will change You can view the current software version on the Module Information tab in the DataFRAME configuration window or on the DataFRAME Details tab in the Network View Prerequisites e MPC DataFRAME 440 Storage System Both flash cards must be present before upgrading the software e Stop any applications that are accessing volumes that reside on the DataFRAME you are upgrading Copying the Upgrade Files from CD or FTP Site Upgrade the Storage System Software on the DataFRAME when a upgrade or a patch is released The Storage System Software upgrade installation takes about 5 to 8 minutes including the DataFRAME reboot Note The 88M must contain both boot flash cards in order to upgrade the storage system software See Configuring Boot Devices on page 52 1 Download the upgrade file from the web site of your approved supplier or from a CD Upgrading the DataFRAME You can install upgrades
209. e DataFRAME Configuration File 46 Figure 35 Restoring the DataFRAME Configuration File 47 Figure 36 Restoring the DataFRAME Configuration File 48 Figure 37 Shutting Down or Rebooting the DataFRAME A 49 Figure 38 Canceling the DataFRAME PDeboct AA Figure 39 Figure 40 Figure 41 Figure 42 Viewing the Feature Registration Tab Using Remote Copy without a License Viewing Boot Device Status with Two Devices A 53 Viewing the Storage Configuration Category Figure 43 Capacity of Disk Pairs in RAID 10 Figure 44 Parity Distributed Across a RAID 5 Set Using Four Diske A 61 Figure 45 Viewing the RAID Setup Heport natt ntEnntuntEntEnntEntEntatE nennen enaena enna 61 Figure 46 RAID 0 on an MPC DataFRAME 440 Storage System cccecceeeeeeeeseeeeeeeeeneeeeeeeeeeseaeeeeesneeeseeeeeenaaes 62 Figure 47 RAID 10 on an MPC DataFRAME 440 Storage Gvstem nnt 63 Figure 48 MPC DataFRAME 440 Storage System RAID 50 Using 4 Disk Sets A 63 Figure 49 MPC DataFRAME 440 Storage System RAID 50 Using 3 Disks Plus a Hot Gpare 63 Figure 50 Monitoring RAID Status on the Main Console Window ssessessesssessresireirseresrrernstretrnntnntnnstnntnntnnennennnnt 71 MPC DataFRAME 440 Storage System Software User Manual XV List of Figures Figure 51 Figure 52 Figure 53 Figure 54 Figure 55 81 Viewing the Disk Setup Tab in the MPC DataFRAME 440 Storage Gvstem 73 Diagram of the Drive Bays in the MPC DataFRAME 440 Storage Syste
210. e Molumg D Description Cluster Cluster_0 h Replication Level None bd Replication Priority Size 2 6063919067381 TB Y Max Hard Threshold 2 6063919067381 TB v Max Soft Threshold 2 6063919067381 B v Max Auto Grow off O Auto O Manual Checksum Data Yes No Figure 153 Creating a New Primary Volume 5 Select primary as the volume type The window for a new primary volume is shown in Figure 153 For information about creating a remote volume see the Remote IP Copy User Manual lf you are creating a remote volume see Creating a Remote Volume on page 342 6 Type a name for the volume 7 Optional Type a description of the volume 8 Select a replication level You must purchase the Scalability Pak to use the N way replication feature beyond the 30 day evaluation period 9 Select a replication priority If you select a replication level of None the replication priority must be Availability See Figure 154 Replication Level None Replication Priority Figure 154 Setting Replication to None 10 Type a size and select the units 11 Type a hard threshold and select the units 12 Type a soft threshold and select the units 226 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Note The system automatically factors replication levels into the settings For example if you create a 500 GB volume and the replication lev
211. e Network View and leaves you logged in to the DataFRAME The Network View window displays the DataFRAME with a pink square underneath indicating the logged in status Module Configuration Overview The module configuration category provides access to detailed information about the DataFRAME backing up and restoring DataFRAME configuration files the software reboot or power off function boot devices and feature registration The module configuration category window for the DataFRAME is shown in Figure 30 40 MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs Changing the DataFRAME Host Name The host name on a DataFRAME is the name that displays below the DataFRAME icon in the Network View It is also visible when the users browse the network Change the host name of the DataFRAME on the Module Information tab The DataFRAME arrives configured with a default host name 1 Log in to the DataFRAME 2 On the Module Information tab click the Host Name field and type the new name If you are operating in a Windows environment the host name should be 15 characters or fewer 3 Click Apply A confirmation message opens 4 Click OK Note Add the host name and IP pair to whatever host name resolution methodology is employed in your environment e g DNS or WINS Changing Passwords Change the password for the user who is logged in to a DataFRAME on the Module Information tab 1 Log in to th
212. e TCP interfaces Change the speed and duplex method of an interface The TCP Status Tab Review the status of the network interfaces on the TCP Status tab shown in Figure 68 ER Configuration for DF140 192 168 0 7 T Storage Module i TCPAP TCP Status DNS Routing Communication E Time Le DataF RAME Adminis Name GE SE SE Status Frame Size Preferred SNMP SG Fees k Reporting Motherb Intel Corp Unknown Unknown Passive F Default Yes Mothelbondo Corp 100Mbs A Half Auto Active Default Set Preferred Figure 68 Viewing the TCP Status Table 18 Status of Information About Network Interfaces Column Description Name Name of the interface Entries are Motherboard PortO Motherboard Port1 e Gott Port e bondO0 the bonded interface s displays only if DataFRAME configured for bonding Description Describes each interface listed For example the bond is the Logical Failover Device Speed Method Lists the actual operating speed reported by the device Duplex Method Lists duplex as reported by the device Status Describes the state of the interface See Table 13 for a detailed description of individual NIC status MPC DataFRAME 440 Storage System Software User Manual 99 Managing the Network Table 18 Status of Information About Network Interfaces Column Description Frame Size
213. e The DataFRAME serial number used to obtain a license key e The license key for that DataFRAME if one has been purchased e Which if any add on modules or applications have been licensed GM Edit Configuration for DF140 192 168 0 7 xi ka Ge d Power Off Feature Registration Boot Devices zi A noge Module Information Backup amp Restore j a TCP IP Network amp DataFRAME Adminis gt SNMP Registration is required for additional software features Current feature Reporting registration license keys are displayed below For additional feature registration please contact your storage provider www mpccorp com 1 888 224 4247 and be prepared to provide the serial number s indicated below SerialNumber 00 30 48 2E BA 5A License Key Not licensed for Multi Node Virtualizetion and Clustering Not licensed for Remote Data Protection Pack 4 Not licensed for Managed Snapshot ay Log Out Figure 39 Viewing the Feature Registration Tab You can register the DataFRAME and purchase a license key to apply in this window First you submit the serial number as instructed Then when you receive a license key copy and paste it into the License Key field MPC DataFRAME 440 Storage System Software User Manual 51 Working with DataFRAMEs Evaluating Features Add on modules and applications are available when you begin using the Storage System Software If you begin using an add on feature or ap
214. e Thresholds 2 Power Supply Status CPU Utilization Memory Utilization RAID Status Network Interface Status Drive Status If the value exceeds 65 alert If the value exceeds 70 alert shut down Save Log Files etails about the selecte D an aad eat Rem variable are displayed here Apply Threshold Actions Log Out Figure 105 Viewing the Monitoring Variable Summary on the Active Window 148 MPC DataFRAME 440 Storage System Software User Manual List of Monitored Variables Reporting Table 31 shows the variables that are monitored for the DataFRAME For each variable the table lists the following information e The units of measurement e Whether the variable is permanent Permanent variables cannot be removed from active reporting e Whether you can change the frequency with which the measurements are taken e The default frequency of measurements e The default action that occurs if the measured value of the variable reaches a threshold Table 31 List of Variables Available for Active Monitoring Perm Specify Default Default Action ic ali Ss Variable Freq Freq Threshold Boot Devices Active No Yes 30 Console alert if not Status i seconds normal Inactive Failed Empty Unformatted Not recognized Unsupported Cache Battery Normal No Yes 1 minute Console alert if Status Charging changes M
215. e following Verify all entries in the EBSD config file Verify the network connection to the DataFRAMEs 3 Re check cat proc ebsd client after making corrections to any issues found During Unmounting If during unmounting you get device filesystem busy you probably have a process accessing dev ebsd disk0 Restart EBSD Service Gives Error Message When the EBSD driver is loaded manually or using the ebsd service the following message appears Warning loading lib modules 2 4 9e enterprise kernel driver addon ebsd ebsd o will taint the kernel non GPL license The EBSD Driver for Linux is a custom driver and will always produce this message when installed 328 MPC DataFRAME 440 Storage System Software User Manual D Remote Copy Remote Copy provides a powerful and flexible method for replicating data and keeping that replicated data available for business continuance backup and recovery data migration and data mining Remote Copy uses the existing volume and snapshot features along with replication across geographic distances to create remote snapshots The geographic distance can be local in the same data center or on the same campus metro in the same city or long distance For example the accounting department in the corporate headquarters in Chicago runs the corporate accounting application and stores the resulting data The designated backup site is in Des Moines Nightly at 11 00 p m accounting
216. e volumes Once removed that data cannot be retrieved Note Be sure to disconnect the device before deleting it 1 Modify the etc ebsd conf file Remove the appropriate device entry in ebsd conf Uninstalling the EBSD Driver 1 Oo ON DO Om A ODN Navigate to the following directory opt Storage_System Storage_System_Software 6 x Drivers EBSD Copy the file install sh to your client system Run install sh u Type cd and press Enter Run Uninstall_EBSD_Driver The Install wizard opens Click Uninstall Click Done when the wizard is finished uninstalling the driver Type cd and press Enter Finishing Up Remove the EBSD volumes from the Storage System Console MPC DataFRAME 440 Storage System Software User Manual 327 Using the EBSD Driver for Linux Troubleshooting Error Could not load the ebsd driver on your system 1 Run uname r to determine what version of Linux you are running 2 Determine whether that version is supported by the EBSD driver Driver successfully loaded but adding device returns failed i e ebsdvm add 0 returns failed 1 Check the error message and correct the problem Driver successfully loaded adding a device appears successful but when you check the config file the device was not added 1 Check cat proc ebsd client 2 If device does not exist add proper entry into the EBSD config file or If device status stuck in starting mode check th
217. edie ation 250 e vledlute Bil ET 251 Rolling Back a Volume to a SnapShot 00 2 0 cceeceeeseeseeesececneeeeseeeeeseaeeeaeenaneceeeseaeeesaeseasesaeesaaessaeeeaeeeaeesneeeeneetaas 251 Requirements for Rolling Back a Volume ee eeeceeeseeeeeneeeeenaeeeeeeeeseaeeeeeaeeeeeaeeeseaeeeesaeeeseaneeeseaeeenenneeeeaes 251 Deleting a Snapshot eege erties eve desc weniad natn dada iti hbase eieei erin EE Ne dE 253 Selecting a Snapshot from the List c ce eececeeeeeeeeeeeeeeeeeeeeeeeeaeeseeeeeaaeseaeeeaeessaeseeeeseaeeeeaeesaeeeaeeseeeseeeenaeeeaeesneeees 254 Selecting a Snapshot Schedule from the List ecceesceeeeeceeeeeeeeeeeeeeeeeaeecaeeeeaeeeaeeseaeeeaeesnaeeseeeeeaeeseeseaeeeneeee 254 Chapter 14 Working with Scripting ccccessseeeeeeeeeeeeeneeeeeeeeeeeeeeneeeeeeeesseeneeeeeeeaeeeeeeenes 255 MOOS AOR e EEN 255 Java commandline COMMANALING 00 0 eee een ee eneeeeeeneeeeeaeeeteeeeeseaeeesesaeeeeeeeeseeeeeseaeeeeseeeeeeeeeenneeneaaes 255 ee EE 258 Scripted Commands for Volumes and Snapshots ceeceeeceeeeeeeeeeeeneeeeeeeeeeeseeeeaeeeaeeseaeeeeeeeaeeseeeseeesnaeeeaeeeeeens 258 VC de ECKE e EE 258 Deleting a SMAPSMOl EE 259 Assigning a LUN Number to a Fibre Channel Volume or Snapshot MPC DataFRAME 440 Storage System ONY BEE 259 Mounting a Snapshot eeeceecceeeseeeeeeeeeeeeeeeeeaeeeaceeeeeessaeeeaeeeseeeeaeesaeeeeaeesaeesaeeseeseaeeseeeeeaesaeessaeesieeeeaeensas 260 Increasing Volume Hard and Soft Threshol
218. ednesday will be deleted You can use Remote Copy to roll a volume back to an old snapshot without losing the interim snapshots Because Remote Copy creates two sets of snapshots primary snapshots and remote copies you can roll a volume back to a snapshot and still retain the other set of snapshots Configuration for Non Destructive Rollback To use remote snapshots for non destructive rollback e Create a remote snapshot schedule e In the schedule specify the same retention policy for the primary and remote snapshots This ensures that you have copies of the same number of snapshots in your primary and remote locations Any snapshots destroyed during rollback of one volume will remain intact on the other volume See Figure 249 for an illustration of this configuration 378 MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations Configuration Diagram Application Workstations Clustered Application Restore Application Serer Server Servers Read wWrite Read Only Primary Volume Storage 5 00 PM Subnet Snapshot 3 00 PME Snapshot 7 00 PMS Snapshot 14 00 AM Figure 249 Non destructive Rollback Example How This Configuration Works for Non Destructive Rollback You can choose to roll back either the primary snapshot or the remote snapshot Rolling back one of the snapshots will cause all the more recent snapshots of that volume to be deleted The other volume retains the
219. eeceaeesaeeseaeeeaeeesaeeeaeeseeeesaeeseaeeeaeesieeetaeeseaeeeneeeeeeenaes 58 Benefits o RBAID WEE 58 RAID Configurations Defined AAA 58 N mber of Disks and RAID sicccnr eka iid anata aia enone acd 58 RAID Set SIZE berg 59 EUR A BEE 59 RAID ai ric RAID NO EE 60 RAID 5 and RAID 50 EE 61 Viewing the RAID Setup Report AAA 61 Planning RAID Configuration cccecceeeceeeeeeeeeeeeeeeneeeeaeeeaeeseeeeseeeeaaeseaeeeneeeeeeeseaeecaeeesaeeeaeeeneeseaeeseaeeeaeesieeeeaeeenaeee 64 Data Replication TEE 64 Using RAID with Replication in a Cluster ec eeeeeeceeeeeeeeeeneeeeeeeeeeesaeeseaeeeeesaeeeaeeseeeeseeseaeseieeenaeeneeseaeen 65 Planning RAID for Capacity Growth uerge EENEKeES pide cin E SSON RTE 66 Requirements for Configuring RAID ecceeceeseesceeeeneeeneeeeaeeeeeeeeaeeseaeeaeesaeeceaeeeaeeeeaeseaaeeeaeeseeeseaeeeeesiaeeeaeeseaeees 66 Placement of Disks in the DataFRAME AAA 66 Management Groups and RAID 0 ecceeecececeeeeeeeneeeeeeeeeeeeeeeeeeseaeeeaeeseeeesaeeseaeeeaeeseaeesaeeseeeeeeeenaeseeeeeeeeaten 67 Clusters and RAID sc2s isientieie ek ase ANG dei eed ce lined ails 67 Configuring RAID ET 67 Setting RAID Rebuild Rate for RAID 1 10 or RAID 5 50 00 eeceeseeeeeeeeeeeeeeeeeeteaeeeaeeseeeesaeeeneeenaeeeaeeeeaeen 68 dude ERAN DD EA A cece te cet tus eu secu EE 68 ERR ell EE 69 Monitoring RAID Status eee eecceseeeseeceeeeeneeceaeeeseenaeeeseeesaeseaeesaaeesacesaeseaeeesaeseaeesaeeseaeeenensaeesneeseeaeenaeee 70 Repla
220. eeeaeeeaeeeaeeseaeeteaeseaeeeseeeeaeeeeeeseaeeneeeeeaees 204 Changing or Removing the Virtual IP oo cece ccceeeeeeeneeeeeeeeeeseaeeeeeeeeeeeaeesaeeeaeesaeseaeesaeeeeaeeeeeeseaeeseaeesieeees 204 MPC DataFRAME 440 Storage System Software User Manual ix Contents Changing or Removing an ISNS Server o ceceeeceseeeeeneeeeeeeeeeeseeeeeaeeesaeeeaeeesaeeeaeeseeeeaeeseeeeeeseaeeeeeeeeeesas 205 Swapping in a Hot Spare oe eeececeeeeeeeeeeeeeeeeeeeeeecaeeceaeeeaeesaeseaeeseeeeseeeseaeeeaeessaeeeeeesaeseaeeseeeeeeesieeseaeessaeseeeeneeeas 206 Repairing a Data FRAME eves setantaus ents iit ee 206 Prerequisites for Using Repair Dat FRAME 206 How Repair DataFRAME Works erneieren a E E AEA E E NE 206 Deleting a er EE 209 Se lecting a Cluster from the EE 209 Chapter 12 Working with Volumes cccccseeeeeeeeeeeeeeeeeeeeeeeneeeneeeesseeeesseeeaeeeeeseneeseneseenes 211 SEllule Weu 211 Planning Volume Type siii isoa aaa a a a E a AEE A AEA E EE aaa aE E Aa E ONE Aaaa 212 Planning Volume EE 212 Measuring Disk Capacity and Volume Size ou eee eecceseeeeeeeeeneeeeeeeeseeeeeeeeeeeeaeeeaeesaeeeaeeseaeseeeeneeeseeseeeenaaes 212 Planning Hard Wal Ve EE 213 Planning Snapshots AA 214 Planning eu le EE 214 Planning Data Replication ccscccscessseeszeee naetseseteneseyseesaeesesevannistescreadetoaneneege teeter ett rteeenens atest 214 Requirements for VOlUMe S i fici nee heer Wheel i E E ee EUREN eee 218 Managing Volume Growth Capacity
221. eeeeeeeeeeeeeeeeeeeeaeeeaeeeeaeetaeteaeeeeeesas 302 Logging in to the ENCHe Kl TEE 303 Configuring Administrative Users cceececeeeeeeeeeeeeeeeeeeeeeeeseaeeeeeceaeeseaeesaeeeaeeseeeeeaeeseaeeeaeesieeesieeeeaeesiaeeeaeeeneeeeas 303 Configuring a Network Connection ceececceeseeseeeeeeeeseeeeeaeeeeaeeeeeseaeeeeaeesaeeeaeeseeeeeseeseaeeeeeeeeaeeseeseaeesiaeseeeeieeeeas 304 Deleting a NIC Bond EE 305 Setting the TCP Speed Duplex and Frame Size ooo eeceeeceeeeeeeeeeeeeeneeeeeeeeaeeeeeeeeseeseaeeeaeeseeesaeeseaeeeeeseaeeeneeeeas 306 Removing a DataFRAME from a Management Group ceceeeceeseeeeeeeeeeeeeeeeeeeeeaeeeseesaeeeaeeseaeeseeeseaeesieeeeeeeraaes 308 Resetting the DataFRAME to Factory Defaults AA 308 Appendix B SNMP MIB Information ccccseeeeeeeeeeeeeeeeeeeeeeeeeseeeeeeeeeeeseeeeenenees 311 SNMP AGenit EE 311 Supported MIBS TEE 311 EXC ptonS EE 311 Appendix C Using the EBSD Driver for LINUX cccssseeeeeseeeeeeeeeeeeeeeeeeeees 315 Installing the EBSD Driver for LINUX eceeeceeeeeeeeeeeeneeeeeeeneeeeaeeeaeseaeeseaeeeeenaeeseaeeeaeessaeeeaeesieeesaeeseessaeetieeeneeensaes 315 Options Available When Using the Installation Wizard AA 315 Copying Driver Bundle to a Network Share Optional cccccesceseceeeeeeeeeeeeeeteeeseeeeeaeeceeeeaeeseaeeeeeeeeeesas 315 Installing the EBSD Driver Using the CD or the Driver Bundles AA 316 Location of the Installed Driver Files sssrinin n aaeeeiai e a ad aa
222. eesseeeesreerrerieerrererreenens 271 Planning Volumes and Fibre Channel AAA 271 Assigning LUN Numbers to VOIUMES A 271 Creating an Authentication Gro p sresti E TEE eee eed eine 273 en le Hilfe SCSI NEE 274 Finishing iSCSI Configuration eeceecceseceeeeeecneeeeeeeeneeeeaeeeeeeesaeeseeeeeaeesaeeeaeeceeesaeeeseeseaeeeaeesneeeneeseaesenees 276 COmMPiQuring EBSD RE 276 Finishing EBSD Configuration 277 Configuring Fibre Channel MPC DataFRAME 440 Storage System only eeeceesceeeeeeeeeeeeeeeeeeetenees 278 Finishing Up the New Authentication Group cceecceeeeeeeeeeeeeeeeeeseeeeeaeeseaeesaeeeaeeeeaeeseeeseseeseeeeeaeesieeeeaeenaas 279 Editing an Authentication Group oo eeceeceeeceseeeeeeeeeeeeeeeeeeeeeeeaeeeeaeenaeeseeeeseeesaesesaeeeaeesaeeseaeeeaeeseaeseaeeseeeeeeeseaeeeaees 279 Deleting an Authentication Group cceeccceeceeeeeeeeeeeeneeeeeeeeeeseaeeeaeeeaaeseaeeseeecaeeseaeeeaeesaeesaeeseeeeeeeseeeeeeesieeeeeesias 280 Volume Lists Overview suc sicceitaan dias ANNE tails eee eect dace el iia geen 281 Requirements for Volume Lists A 281 Planning We tu 281 Creating a Volume List 2 scccd c2c3ec02ccerevepencdezcepenccneee cesar eaa a arada d dane eect 281 Adding Volumes to the Volume List ccesececeeeeeeeseceeeeneeeeeeeeeessaeeneseneecaeeseeseaeeseeaeeseeeeeseeeeeneenseneneees 282 Adding Authentication Groups to the Volume List ecceeeeeeeeeeeeeeeeneeeeeeeeeeeseeeeeeaeseaeeeaeeseaeeseeeeeaeeeeeeeae
223. egation provides bandwidth gains because data is transferred over both NICs simultaneously For NIC aggregation both NICs must be plugged into the same switch and that switch must be LACP capable and support 802 3ad aggregation Because both NICs are plugged into the same switch NIC aggregation does not protect against switch failure For active backup plug the two NICs on the DataFRAME into separate switches While NIC aggregation will only survive a port failure active backup will survive a switch failure Table 11 Comparison of Active Backup and NIC Aggregation Bonding Feature Active Backup NIC Aggregation Bandwidth Use of 1 NIC at a time Simultaneous use of both provides normal bandwidth NICs increases bandwidth Protection during port Yes Yes failure Protection during switch Yes NICs are plugged into No Both NICs are plugged failure separate switches into the same switch Requires support for No Yes 802 3ad link aggregation Allocate a static IP address for the logical bond interface You cannot use DHCP for the bond IP MPC DataFRAME 440 Storage System Software User Manual 85 Managing the Network Physical and Logical Interfaces The NICs in the DataFRAME are labeled Motherboard PortN and SlotN PortN where N is a number depending on whether the NIC is located in the motherboard or in a PCI slot If 2 or 4 physical interfaces are bonded the logical interface is labeled bondN an
224. el is 2 the system automatically allocates 1000 GB for the volume 13 Select the cluster you want to contain the volume 14 Optional Select the auto grow setting you want 15 Optional Select whether you want to enable checksumming 16 Click OK The Storage System Software creates the volume and it is attached to the cluster shown in Figure 155 17 Select the new volume in the Network View The volume Tab View opens also shown in Figure 155 mrc Computers DataFRAME Management Console loj xj File Find Tasks View Help Available DataFRAMEs T Rea a a Cluster_0 DF420 1 g b kb DF140 DF140 1 Volume H Volume Remote Snapshot Schedules Target Information Volume List Memberships Details Snapshots Snapshot Schedules Hosts Name Volume 0 Remote Snapshot Size 500 GB Replication Level None Created 11 22 200 Cluster Cluster 0 Checksum Data No Hard Threshold 500 GB Replication Priority Availability Soft Threshold 500 GB Type Primary Created by Script No Status Normal Volume Tab View Description Figure 155 Viewing a Volume in a Cluster The Volume Tab View The tabs provide access to volume information and features such as creating snapshots and associating authentication groups with the volume Details Tab Displays information about the selected volume You can add create snapshots add volumes to volume lists and edit an
225. elect TCP IP Network from the DataFRAME configuration categories 3 Click the Routing tab to bring it to the front shown in Figure 71 eit Configuration for DF140 192 168 0 7 xj TCPAP TCP Status DNS Routing Communication amp Module Storage Time Routing Info in order Ss DataF RAME Adminis fe SNMP Dees Net Gateway Netmask Flags Reporting Motherboard 192 168 0 0 0 0 0 0 755 255 255 0 U Motherboard 0 0 0 0 192 168 0 2 0 0 0 0 UG Figure 71 Adding Network Routing Information 106 MPC DataFRAME 440 Storage System Software User Manual Managing the Network 4 Click Add The Add Routing Information dialog opens shown in Figure 72 Add Routing Info x Device Motherboard Port1 Y Net Gateway Netmask 255 255 255 255 Cancel Figure 72 Adding Routing Information 5 Select the port to use for routing in the Device list 6 Type the IP address portion of the network address in the Net field 7 Type the IP address of the router in the Gateway field 8 Select the netmask 9 Click OK 10 Use the arrows on the routing table panel to order devices according to the needs of your network The DataFRAME will attempt to use the routes in the order in which they are listed 11 Click Apply when you are finished Editing Routing Information You can only edit optional routes you have added 1 On the routing ta
226. ement you choose Manual The thresholds will only increase e When there is sufficient room in the cluster to accommodate the increases or 220 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes e To the point where the hard threshold equals the volume length at which point the soft threshold also is increased to equal the volume length so that alerts are not triggered whichever of these conditions occur first Best Practice The goal when using auto grow is to grow smoothly and just in time growing as little as possible when you can get away with it and limiting the total number of auto grow events to a reasonable number Also you want the volume to grow before the client s hit the hard threshold which would cause the client s to stall for a few seconds Setting Auto Grow for Manual Use manual auto grow to select the amount by which you want the hard and soft thresholds to increase Both thresholds will increase by the same amount at the same time The increment must be at least 1 of the volume length or 1 MB whichever is greater Setting Auto Grow for Automatic The auto setting for auto grow begins with small increments which gradually increase in size After a maximum of ten auto grow events the volume reaches its full size which is the volume length The increment schedule for automatic auto grow is illustrated belowin Figure 149 Maximum instances of auto grow mun SE i 1 2 3 4
227. ents for Snapshots A 236 Managing Capacity Using Volume and Snapshot Thresholds AAA 237 Easiest Method for Planning Capacity ccccceeceeseeeeeeeeeeeeeneeeeeeeeaeeseeeeseeeesaeeeaeeeeaeeseeeeeaeeeeeeeeaeesnaeeeeeessaes 237 Most Flexible Method for Planning Capacity ceceecceeeceseeeeeeeeeeeeseeeeeeeeseneseaeeseeeeaeeseaeeseesseeeneeeneeeeaes 238 N elleRe ETC 239 Source Volumes for Data Mining or Tape Backups or Data Preservation Before Upgrading Software 239 Protection Against Data Corruption cecccccceeeeeeeeeeeeeeeeeeeseaeeeaeeeeeesaeeseaeeeaeessaeseaeeseeesaeeeeeeseaeesieeeeeeesaaes 239 Creating a SMAPSMOL EE 240 The Snapshot Tab View eg dani Eeer deet 242 Mounting or Accessing a Snapshot cee eccceeceeeeeeeneeeeeeeeneeeeaeeeeeseaeeseeeesaeeeaeeseeeessaeseaeeseeeenaeeseeseaeesiaeseaeeeneeeeas 243 Snapshot EE EE 243 Deleting a Snapshot s Writable Space A 244 Editing a Snapshot EE 244 Manually Copying a Volume from a Snapshot eeceeceeceeceeeeeeeeceeeeeaeeeaeeesaeeeaeessaeesaeseaeseaeeseeeeseeeeeeseaeeeneeesas 246 Creating Snapshot Schedules A 246 Requirements for Scheduling Snapshots AA 246 x MPC DataFRAME 440 Storage System Software User Manual Contents Creating Snapshot Schedules eee eeeeceeceeeneeeeeeeeeeeeeeeeeeeseeseaeeseaeeaeetaeeeeaeeeseeseaeseaeesnaeeseeeeeaeeseaeeeaeesneeees 247 Editing Snapshot Schedules AA 249 Deleting Snapshot Schedules siirsi iniii hanes dicen tees aeedi
228. er The Log Type list only contains logs that support syslog 5 In the Destination field type the IP address or host name of the computer that will receive the logs 6 Click OK The remote log displays in the Remote logs list on the Log Files tab Configuring the Remote Log Target Computer Configure syslog on the remote log target computer Refer to the syslog product documentation for information about configuring syslog Note The string in parentheses next to the remote log name on the Log Files tab includes the facility and level information that you will configure in syslog For example in the log file name MPC DataFRAME 440 Storage System Software User Manual 143 Reporting auth error auth warning the facility is auth and the level is warning Editing Remote Log Targets You can select a different log file or change the target computer for a remote log 1 On the Log Files tab select the log in the Remote logs list 2 Click Edit The Edit Remote Log Target window opens 3 Change the log type or destination 4 Click OK Deleting Remote Logs To delete a remote log 1 On the Log Files tab select the log in the Remote logs list 2 Click Delete A confirmation message opens 3 Click OK Note After deleting a remote log file from the DataFRAME remove references to this log file from the syslog configuration on the target computer Using Active Monitoring Use active monitoring to track
229. er Manual Remote Copy Creating a New Remote Volume When you create the remote snapshot use the Remote Snapshot window shown in Figure 217 to create the volume Alternately you can create a new volume from the cluster details panel and select the Remote option in the New Volume Window Note The fastest way to create a remote volume is to create it as part of creating the remote snapshot using the Remote Snapshot window To create the remote volume from the New Remote Snapshot window 1 In the Remote section select the Management Group to contain the remote snapshot You must be logged into the management group to continue 2 To create a new remote volume click New Volume The Cluster List window opens shown in Figure 222 Cluster List ES Select a cluster to contain the volume Clusters from management groups that you are logged in to Cluster SS Se Cancel Figure 222 Selecting a Cluster for the Remote Volume MPC DataFRAME 440 Storage System Software User Manual 343 Remote Copy 3 Select a cluster for the remote volume and click OK The New Volume window opens shown in Figure 223 See Chapter 12 Working with Volumes for detailed information about creating volumes New Remote Yolume New Remote Volume Type Volume Name Volume D Description Cluster TransData_BkUp Replication Level None Replication Priority Size Hard Threshold
230. er_0 DF420 1 Snap_Sch_O Snapshot_0 Available DataFRAMES I Management Groups Alert Messages EBSD Hosts These DataFRAMEs are available for use Name IP Software Version Model DF140 1 192 168 0 9 6 3 41 2605 0 NSM160 Figure 10 Network View Features 26 MPC DataFRAME 440 Storage System Software User Manual Getting Started Changing Which DataFRAMEs Appear in the Network View 1 Click the Find menu 2 Select Clear All Found Items to remove all DataFRAMEs and EBSD drivers from the Network View 3 Perform a Find using either method Finding by Subnet and Mask on page 21 or Finding by Module IP or Host Name on page 23 to find the desired set of DataFRAMEs Status of DataFRAMEs The Network View graphically depicts the status of each DataFRAME DataFRAMEs on the network are either available or part of a management group Other graphical information in the Network View depicts the storage architecture you create on your system An example setup is shown in Figure 10 Management Groups Management groups are groups of DataFRAMEs within which one or more DataFRAMEs are designated as managers e Clusters Clusters are sub groupings of DataFRAMEs within a management group Volumes Volumes are data storage areas created on clusters e Snapshots Snapshots are read only copies of volumes created at specific points in time If you are logged into the module a pink
231. es 283 Completing the Volume List ccecceeceeeeeeeneeeeeeeeeeeceaeeeaeeesaeeeaeeseeeeaeeseaeeeaeeseaeeseeeaeeeaeeeeeeseaeeseeeeeeesiaeseaeesieeees 284 Editing EW de TEE 284 MPC DataFRAME 440 Storage System Software User Manual xi Contents Opening the Volume List to Edit EE 284 Editing Volume Permission Levels AAA 285 Changing Authentication Groups in a Volume List 0 0 eeceeeceeeeeeeeeetneeeeeeeeeeeeeeeeeaeesaeeeeeeeeaeeseaeeseaeenaeeesas 286 Removing a Volume from a Volume List A 286 Deleting a Volume List uge Sel eee doled deanna eed 287 Selecting an Authentication Group from the List c ceceeccceeeeeeeeeseeeeeeeeeeeeceseeseeeeeseeseaeeeaeesaeeesaeeseaeescaeseaeeeneeeeas 287 Selecting a Volume List from the List EE 288 Chapter 16 Feature Registration cccsssseesseeeeeesseeeeeeesseeseeseseeeeesneeneeseseeneeeenseenens 289 Add On Features and Applications Registration Overview e cceeceseeeeeeeceneeeeeeeeeeeteaeeeeeenaeesaeeeeaeeseaeeeaeeeneeesas 289 Evalliating Features inaa aaraa E E a Taa aA a a E TAAA a EA AAE aE EAA aA Aar ANE 289 30 Day Evaluation Pernod jxzccciiccsisiecedasictiazecnettslegacacetesececeasiedususechdsacdessdeannaspibacdaedascfenpandasstgenanensPeaeesnes 289 Tracking the Time Remaining in the Evaluation Period AAA 290 Evaluating the Scalability Pak unze incite fot dae eee tec eed ENEE Een Starting the License Evaluation Period Backing Out of the License Evaluation Period AA 2
232. es a new volume Access Volume Wizard The Access Volume Wizard creates access privileges for a volume Ca C Show Launch Pad on startup This Wizard Launch Pad and the wizards are available via the Help menu 30 Click on Next to continue Create Yolume Wizard EZ Welcome to the Create Volume Wizard This wizard helps you mpc Create a volume for the first time or Add another volume to an existin DataFRAME 3 management group www mpccorp com Before you start this wizard you must Discover your DataFRAMEs on the network Run the Find DataFRAMEs Wizard You can exit the wizard at any time by clicking Cancel To continue click Next MPC DataFRAME 440 Storage System Software User Manual 11 Getting Started 31 Select New Management Group and click on Next Management groups allow you to manage and configure storage systems as a group See Working with Management Groups on page 157 for additional information on management groups x Choose a Management Group Select new or existing management group mpc Do you want to create a new management group or use an existing one New Management Group Existing Management Group A management group is a container in which you cluster DataFRAMEs and create volumes To continue click Next 32 Enter the name of your new management group and click on Next Create Volume Wizard xi Create a Management
233. eshold and soft threshold change to 0 length 1 Log in to the management group containing the volume that you want to convert 2 Right click the volume in the network view and select Edit Volume The Edit Volume window opens 3 Change the Type from Primary to Remote Notice that the window changes to the Edit Remote Volume window and all the fields are greyed out as shown in Figure 234 Additionally the values in the size hard threshold and soft threshold fields are set to 0 Edit Volume Edit Remote Volume Type Primary Remote Type Primary Remote Volume Name Volume Name Description Description Created Created Cluster TransData05 Cluster Replication Level 2 Way v Replication Level Replication Priority Availability Redundancy Replication Priority Size 8 0 GB v Max Size Hard Threshold 3 0 GB v Max Hard Threshold Soft Threshold 3 0 GB v Max Soft Threshold Auto Grow Off Auto Manual Auto Grow Checksum Data Yes No Checksum Data Cancel Cancel E nmary volume with size hard and electing Remote C anges the volume to soft thresholds defined a remote volume with the size hard and soft thresholds set to 0 Figure 234 Volume Changed from Primary to Remote 356 MPC DataFRAME 440 Storage System Software User Manual Remote Copy 4 Click OK The Make Volume Remote window opens shown in Figure 235 Make Yolume Remote Make Volume
234. estored volume to the original volume list For Fibre Channel volumes add a new LUN number as well 9 Reconfigure hosts to access the new volume Warning The original volume is deleted as part of the rollback Deleting a Snapshot Deleting a snapshot removes that snapshot s data from the DataFRAME and removes the snapshot from the Network View The writable space associated with the snapshot is deleted Prerequisites Note These prerequisites do not apply to Fibre Channel snapshots e Stop applications from accessing the snapshot e Disable the snapshot if it is mounted by a host 1 Log into the management group that contains the snapshot that you want to delete The management group Tab View opens 2 Select the snapshot that you want to delete 3 Review the Details tab to ensure you have selected the correct snapshot 4 From the Tasks menu on the Details tab click Delete Snapshot A confirmation message opens 5 Click OK Warning Deleting a snapshot causes that snapshot s data to be unavailable from the DataFRAME MPC DataFRAME 440 Storage System Software User Manual 253 Working with Snapshots Selecting a Snapshot from the List Select a snapshot from the list and click OK Selecting a Snapshot Schedule from the List Select a snapshot schedule from the list and click OK 254 MPC DataFRAME 440 Storage System Software User Manual 14 Working with Scripting The Storage System Software provides applica
235. eteaeeeaeeesaeesaeeseeeesaeseaeeeeeseaeeseeeseaees 32 ETIKETT 32 Logging In to and Out of the DataFRAME 33 Logging In to Additional Data FRAMES 33 Logging Out of the DataFRAME AA 34 DataFRAME Configuration Window c cceccececeeeeeeeneeeeeeeeeeeseaeeeaeeeeeceaeesaeeeeaeesaaesaeeeaeeesaeeeieeseaeesaeseeeenaeeeeeeseaees 34 Configuring Multiple DataFRAMES AA 35 Chapter 2 Working with DataFRAMES ccccccssseeseeeeeeeeeseeeeeeeeeneeeeenseeeeeeenneeneeeeeeneeeeneeeees 37 DataFRAME Configuration Window Overview ccccccceeseeeeneeeeeeeeeeeeneeeeeeeeaeeseaeeeaeeeseeesaeesieeseaeeseaeeeaeessaeeneeeteaees 37 Logging In to the DataFRAME AAA 38 Logging In to Additional DataFRAMES AAA 38 Closing the DataFRAME Configuration Window c eecceesceeseeeeeeeeneeseaeeeaeesaeeseeeseaeessaeseaeseaeeseeeeseeeeeaeesieeeneeenaes 39 Logging Out of the Data FRAME AA 39 Closing the DataFRAME Configuration Window without Logging Out 40 Module Configuration Overview cccccesceeeeeeeeeeeeeeeeeeeaeeeaeeeeaeeseaeeaeeeaeeeeaeeseeseaeesaeeseaeeseeeneeesaeseaeseneesnaeeseeeseaees Changing the DataFRAME Host Name tntu ntEunE AAE AEEENEEAEEEEEEEA EENE EE EEn EE nennen nenne eant Changing SE EE Upgrading the Storage System Software Copying the Upgrade Files from CD or FTP Site oo eeceeeeeeesceeeeeeeeeeceeeeeeeeeeeeeeaeeseaeenaeeseeeseaeeeseeseaeeeneeeeaes 42 Upgrading the DataFRAME AAA Backup and Restore of DataFRAME Configuratio
236. etwork should automatically start The Share Optional on page 315 for autorun window opens asking you to information about the tar gz files verify that you want to run Autorun 2 Run the script install sh located in opt The autorun window may be behind Storage_System_Software 6 x Drivers the CD folder directory which also EBSD opens 3 Click Yes to have the automatic install process run The automatic installation starts and steps you through the install process 4 On the Choose Product Component window select EBSD Driver 5 Click Next The Choose Install Folder window opens 6 Choose to accept the default directory opt Storage_System Storage_System_Software 6 x Drivers EBSD 7 Review the Pre installation Summary window and click Install 8 Click Done on the Congratulations window NOTE See Configuring the EBSD Driver for Linux on page 321 for information about configuring the EBSD driver Location of the Installed Driver Files The installation installs the EBSD driver into the following locations e usr local sbin for ebsdvm e opt Storage_System Storage_System_Software 6 x Drivers EBSD uname r etc for ebsd conf sample e etc init d for ebsd e Script also installs run level links in etc rc d Upgrading the EBSD Driver Using the CD or Driver Bundles 1 Stop all operations to ebsd devices e unmount ebsddisk 2 Install driver from Installation CD or the driver bund
237. evaluation begins During the 30 day evaluation period you can create volumes with 2 or 3 way replication add a hot spare to the cluster or configure a virtual manager MPC DataFRAME 440 Storage System Software User Manual 291 Feature Registration Please read Using Hot Spares on page 191 Planning Data Replication on page 214 and Chapter 10 Disaster Recovery Using A Virtual Manager before working with these features Backing Out of the License Evaluation Period If you decide not to purchase the Scalability Pak you must reduce the cluster to one DataFRAME The features you are evaluating dictate the steps required to safely back out of the evaluation configuration particularly if you want to save any volumes or snapshots in the test configuration 1 Back up any volumes you plan to retain The table below describes additional steps to safely back out of the Scalability Pak evaluation Table 52 Safely Backing out of Scalability Pak Evaluation Feature Being Evaluated Steps to Back Out Multiple DataFRAMEs with a large If volume is too large to fit on a cluster with one DataFRAME volume do one of the following Delete the volume Move the volume to another single note cluster with adequate capacity Add storage to the DataFRAME 2 or 3 way replication Set volume replication level to none Virtual manager Stop virtual manager 2 Remove the extra DataFRAMEs from the cluster Ev
238. ew name for the file 4 Click Save The configuration file is saved as a binary file in the folder you selected 5 Click OK to close the Backup Configuration window Backing Up a Management Group With Remote Copy Relationships If you back up a management group that is participating in Remote Copy it is important to back up the other Remote Copy management groups at the same time If you back them up at different times and then try to restore one of the groups the back up files will not match which could cause problems with the restore MPC DataFRAME 440 Storage System Software User Manual 177 Working with Management Groups Saving the Management Group Configuration Description Click Save Configuration Description The Save window opens Navigate to the location in which to store the configuration description text file Use the default name or type a new name for the file gt UO N Click Save The configuration description is saved as a text file in the folder you selected 5 Click OK to close the Backup Configuration window Restoring a Management Group For disaster recovery you can use the management group binary file to recreate a management group The restore procedure restores everything except snapshots since the data stored in volumes and snapshots is gone Requirements for Restoring a Management Group e Hardware You must have the same number of DataFRAMEs available that are the same capacity or
239. experiences a failure and a loss of quorum To regain quorum you start the virtual manager on a DataFRAME that is operating and in the site that is operational or primary depending upon your situation Requirement What it Means Use a Virtual Manager with an Even Number of Regular Managers Running on of DataFRAMEs o DataFRA Total of Disaster MEs Managers Recovery Runnin Including the Scenario g Virtual Regular Manager Managers Two sites with 4 5 shared data Two DataFRAMEs in 2 3 Management Group Add a Virtual Manager When You cannot add a virtual manager after quorum has been lost The Creating Management Group virtual manager must be added to the management group before any failure occurs A Virtual Manager Must Only Only one instance of a virtual manager must run at a time Once Be Started Once and Run you start a virtual manager you must not start that virtual manager Only Until the Site is Restored a second time The virtual manager should run only until the site is or Communication is Restored restored and data is resynchronized or until communication is restored and data is resynchronized Illustrations of correct uses of a virtual manager are shown in the first example of two site failure scenarios Figure 128 It is important to only start a virtual manager once In the second exampleFigure 129 illustrations of incorrect uses of a virtual manager are shown MPC
240. face using the Configuration Interface to access the DataFRAME See Configuring a Network Connection on page 304 Disabling a Network Interface 1 Select from the list on the TCP IP window the interface to disable 2 Click Edit The Edit TCP IP Configuration window opens shown in Figure 58 3 Click Disable Interface 4 Click OK A confirmation message opens If you are disabling the only interface the message warns that the DataFRAME may be inaccessible if you continue 5 Click OK If DataFRAME is in a Management Group If the DataFRAME for which you are disabling the interface is a manager in a management group a window opens which displays all the IP addresses of the managers in the management group and a reminder to reconfigure the application servers that are affected by the update Configuring a Disabled Interface 98 If one interface is still connected to the DataFRAME but another interface is disconnected you can reconnect to the second interface using the Console See Configuring the IP Address Manually on page 83 If both interfaces to the DataFRAME are disconnected you must attach a terminal or PC or laptop to the DataFRAME with a null modem cable and configure at least one interface using the Configuration Interface See Connecting to the Configuration Interface on page 301 MPC DataFRAME 440 Storage System Software User Manual Managing the Network TCP Status Review the status of th
241. figuration_Backup Files of Type All Files ba Save Cancel Figure 34 Backing up the DataFRAME Configuration File 2 Navigate to a folder on the Console computer to contain the DataFRAME configuration backup file 3 Accept the default name DataFRAME_Configuration_Backup or enter a new name for the backup file Note The configuration files for all DataFRAMEs that you back up are stored on the computer running the Console If you back up multiple DataFRAMEs be sure to give each DataFRAME configuration file a unique and descriptive name This will make it easier to locate the correct configuration file if you need to restore the configuration of a specific DataFRAME 4 Click Save 46 MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs Restoring the DataFRAME Configuration from a File Use Restore to restore the configuration of a DataFRAME 1 On the Backup and Restore tab click Restore The Restore DataFRAME window opens shown in Figure 35 xi File Name Browse Version No file name Description No file name Install DataFRAME Version Management Group Cluster 4 DF420 1 192 168 6 3 33 0007 0 Available DataFRA vi DF140 192 168 0 7 6 3 41 2605 0 Available DataFRA J DF140 1 192 168 6 3 41 2605 0 Available DataFRA T DF420 192 168 0 3 hy Check all Clear All Install file on selected DataFRAMEs one at a time Recommended Install f
242. figurations cccccssssssssseeeeeeeeeeeeees 367 WEE 367 Using Remote Copy for Business Continuance cceecceeeceeeeeeeeeeeneeeeeeeeaeesaeeeeaeeeaeeseaeseaeeeeeeeseeesnaeesaeeeaeeeeeeesiaes 367 Achieving High ET EE 367 Configuration for High Availability 367 How This Configuration Works for High Availability 368 Best EICHER Zeeche eege gereest Eder eege eege 370 Achieving Affordable Disaster Recovery esccccesseeceeneeeeeneeeseeneeeeeaeeeeeneeeseeeeseaeeeenaaeesesneeeneaeeenennaeeneaes 372 Configuration for Affordable Disaster Recovery ccecccceceeeeeeeeeeeeeeeneeeeaeeeeeeteaeeseeseaeesaeeseaeeneeteaeeeeeesas 372 How this Works for Affordable Disaster Recovery escceeseseeeeeseeeteeeeeeeneeeteneeeeesaeeeesaeesenaeeeseneeeeeeanereeaes 373 Best PIACHCOS gs ccencnccscedavucecsescecageesaceuuubenzenecaggeteupaaetssteennszeravisvass exdsaawidenes sagen Ee 374 Using Remote Copy for Off site Backup and Recovery cecceeeceeseeeeneeeneeeeeeseeeeeaeeeaeeeaeeseeeesieeseaeteaeeeeeeneeensaes 376 Achieving Off site Tape Backup ccccesceeeeeesneeeeeeeeeeseaeeeaeecaeeeneeeeaaeseaeeseaeeeaeeseeeeeeeseaeeeaeesneeesaeeseeeeeaeees 376 Configuration for Off site Backup and Recovery ccecceeceeeeeeceeeeeeeeeaeeeaeeseeeeeeeeteeeeeaeeseaeeeaeeteaeeseeeeaeeee 376 How This Configuration Works for Off site Tape Backup cccceeceeeseseceeeeeeeseeeeeeeeseeteaeeeaeeeaeeseeeneetenees 377 Best ee ET
243. figurations for each network interface and to bond the network interfaces Identifying the Network Interfaces The DataFRAME comes with two onboard Gigabit Ethernet ports These ports are named Motherboard PortO and Motherboard Port1 and are labeled on the back of the DataFRAME as listed in Table 9 Table 9 Identifying the NICs in the Motherboard Motherboard Interfaces Where labeled What the label says TCP IP Network Configuration Category in the Console TCP IP tab e TCP Status tab Name Motherboard Port0 Motherboard Port1 Description MPC Gigabit Ethernet Configuration Interface Name Motherboard Port1 Motherboard PortO Label on the back of the DataFRAME NICs 1 amp 2 80 MPC DataFRAME 440 Storage System Software User Manual Managing the Network Table 10 Identifying Add on NICs Add on Interfaces Where labeled What the label says TCP IP Network Configuration Name Gott Port Slot1 Port1 and so on Ge Description MPC Gigabit Ethernet e TCP IP tab e TCP Status tab Configuration Interface Name Slot1 PortO Slot1 Port1 and so on Label on the back of the Port A DataFRAME Port B Port C Port D The motherboard interfaces are labeled NICs 1 and 2 on the back of the DataFRAME Figure 55 illustrates the MPC DataFRAME 440 Storage System Motherboard Port 1 64 bit PCI Slot Motherboard Port 0 32 bit PCI Slots Figure 55 Network Interface Ports
244. full set of snapshots You can continue to make snapshots even though one side was rolled back and the other side was not When deciding whether to roll back the primary or remote volume consider the following e When you roll back the primary snapshot to a primary volume any applications accessing the primary volume will no longer have access to the most current data as the primary volume has been rolled back to a previous state If the primary volume must be synchronized with other volumes accessed by the same application consider rolling back the remote volume instead Figure 250 shows rollback of the primary snapshot while leaving the remote snapshots intact MPC DataFRAME 440 Storage System Software User Manual 379 380 Sample Remote Copy Configurations Application Workstations Clustered Application Application Serer Servers Read Mitte Primary Storage Subnet Primary Snapshot 1 00 PM Primary Snapshot 77 00 AM Restore Serer 47 00 AM Figure 250 Non destructive Rollback from the Primary Snapshot e To roll back the remote snapshot you must first make the remote volume into a primary volume This will stop scheduled creation of remote snapshots which may jeopardize your high availability disaster recovery or routine backup strategies Figure 251 shows rollback of the remote snapshot MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations Application
245. ge 281 3 Configure client access to the snapshot volume Accessing the snapshot e as a source volume for data mining and other data use e as a source volume for creating backups e for data and file system preservation before upgrading software e for protection against data and file system corruption e for file level restore without tape or backup software Creating Snapshot Schedules Using the Console you can schedule recurring snapshots Recurring snapshots can be scheduled in a variety of frequencies and with a variety of retention policies Note Scripting snapshots can also take place on the client side Scripted snapshots offer greater flexibility for quiescing hosts while taking snapshots and for automating tasks associated with volumes and their snapshots Scripting of snapshots is an add on feature You must purchase the Configurable Snapshot Pak to use snapshot scripting beyond the 30 day evaluation period Requirements for Scheduling Snapshots Scheduled snapshots require particular attention to capacity management Additionally you must ensure that the time settings on theDataFRAMEs running managers and the time setting of the management group are synchronized Note Use NTP to ensure that all the DataFRAMEs in the management group have synchronized time settings 246 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Table 43 Requirements for Scheduling Snapshots Requirement What it
246. ghty fT 18 87 2 RAD States normal Veter Changed 82 G607 2005 125111 Pu MST Anni Groen temestore 10 0 3 199 CMU Fan Speed D Vatu is baiow 3000 zt DR 7 2005 125011 OR MST Aan trom Maiti F117 7 Gerace fash 1 Status Acia Vater Chaneod Figure 14 Viewing messages in the Alert Messages Tab The most recent alert is at the top The messages are continuous while the Console is open When you close the Console the messages are cleared MPC DataFRAME 440 Storage System Software User Manual 29 Getting Started EBSD Hosts Tab The EBSD Hosts tab shown in Figure 15 lists all versions of the EBSD drivers that are currently installed on the network It also lists information about the hosts that are using that driver See the EBSD Linux Manual for more information about the EBSD drivers Available DataFRAMEs Management Groups Alert Messages EBSD Hosts HostName I P Groups amp Volumes Type I INTG SVR 1 1006136 Volume_3intest_610rv Windows 2003 61 200001 INTG SRVR 1 10 0 11 219 Windows 2003 6 1 200001 Figure 15 EBSD Hosts Tab Viewing Storage System Module Details Select a DataFRAME from the Network View and the DataFRAME Details tab opens in the Tab View shown in Figure 16 mec Computers DataFRAME Management Console E DI sl File Find Tasks View Help SS Available DataFRAMEs DF140 DF140 1 Details for selected DataFRAME DF420 DF 420 1 Details Management Gr
247. gical Motherboard PortO and Slot1 PortO are interface both active Motherboard Port0 is Active e Glo Port is Active 2 Motherboard Porto interface fails e Motherboard PortO status becomes Because NIC aggregation is Passive Failed configured Slot1 Port0 continues Slot1 PortO status remains Active operating 3 Motherboard Porto link failure is e Motherboard PortO resumes Active repaired status e Slot1 PortO remains Active Summary of NIC States During Failover Table 17 shows the states of Motherboard PortO and Slot1 PortO when configured for NIC aggregation Table 17 NIC Status During Failover with NIC Aggregation Failover Status Status agate erboard Status of Slot1 PortO Normal Operation Preferred No Preferred No Status Active Status Active Data Transfer Yes Data Transfer Yes MPC DataFRAME 440 Storage System Software User Manual 91 Managing the Network Table 17 NIC Status During Failover with NIC Aggregation Failover Status Status of Motherboard Porto Status of Slot1 Port Motherboard Porto Fails Data Transfer Continues on Slot1 PortO Preferred No Status Passive Failed Data Transfer No Preferred No Status Active Data Transfer Yes Motherboard Port Restored Preferred No Status Active Data Transfer Yes Preferred No Status Active Data Transfer Yes Example Network Configurations with NIC Aggregation Two sim
248. gle logical interface This logical interface acts as the master interface controlling and monitoring the physical slave interfaces 84 MPC DataFRAME 440 Storage System Software User Manual Managing the Network Bonding two interfaces for failover provides fault tolerance at the local hardware level for network communication Failures of NICs Ethernet cables individual switch ports and or entire switches can be tolerated while maintaining data availability Bonding two interfaces for aggregation provides bandwidth aggregation and localized fault tolerance Depending on your DataFRAME hardware network infrastructure design and Ethernet switch capabilities you can bond NICs in two ways e Active Backup You specify a preferred NIC for the bonded logical interface to use If the preferred NIC fails then the logical interface begins using another NIC in the bond until the preferred NIC resumes operation When the preferred NIC resumes operation data transfer resumes on the preferred NIC e NIC Aggregation The logical interface uses both NICs simultaneously for data transfer This configuration increases network bandwidth and if one NIC fails the other continues operating normally Warning NIC aggregation requires plugging both NICs into the same switch This means that NIC aggregation does not protect against switch failure You can create bonds of 2 or 4 NICs A NIC can only be in one bond Best Practices NIC aggr
249. group before you can delete any clusters within that group 1 Log in to the management group that contains the cluster you want to delete 2 Select the cluster you want to delete The Cluster Tab View opens 3 From the Tasks menu on the Details tab select Delete Cluster A confirmation message opens If the message says that the cluster is in use you must delete the snapshots and volumes on the cluster first 4 Click OK The cluster is deleted and the DataFRAMEs return to the management group as available Selecting a Cluster from the List The Select Cluster list window opens when you select Tasks gt Cluster from the menu and then select one of the following options MPC DataFRAME 440 Storage System Software User Manual 209 Working with Clusters 210 MPC DataFRAME 440 Storage System Software User Manual 12 Working with Volumes A volume is a logical entity that is made up of storage on one or more DataFRAMEs It can be used as raw data storage or it can be formatted with a file system and used by a host or file server You create volumes on clusters of one or more DataFRAMEs After you create a volume you must add it to a volume list which is associated with an authentication group Volume lists and authentication groups control access to volumes by application servers For detailed information see Chapter 15 Controlling Client Access to Volumes This chapter covers the following topics e Planning volume size and
250. guration of the deleted logical interface The other NIC is disabled and its IP address is set to 0 0 0 0 When you delete a NIC aggregation bond one of the active interfaces in the bond retains the IP address of the deleted logical interface The other NIC is disabled and its IP address is set to 0 0 0 0 1 On the TCP IP tab select the bond that you want to delete 2 Click Delete Bond Because the IP addresses change the Search for DataFRAMEs window opens For detailed information see Finding Storage System Modules on the Network on page 19 of Getting Started on page 1 3 Finish searching for the DataFRAME using the desired method Finding the DataFRAME might take a few minutes You can exit the search window and use the Find menu at your convenience MPC DataFRAME 440 Storage System Software User Manual 97 Managing the Network Disabling a Network Interface You can disable the network interfaces on the DataFRAME e You can only disable top level interfaces This includes bonded interfaces and NICs that are not part of bonded interfaces e To ensure that you always have access to the DataFRAME do not disable the last interface If you want to disable the last interface first enable another interface Warning If you disable an interface be sure you enable another interface first That way you always have access to the DataFRAME If you disable all the interfaces you must reconfigure at least one inter
251. guratton nen nenn 368 High Availability Configuration During Failover ccceceesceseeeseeeceneeeneeeeeeesaeeseaeeeseeseaeeeaeeseeeeseeeeeeenaas 369 High Availability Configuration During Failback A 370 High Availability During Failover Example Configuration cccceeseeeseeeseeeeeeeeeeeseneeeeeeseeeeseeeeeneenaas 371 Affordable Disaster Recovery Example Configuration cccceecccsseeeeeeeeseeteneeeneesseeeseeeseneeseaeeeeesieees 372 Restoring from a Remote Volume ccesccccesseeeeeeceeeeneeeeeeeeeesneecenaeeeeeseaseesaeeseeeeeeeeeeeesneeseneneeeeenees 373 Restoring from Tape Backup 374 Off site Backup and Recovery Example Configuration ccccescceeseeseeeeeneeeeneeeeeeneeeseeeeeaeestaeseaeeeeees 377 Non destructive Rollback Example AAA 379 Non destructive Rollback from the Primary Snapshot A 380 Non destructive Rollback from the Remote Snapshot cccesceeeeeseeeeeeeeeeeeeeseeseeeeeeeeeeeenaeeseeeeeneenea 381 Data Migration Example Confiouration 382 Configuration after Data Migration ccccseeeeseeeseeeeseeeeeeceeeeeeeeeneecaeeeaeessaeesaeeseeseaeeseeeeeaeeeieeeeeeeieees 383 MPC DataFRAME 440 Storage System Software User Manual xix List of Figures XX MPC DataFRAME 440 Storage System Software User Manual List of Tables List of Tables Table 1 DataFRAME Contiguration Tasks s 2 c0scscedunad adi e a SN NEEN ge 14 Table 2 Boot Flash Card Giatus AE 54 Table 3
252. h cards If one card fails or is removed the system can still boot If you remove and replace one of the cards you must activate the card to synchronize it with the other card There must always be at least one active flash card in the DataFRAME If you are upgrading the Storage System Software a dual boot device DataFRAME must contain both flash cards MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs Checking Boot Device Status in a DataFRAME You can view the compact flash card status on the Boot Devices window 1 Log in to the DataFRAME 2 Click the Boot Devices tab The Boot Devices window opens shown in Figure 41 e Edit Configuration for DF140 192 168 0 7 Ka T Storage Time a TCP IP Network Sm DataFRAME Adminis ga SNMP ki Reporting Power Off Feature Registration Boot Devices Module Information Backup amp Restore Device Status Active Active Figure 41 Viewing Boot Device Status with Two Devices MPC DataFRAME 440 Storage System Software User Manual 53 Working with DataFRAMEs 3 The status of each compact flash card is listed in the Status column Table 2 Boot Flash Card Status Flash Card Status Description Active The device is synchronized and ready to be used Inactive The device is ready to be removed from the DataFRAME It will not be used to boot the DataFRAME
253. hanges rather than the entire volume e Remote Copy is robust If the network link goes down during the process copying resumes where it left off when the link is restored MPC DataFRAME 440 Storage System Software User Manual 333 Remote Copy Planning for Remote Copy Remote Copy works at the management group cluster volume snapshot and DataFRAME level Table 59 Remote Copy and Management Groups Clusters Volumes Snapshots and DataFRAMEs Storage System Level Remote Copy Configuration Management Groups Remote snapshots can be created in the same management group or in a different management group than the primary volume e If using different management groups the remote bandwidth setting of the management group containing the remote volume determines the maximum rate of data transfer to the remote snapshot Clusters Remote snapshots can be created in the same cluster or ina different cluster than the primary volume Volumes Primary volumes contain the data to be copied to the remote snapshot e Data is copied to the remote snapshot via the remote volume The remote volume is a pointer to the remote snapshot The remote volume has a size of 0 Snapshots e Once data is copied from the primary snapshot to the remote snapshot the remote snapshot behaves as a regular snapshot DataFRAME Active monitoring of each DataFRAME notifies you when copies complete or fail Active monitoring also
254. hat is part of a management group If you want to reduce the capacity of a DataFRAME first remove it from the management group Then remove disks from the DataFRAME and reconfigure RAID 76 MPC DataFRAME 440 Storage System Software User Manual Storage Memory Requirements for Adding Disks Before you add disks to the DataFRAME confirm that the DataFRAME has enough memory to use the additional disks Table 7 summarizes the memory requirements by disk capacity of fully populated DataFRAMEs with RAID 0 RAID 1 10 and RAID 5 50 configurations Contact your DataFRAME supplier for additional memory Table 7 Memory Requirements for Fully Populated DataFRAME MPC DataFRAME Memory Requirement for 12 or 16 Disks Fully 440 Storage Populated System RAID Level For 250 GB Disks For 400 GB Disks RAID 0 1 GB 2 GB RAID 1 10 1 GB 1 GB RAID 5 50 1 GB 2 GB Adding Disks Prerequisite Before you add disks to the DataFRAME be sure that the DataFRAME has enough memory to use the additional disks See Memory Requirements for Adding Disks on page 77 Warning Adding a disk to the RAID deletes any existing data on that disk 1 Add the new disks to the DataFRAME You must add disks in contiguous disk bays from left to right as shown in Figure 53 2 Using the Storage System Console log in to the DataFRAME 3 Select the Storage configuration category 4 Click the Disk Setup tab to bring it to the front
255. hedule may create primary snapshots more frequently than the remote copy process can keep up with The retention policies for scheduled remote copies ensure that such factors do not cause primary and remote snapshots to become unsynchronized Regardless of the retention policy defined for scheduled remote copies up to 2 additional snapshots may be retained by the system at any given time These two additional snapshots include the snapshot that is in the process of being copied and the last fully synchronized snapshot A fully synchronized snapshot is one that has completed copying so that the remote snapshot is a complete mirror of its corresponding primary snapshot Up to two additional snapshots may be retained by the system at any given time Because the system will never delete the last fully synchronized primary snapshot a remote copy schedule may retain N 2 copies for a retention policy of N the currently copying remote snapshot plus the last fully synchronized snap shot Using the example above if you have a retention policy for your remote MPC DataFRAME 440 Storage System Software User Manual Remote Copy copy schedule of two primary and five remote snapshots the system may retain up to four primary and seven remote snapshots for a period of time Table 60 Snapshot Retention Policy and Maximum Number of Snapshots Retained Remote Schedule Retention Policy Maximum Number of Snapshots Retained n of primary snaps
256. her location on the network such as a file server 8 Click Next Review the Pre installation Summary window 9 Click Install The EBSD driver bundle is copied into the directory you specified The driver bundle contains tar gz files for all the versions of Linux that are currently supported plus an install sh file to install the driver With Autorun Not Enabled 1 Insert the driver CD in to the CD drive of the EBSD host server 2 Open a terminal window 3 Navigate to the VM directory on the cd cd mnt cdrom Disk1 InstData VM This directory contains the EBSD6x_setup bin binary file that will launch the automatic install 4 Launch the automatic install by issuing the following command sh EBSD6x_setup bin The automatic installer launches Installing the EBSD Driver Using the CD or the Driver Bundles You can install the EBSD driver locally using the CD or using a driver bundle from a network share See Copying Driver Bundle to a Network Share Optional on page 315 316 MPC DataFRAME 440 Storage System Software User Manual Using the EBSD Driver for Linux Installing from the CD Installing from a Network Share 1 Start the Windows Manager of your 1 Optional Copy one or all tar gz file s and choice such as KDE or GNOME the install sh file to the computer on which 2 Insert the driver CD into the CD drive you want to install the driver of the EBSD client PC Autorun See Copying Driver Bundle to a N
257. hile setting up active monitoring specify alert actions for one variable and then check the box to apply those actions to all variables on the DataFRAME This setting applies the same email address and other alert settings to all DataFRAMEs Then if you need to customize alert actions for a particular variable you can edit that variable 6 Click Finish when you have configured all the threshold items in the list Downloading a Variable Log File To save the history of a variable download a copy of the log file 1 In the list of monitored variables click the variable for which you want to save the log file 2 Click Download Log on the Active Reporting window The Save Variable Log File window opens 3 Choose a location for the file 4 Optional Change the name of the log file 5 Click Save The file is saved to the location you specified Viewing the Variable Summary You can review the frequency settings and the triggers for a variable in the Monitored Variables list without editing the variable 1 In the list of monitored variables select a variable The frequency thresholds and notification settings display to the right of the list ER Configuration for DF140 192 168 0 7 xi Module T Storage Time D TCP IP Network e DataFRAME Adminis ge SNMP p Passive Log Files Active Email Diagnostics Alerts Monitored Variables Summary Frequency 15 seconds CDU Temperatur
258. hold as necessary to manage capacity growth Table 40 and Table 41 illustrate how storage can be effectively managed by setting the hard threshold below the original volume size in a series of snapshots as illustrated below Table 40 Space used by Snapshots when Hard Threshold is set to the Original Volume Size Snapshot Size Da Volume Data Stored w No ome song y Snapshot or Changed Threshold Cluster Change Mon Original Volume N A 50 GB 50 GB 50 GB Tue Snapshot 1 lt 15 GB 50 GB 100 GB Wed Snapshot 2 lt 10 GB 50 GB 150 GB Thur Snapshot 3 lt 8 GB 50 GB 200 GB Table 41 Space used by Snapshots when Hard Threshold is Reduced Snapshot Size Total Space Data Stored or w Hard Day Volume Snapshot Changed Threshold pte Reduced Mon Original Volume N A 50 GB 50 GB 50 GB Tue Snapshot 1 lt 15 GB 15 GB 65 GB Wed Snapshot 2 lt 10 GB 15 GB 80 GB Thur Snapshot 3 lt 8GB 15 GB 95 GB Note Note the dramatic savings in storage space in the table above Note Deleting files on a file system does not create space on the volume For file level capacity management use application or file system level tools 238 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Planning Snapshots When planning to use snapshots take the purpose and size considerations into account Note When considering the size of snapshots in the cluster
259. hot The process of copying the data may take some time The remote snapshot appears below the remote volume as shown in Figure 221 MPC DataFRAME 440 Storage System Software User Manual 341 Remote Copy Note If you create a remote snapshot of a volume with a remote snapshot still in progress the second remote snapshot will not begin copying until the first remote snapshot is complete a Remote_Backup a Transaction_Data a a Transaction_Data_Backup a a Transaction_Data s Denver 1 Denver 2 Boulder 1 Boulder 2 Remv VolCD04 RemSSCD04 Remote Volume Primary Volume and Remote Snapshot and Primary Snapshot Figure 221 Viewing the Remote Snapshot Creating a Remote Volume Then go to Completing the Remote Snapshot on page 341 You can create a remote volume by any of the following methods e Make an existing volume into a remote volume e Create a new remote volume during creation of a remote snapshot e Create a new volume from the cluster Details panel and then select the Remote option on the New Volume window Making an Existing Volume into a Remote Volume Selecting an existing volume to become a remote volume will cause e A snapshot of all existing data to be created for that volume and then e All the data in that volume will be deleted so that the remote volume will have zero length and zero hard and soft thresholds 342 MPC DataFRAME 440 Storage System Software Us
260. hot becomes the hard threshold of the writable volume and defines the amount of space allocated for changes to the original volume Note Setting the hard threshold smaller than the size of the original volume allows you to create snapshots that require less space on the cluster See Managing Capacity Using Volume and Snapshot Thresholds on page 237 8 Click OK when you are finished The Snapshots tab opens with the new snapshot listed The new snapshot also displays in the Network view as shown in Figure 159 MPC DataFRAME 440 Storage System Software User Manual 241 Working with Snapshots G Available DataFRAMEs J Group_0 E DF140 1 asks View Help DF420 1 mrc Computers DataFRAME Management Console File Find New Snapshot b k Remote Snapshot Schedules Targe Volume List Memberships Details Snapshot Schedules Hosts Remote Snapshot Name Hard Threshold Sof Threshold Created Snapshot_0 500 GB 500 GB 11 24 2005 03 46 12 AM Figure 159 New Snapshot K Note Snapshots are listed below the volume in descending date order from newest to oldest The Snapshot Tab View Clicking on the snapshot itself opens the snapshot Tab View shown in Figure 160 Target Information Volume List Memberships Remote Snapshot Schedules Details Name Created Cluste
261. hots X of remote snapshots n 2 primary snapshots X 2 remote snapshots n of hours for primary snapshots x of hours for remote snapshots n 2 primary snapshots older than n X 2 remote snapshots older than xx n of days for primary snapshots x of days for remote snapshots n 2 primary snapshots older than n x 2 remote snapshots older than xx n of weeks for primary snapshots x of weeks for remote snapshots n 2 primary snapshots older than n x 2 remote snapshots older than xx Remote snapshots will only be deleted after their corresponding primary snapshot is deleted Additionally a remote snapshot will only be deleted after its counterpart primary snapshot Therefore you can not retain fewer scheduled remote snapshots than primary snapshots when setting your retention policies Note If you retain more remote snapshots than primary snapshots the remote snapshots become regular snapshots when their corresponding primary snapshots are deleted You can identify them as remote snapshots by their names since the naming convention is established as part of creating the remote snapshot schedule Best Practices e Retain at least two primary snapshots to ensure that only incremental copying is required for primary snapshots e Review your remote copy schedule to ensure that the frequency of the remote copies correlates to the amount of time required to complete a copy Use the che
262. hots review Chapter 13 Working with Snapshots Planning Soft Thresholds Soft thresholds trigger alerts to system administrators to help ensure that hard thresholds are not exceeded Upon receiving an alert the system administrator can take steps to increase capacity according to planned capacity management See Managing Volume Growth Capacity on page 219 for strategies to manage volume growth Best Practice If Not Using Snapshots If the hard threshold is equal to the volume size set the soft threshold equal to the volume size as well Use application level monitoring to manage capacity growth Best Practice If Using Snapshots If the hard threshold is less than the volume size set the soft threshold to a percentage of the hard threshold When a soft threshold alert is received e Provision more storage for the cluster if required e Increase the hard threshold e Re adjust the soft threshold to be a percentage of the new hard threshold Planning Data Replication Data replication creates redundant copies of a volume You can create up to three copies using 3 way replication Because these copies reside on different DataFRAMEs replication levels are tied to the number of available DataFRAMEs in a cluster Hot spare DataFRAMEs are not available for data storage and therefore not available when calculating replication levels The Storage System Software and the Storage System Console provide flexibility when planning data
263. ia 317 Upgrading the EBSD Driver Using the CD or Driver Bundles AA 317 Installing the EBSD Driver with RPM Packages A 318 RPM Package Naming Convention ccccccsceeeeeeeeeeeeeeeeeeseaeeeaeeeneeesaeeseaeeeaeesaeseaeeseeeeeaeeseeseaeesieeeneeesaes 318 xii MPC DataFRAME 440 Storage System Software User Manual Contents Installing A Binary RPM Package A 318 Querying An Existing RPM Package ccccceeseeeeeseeeeeeceeeseaeeseeeeeaeetaaeeeaeeseaeeeaaeeeaeenaeeseeeeeaeesnaeeeaeesneeees 319 Uninstallingithe RPM siesta tev ied iste a dnl aes 320 Using the Source RPM Package to Build a Driver for a New Kernel Version cccssscceeeteeeesteeeees 320 Configuring the EBSD Driver for LINUX EN 321 Creating ebe cont AAA 321 Connecting the EBSD Devices to the DataFRAME EBSD Server assesssssseieeisersieseresrisrrsrrnrrssensirsersee 323 Venting SEIN 323 Mounting the Block Device EBSD Disk cccecceeeeeeeeeeeeeeeeeeeeseeeeeeesieeseaeeeaeesaeenaeeeaeeseaeesaeesnaeetieeseaeenaas 324 Adding an EBSD Disk at Runtime 0 e eee eee ceteeeeeetneeeeeeeeeeeteneeeaeecaaee cesses eesaesaeesaeeseeeesaeseaeseeesneeeseeseaeeseeetenees 324 Starting the EBSD Semice sss isaisip aboia laude bauedealeddaceadadeute easi adea aioa Seassa aeaa ina a aE Ta 324 Stopping the EBSD Driver as ccscccskecs codesecedstedeneheseexsuphesedvceyctetnagseadesdepsedateaiuey eacetucsdesestandtusensuepesdaeasaeelonsacavteaseeenass 325 Status of the EBSD Driver and Dev
264. ices 000 ec ceeccceeeeeeeeeeeeeneeeeeeeeaeeceaeeeaeeeaeeseaeeeaeeeaeseaeeeeeeeaeeseeeseaeesieeseaeeeneeens 325 Disconnecting an EBSD Device vicic s detec eeessesee tes avec idl tired ets Gnade Ea E Aenea tind divin 326 Disabling an EBSD Device AAA 326 Deleting an EBSD Device siirre iuen r al edt she RA A it cee lees 327 Uninstalling the EBSD Driver e cc ecceeeeeeeeeeceeeeeeeeeneeseeeeeaeesaeeeaeeeaeeseaeeeaeeeeaeeseaeeaeeeseeseaeeeaeeseeeseeeeneesieeeeaeenias 327 Finishing le EE 327 Troubleshooting izeten rr aE nero tte ree eevee aaeeee EE 328 Appendix D Remote CODY veer 329 Registering Remote Copy EE 329 Number of Remote Copy Licenses Required AA 329 Glossary for Remote Copy EEN 330 How Remote Copy Works AAA 331 Graphical Representations of Remote Copy eecceeeseeeeeeseeeeneeeeeeeeseeseaeeeaeecaeeeaeeeaeeeeaeeeeeeeeeseeeeseeeneeseaeesneeees 332 Remote Copy and Volume Replication EE 333 Uses for Remote Copy AE 333 Benefits of Remote Copy net NEE ach nin nti Milne lied 333 Planning for Remote Copy ceeccecceeseeeeeeeeeeeeeaeeeaeeeeeeeaeeeseeseaeesaeeesaeesaeeseeseaeeseaeenaeesaeeseaeeeeeseaeseaeeseeeesaeeseaeeeaees 334 Planning th Remote Snapshot sisii sssusa EENS de en ee ede 334 Using Schedules for Remote Copy c ceecceecceseeseeeeeneeeneeeeeeeseaeeeeeeeaeeeaeeeeeeseeseaeeeeesaeeeaeeseeeeseaseeeeneeesieeeeaeesaas 335 Planning the Remote Copy Schedule 335 Beet Praci OS sicihiecciccidedacsecidstacayesiaghage
265. icient number of DataFRAMEs are in the cluster to make the volume available For a detailed explanation see Table 35 on page 218 For example if you have 2 way replication with 3 DataFRAMEs in the cluster you can change from Availability to Redundancy if all the DataFRAMEs in the cluster are available and have enough space for replicating the data Size To increase the size of the volume one of the following conditions must be met There must be sufficient unallocated space in the cluster e You can move the volume to a cluster that has enough unallocated space e You can add a DataFRAME to the cluster To decrease the size of the volume Ifthe volume has been or is mounted by any operating system you must shrink the volume from the client operating system before shrinking the volume in the Console The size entered must be greater than the hard threshold You cannot decrease the volume size to a value less than the hard threshold e You also cannot decrease the size of the volume below the size needed for data currently stored on the volume Warning Decreasing the volume size or hard threshold is not recommended If you shrink the volume in the Storage System Console before shrinking it from the client file system your data will be corrupted or lost MPC DataFRAME 440 Storage System Software User Manual 229 Working with Volumes Table 38 Requirements for Changing Volume Parameters Item Requirements fo
266. iguring hot backup sites you can still use Remote Copy to protect your data during an emergency Using remote snapshots you can maintain copies of your volumes in remote sites Set up a schedule for creating remote copies and if your primary storage site becomes unavailable you can easily access the most recent remote copy of your data volumes You can also use remote snapshots to transfer data to a backup location where tape backups are then created This eliminates the backup window on your primary volumes and ensures that you have copies of your data in the remote site on DataFRAMEs as well as on tape Configuration for Affordable Disaster Recovery To configure affordable disaster recovery create remote snapshots of your volumes in an off site location In addition you can create tape backups from the remote snapshots in the off site location e Designate one or more off site locations to be the destination for remote snapshots e Set up a schedule for creating remote snapshots in the designated off site locations If your application server uses multiple volumes that must be in sync use a script to quiesce the application before creating remote snapshots e Create routine tape backups of the remote snapshots in the off site locations Configuration Diagram Application Workstations EH Tape licati Production EH Library Production Application Application Application Servers Serer Primary Remote Primary
267. ile on page 45 1 Log in to the management group 2 From the Tasks menu on the Details tab select Back up Configuration of Management Group The Back up Configuration of Management Group window opens shown in Figure 124 MPC DataFRAME 440 Storage System Software User Manual 175 Working with Management Groups Back up Configuration of Management Group Save Backup Configuration Management Group Name Transaction_Data Local Bandwidth 32768 Kb sec Remote Bandwidth 32768 Kb sec Database Version 6 3 Virtual Manager None Communication Mode Multicast Off Unicast On Unicast IPs 10 0 33 18 10 0 33 25 Authentication Groups Name CreditTrans Description Volume list Credit ZZ D Save Configuration Description Back Up Configuration Figure 124 Backing up the Management Group Configuration 176 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Backing Up a Management Group Configuration 1 Click Back Up Configuration The Save window opens shown in Figure 125 Save In Ef config_Backup x el EA eil E Ge File Name ManagementGroupDescription tet Files of Type Management Group Description Files Save Cancel Figure 125 Save Window for Backing up the Management Group Configuration 2 Navigate to the location in which to store the configuration binary file 3 Use the default name or type a n
268. ile on selected DataFRAMEs simultaneously Advanced Close Figure 35 Restoring the DataFRAME Configuration File 2 Select the Install check box next to the DataFRAME you want to restore You can select multiple DataFRAMEs to restore from the list 3 Select Install file on selected DataFRAMEs one at a time Recommended 4 Click Browse to navigate to the folder on the Console computer where the configuration backup file is saved 5 Select the file to restore and click Open Backup File 6 Review the version and description to ensure you are restoring the correct file 7 Click Install The Install Status window opens When the restoration is complete the Save to File and Close buttons become enabled To save a log file of the restore operation before rebooting click Save to File MPC DataFRAME 440 Storage System Software User Manual 47 Working with DataFRAMEs G install status DF140 192 168 0 7 In Progress Pre q jualification tests started Pre qualification tests finished Setting the user upgrade information started Setting the user upgrade information finished Gathering the install destinati wri Gathering the install destination information finished Opening the install connection started X Automatically scroll Savetorite Cancel After Current DataFRAME close Figure 36 Restoring the DataFRAME Configuration File 8 Click Close to fin
269. ime 5 27 05 2 12 PM Primary Snapshot CredDataRS1 Elapsed Time 3m 14s Remote Mgmt Group TransData_Remote Est Time Remaining N A Remote Snapshot May05RemSS Completion Time 5 27 05 2 16 PM Status Data Manual Scheduled MANUAL Data Copied 61 5 MB State COMPLETE Data Remaining N A Complete 100 Current Rate NA Avg Rate 2 596 Kbisec KS Figure 229 Remote Snapshot Details for a Completed Remote Copy During the remote copy process the Details window reports current data for the statistics When the copy is completed the statistics show summary data Figure 229 shows a completed remote copy Table 62 lists the values for the statistics reported in the Details window Table 62 Values for Remote Copy Details Window Statistic Values Source Info Section Primary Mgmt Group Name of the management group containing the primary volume and snapshot Primary Snapshot Name of the primary snapshot Remote Mgmt Group Name of the management group containing the remote snapshot Remote Snapshot Name of the remote snapshot Status ManuallScheduled Whether the snapshot was created using a snapshot schedule or manually State Started Copying Stalled Complete Current state of the copy process Complete 0 100 Percent of the copy process that is completed Time Start Time MM DD YY HH MM AM PM Date and time copy started Elapsed Time Xd Xh Xm Xs X a number and the days
270. ion 6 3 41 2605 0 Figure 18 Icon Showing RAID is Off s if RAID is degraded a yellow circle displays as shown in Figure 19 See Monitoring RAID Status on page 70 for information about fixing degraded RAID MPC DataFRAME 440 Storage System Software User Manual 31 Getting Started Details Management Group Information Availability HostName DF140 RAID Status Degraded RAID Configuration RAID 0 IP Address 192 168 0 7 Model DF 100 Series Serial Number 00 30 48 2E BA 5A Storage Mode block Software Version 6 3 41 2605 0 Figure 19 Icon Showing RAID is Degraded If RAID is rebuilding for RAID 5 or 50 or for RAID 10 a blue circle displays as shown in Figure 20 192 168 0 7 DF 100 Series Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Figure 20 Icon Showing RAID is Rebuilding Management Group Information Tab The Management Group Information tab provides detailed information about the management group to which the DataFRAME belongs Details Management Group Information Availability Management Group Transaction_Data Manager No Hot Spare No Virtual Manager No Migrating Data No Status Storage Server Normal RAID Normal Not a Manager Not a Virtual Manager Figure 21 Management Group Information Tab Availability Tab The Availability tab displays which volumes and snapshots availability This tab depends on this DataFRAME staying online Details i
271. ion Group The Edit Authentication Group window opens shown in Figure 184 MPC DataFRAME 440 Storage System Software User Manual 279 Controlling Client Access to Volumes Edit Authentication Group Edit Authentication Group Name Description Volume List Credit iSCSI EBSD 1 Fibre Channel Z Allow access via iSCSI Authentication CHAP not required Initiator Node Name _lign 1991 05 com microsoft kwegner wxp CHAP required Cancel Figure 184 Editing an Authentication Group 5 Select the appropriate tab for the group you are editing 6 Change the appropriate information 7 Click OK when you are finished Deleting an Authentication Group Deleting an authentication group will stop access to volumes by clients using that group Access to the same volume by other authentication groups continues 1 Log into the management group and select that management group in the Network View The management group Tab View opens 2 Click the Authentication Groups tab to bring it to the front 3 Select from the list the group you want to delete 4 From the Tasks menu select Delete Authentication Group A confirmation window opens 5 Click OK to delete the group 280 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Volume Lists Overview The volume list for an authentication group is the list of volumes accessible to that group
272. iple DataFRAMEs with the same configuration you can create a single configuration backup file and use it to restore the configuration on any of these DataFRAMEs Any DataFRAME that you restore from the backup file will have exactly the same configuration If you back up the configuration of a DataFRAME that has a static IP address and then restore that configuration to a second DataFRAME the second DataFRAME will have the same IP address You must manually change the IP address on the second DataFRAME 1 Log in to the DataFRAME 2 Click the Backup and Restore tab The Backup and Restore window opens shown in Figure 33 edit configuration for DF140 192 168 0 7 xj ka o d Power Off Feature Registration Boot Devices ar w bh Module Information Backup amp Restore D TCP IP Network Se DataFRAME Administ g SNMP e IS Reporting Back up the DataFRAME configuration files to this machine Restore the backup configuration to the DataFRAME if required Back Up Restore Log Out Figure 33 Viewing the Backup and Restore Window Up the DataFRAME Configuration File Use Backup to save the DataFRAME configuration file to a directory on your local machine 1 Click Backup The Save window opens shown in Figure 34 MPC DataFRAME 440 Storage System Software User Manual 45 Working with DataFRAMEs x Sach E DF140 Back up a e e sl File Name DataFRAME_Con
273. ique MAC media access control address Address IP address of the NIC Mask Network mask for NIC Gateway Gateway that the DataFRAME is using Mode Shows manual auto disabled Manual equals a static IP auto equals DHCP disabled means the interface is disabled DNS Data Information about DNS if a DNS server is being used Server 1 Server 2 IP address of the DNS servers Memory Information about RAM memory in the DataFRAME Total Total amount of memory in KB Free Total amount of free memory in KB CPU Information about the CPU Model Name Model name manufacturer and capable speed of the CPU Speed Clock speed of the microprocessor Load Average Information about the average load on the system Machine Uptime The total time the DataFRAME has been running from initial boot up Enclosure Firmware Version Firmware version number for the midplane Drive Temperature The temperature of the drive in centigrade Drive Status Drive Number Information about the drives in the DataFRAME MPC DataFRAME 440 Storage System Drive 1 through 16 RAID Rebuild Rate Statistics Unit Number Information about RAID RAID Rebuild Rate is a percentage of RAID card through put Information about the RAID for the DataFRAME Identifies disks that make up the RAID configuration their RAID level chunk size and device name Power Supplies Number 1 Number 2 Status information abou
274. ish restoring the configuration The DataFRAME reboots and the configuration is restored to the identical configuration as that in the backup file Completing the Restore After you restore the DataFRAME configuration from a file up to three manual configuration steps are required e You must manually configure RAID on the DataFRAME e You must manually add network routes after the restoration Restoring a DataFRAME configuration file from one DataFRAME to a second DataFRAME does not restore network routes that were configured on the DataFRAME e f you restore multiple DataFRAMEs from one configuration file you must manually change the IP address on the additional DataFRAMEs For example if you back up the configuration of a DataFRAME with a static IP address and then restore that configuration to a second DataFRAME the second DataFRAME will have the same IP address Rebooting the DataFRAME Reboot the DataFRAME from the Console without powering off Set the amount of time before the reboot begins to ensure that any activity to the module has stopped 1 Log in to the DataFRAME 2 Select the Power Off tab The Power Off window opens shown in Figure 37 48 MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs E Edit Configuration for DF140 192 168 0 7 i x Power Off Feature Registration Boot Devices Module Information f Backup amp Restore T Storage Time ga TO
275. ished Changing or Removing the Virtual IP Anytime you change or remove the virtual IP address for iSCSI volumes you are changing the configuration that clients are using Therefore it is important to disconnect any clients before making this change Preparing Clients e Quiesce any applications that are accessing volumes in the cluster e Log off the active sessions in the initiator for those volumes Changing or Removing the Virtual IP Address 1 In the Edit Cluster window click the iSCSI tab 204 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters 2 Change or delete the entries in the IP Address Subnet Mask and Default Gateway fields Finishing Up 1 Click OK when you are finished changing or removing the virtual IP 2 Reconfigure the iSCSI initiator with the changes 3 Reconnect to the volumes 4 Restart the applications that use the volumes Changing or Removing an iSNS Server If you change the IP address of an iSNS server or remove the server you may need to change the configuration that clients are using Therefore you may need to disconnect any clients before making this change Preparing Clients e Quiesce any applications that are accessing volumes in the cluster e Log off the active sessions in the initiator for those volumes Changing an iSNS Server 1 Select the iSNS server to change 2 Click Edit The Edit iSNS Server window opens 3 Change the IP address 4 Cli
276. ished adding hosts MPC DataFRAME 440 Storage System Software User Manual 137 Using SNMP Editing the Trap Recipient 1 Select the host you want to change from the list of Trap Recipients 2 Click Edit The Trap Recipient window opens 3 Change the host name or IP address 4 Click OK 5 Click Apply when you are finished editing trap recipients Removing the Trap Recipient 1 Select the host you want to remove from the list of Trap Recipients 2 Click Remove A confirmation window opens 3 Click OK to remove the trap recipient The host is removed from the list 4 Click Apply when you are finished removing trap recipients Disabling SNMP Traps To disable SNMP traps you must delete all of the settings in the SNMP Traps window 1 Remove the Trap Recipient hosts 2 Delete the Trap Community String 3 Click Apply 138 MPC DataFRAME 440 Storage System Software User Manual 8 Reporting The Reporting category includes multiple types of information and reporting capabilities Review a passive report of system statistics hardware and configuration information save log files set up email alerting and review alerts generated automatically by the operating system Reporting Overview Use reporting to e View real time statistical information about the DataFRAME e View and save log files e Set up active monitoring of selected variables e Set up email notification View alerts Run hardware diagnostics When y
277. issing Faulty CPU Utilization Percent No Yes 1 minute Console alert at gt 90 CPU Celsius Yes No 15 Warning at 65 Temperature seconds Console alert SSR316MJ2 Critical at 70 Console alert Shutdown Drive Health Normal Yes Yes 1 minute Console alert if not Status normal Marginal Faulty Drive Status On and No Yes 1 minute Console alert if secured Off changes and secured Off or removed MPC DataFRAME 440 Storage System Software User Manual 149 Reporting 150 Table 31 List of Variables Available for Active Monitoring i Perm Specify Default Default Action eee eae ee Variable Freq Freq Threshold Drive Celsius Yes No 1 minute Warning at 60 Temperature Console alert Critical at 65 Console alert Drive Power Off Fan Status Normal No Yes 1 minute Console alert if Faulty changes Fibre Channel Waiting for No Yes 1 minute Console alert if Status firmware changes Active Enabled Not ready Initialized Hot Spares No Yes 15 Console alert if a Activated minutes DataFRAME hot spare is activated Memory Percent No Yes 1 minute Console alert at Utilization gt 90 Motherboard Celsius Yes No 15 Warning at 65 Temperature seconds Console alert Critical at 70 Console alert Shutdown Network No Yes 1 minute Console alert if NIC Interface Status status changes NVRAM Status Yes Yes 1 minute Cons
278. it is correctly detected Find DataFRAMEs Wizard xj Search by DataFRAME IP or hostname mpc Enter IP address or hostname Wen Click Add to enter an IP address or host name Use Edit or Delete to refine your list IP or Host Name Status 192 168 0 40 Found Eat _ petete To find your DataFRAMEs on the network click Finish MPC DataFRAME 440 Storage System Software User Manual 7 Getting Started 20 Click on Close to return to the Wizard Launch Pad x Summary Found DataFRAMEs are listed below mpc DataFRAME search results Drun DF 440 1 IP 192 168 0 40 Found To find additional DataF RAMES e Repeat this wizard using different search criteria Search again using the Find menu on the DataFRAME Management Console Repeat this Wizard Run the Create Volume Wizard a You must set up your RAID configuration before using the DataFRAME Close out of the Wizard Launch Pad by clicking on OK Welcome to the Wizard Launch Pad The Varad Launch Pad provides wizards for common tasks to get you started Find Datel RAMLs Wizard The Find Oster RAMEs Wizerd finds Deter RAMES on your r toron Creste Volume Wizard e The Creste Volume Wizard crestes a new voume Access Volume Wizard The Access Volume Wizard crestes eccess prvieges for volume r C Show Launch Pad on startup This Wizard Launch Pad and the wizards are ara sbie vie
279. ity at some cost to system performance Managing Volume Growth Capacity Note When creating a volume for which you plan to use snapshots you can set the soft threshold value to help manage capacity growth This threshold value triggers an alert providing you the opportunity to increase the capacity of the volume before it is full Volume size replication level and snapshots should be planned in conjunction if you intend to use Snapshots review Chapter 13 Working with Snapshots Creating the Volume and Setting Thresholds MPC DataFRAME 440 Storage System Software User Manual e First create the volume and designate the size This size is the logical size on the cluster For example you have a 750 GB cluster and you create a 500 GB volume e Second set the hard threshold to some size smaller than the actual volume size For our example 500 GB volume you set the hard threshold at 300 GB 219 Working with Volumes e Third set the soft threshold lower than the hard threshold The soft threshold triggers an alert to the system administrator notifying that the soft threshold has been reached This alert gives you time to increase the volume size and hard threshold For our example set the soft threshold at 485 GB Warning If the hard threshold is set lower than the volume size and the hard threshold is reached then other applications that are accessing the volume will hang until you increase the hard threshold
280. k dewscsi host0 busOttargetS unOdisc Disk dewscsi host0 busOitarget6 lunOidisc Disk dewscsifhost0 busOitarget8 lunOidisc Disk dewscsi hostt busOitargetO unOidisc Disk MPC DataFRAME 440 Storage System Software User Manual 9 Getting Started 24 Enter the RAID level and rebuild rate See Storage on page 57 for more information on RAID levels 25 Click on Configure RAID to set up the RAID level Beat Configuration for DF440 1 192 168 0 40 Module RAID Setup Disk Setup itp Borage RAID Setup Disk Setup Time or TCP IP Network Sn DataFRAME Administ fe SNMP Reporting RAID Status Start RAID Reconfigure RAID RAID Configuration RAID 0 7 The RAID Rebuild Rate is a percentage of RAID card throughput Percent d DR ah s0 90 100 Amy 40 70 80 Device Type j Subdevices Disk 0 Disk Disk Disk Disk Disk Disk Disk Disk boti 50 60 Device Name dewscsifhost0 busOitargetO unOidise dewscsi host0 busOitargett lunO disc dewscsi host0 busOitarget2 unOidisc dewscsi host0 busOitarget3 lunO disc dewscsihost busOtarget4 unOidisc dewscsi host0 busOitargeti lunO disc dewscsi host0 busOitarget6 unO disc dewscsifhost0 busOitarget8 lunOidisc dewscsi hostt busOitargetO unOidisc Click on Done 26 The RAID Status should indicate normal as illustrated in the following figure Edit configuration for DF440 1
281. l However this replication is configured independently of the volume replication configured on the primary volume If one of the DataFRAMEs in the primary location fails the primary volume will still be available If all of the DataFRAMEs fail or if the application server fails then you can recover data from the remote snapshots or tape backups in the off site location MPC DataFRAME 440 Storage System Software User Manual 375 Sample Remote Copy Configurations Using Remote Copy for Off site Backup and Recovery For backup and recovery systems Remote Copy can eliminate the backup window on an application server Using scripting configure the EBSD driver to mount remote snapshots on a backup server either local or remote and then back up the remote snapshot from the backup server The remote snapshot is available if the primary volume fails Achieving Off site Tape Backup Rather than creating tape backups and then transporting them to a secure off site location you can use Remote Copy to create remote snapshots in an off site location and then create tape backups at the off site location Configuration for Off site Backup and Recovery To use remote snapshots for off site tape backup create remote snapshots for access by your tape backup application e Create remote volumes in your backup location e Configure your backup application to access the remote snapshots e Configure schedules to create remote snapshots in the de
282. l administrator Add Remove Password Confirm Password 5 40 characters Figure 90 Editing an Administrative User 5 Change the necessary information 6 Click OK Deleting Administrative Users 1 Select DataFRAME Administration from the configuration categories 2 Click the Users tab to bring it to the front 3 Select the user to delete from the list of users 4 Click Delete A confirmation message opens 5 Click OK Note If you delete an administrative user that user is automatically removed from any administrative groups MPC DataFRAME 440 Storage System Software User Manual 129 Administrative Users and Groups 130 MPC DataFRAME 440 Storage System Software User Manual D Using SNMP The DataFRAME can be monitored using an SNMP Agent You can also enable SNMP traps The DataFRAME Management Information Base MIB is read only and supports SNMP versions 1 and 2c See Installing the Storage System MIB on page 135 for a list of Storage System MIBs Getting There 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select SNMP from the configuration categories The SNMP General tab opens shown in Figure 91 Gear Configuration for DF140 192 168 0 7 es Module W Storage Time Disable SNMP Agent se TCP IP Network a DataFRAME Administ O Enable SNMP Agent ei Repo
283. l window select Disable SNMP Agent 2 Click Apply Enabling and Disabling SNMP Traps Enable SNMP Traps if you plan to use an SNMP tool to notify you when a monitoring threshold is reached 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select SNMP from the configuration categories The SNMP General window opens shown in Figure 91 on page 131 3 Select the SNMP Traps tab The SNMP Traps window opens shown in Figure 96 136 MPC DataFRAME 440 Storage System Software User Manual Using SNMP ER Configuration for DF140 192 168 0 7 es Module 30 Storage Time SNMP General SNMP Traps Trap Community String d D TCP IP Network DataFRAME Administ E ES Trap Recipients IP Host name eg ki Reporting Figure 96 Enabling SNMP Traps Enabling SNMP Traps 1 Enter the Trap Community String This is required if you want to use SNMP traps Note The Trap Community String is used for client side authentication 2 Click Add in the Trap Recipients area to add specific trap recipients The Trap Recipient window opens Add IP or Host Name xj IP or Host Name Figure 97 Adding an SNMP Trap Recipient 3 Enter the host name or IP address for the SNMP client that is receiving the traps 4 Click OK 5 Repeat steps 2 through 4 for each host in the trap community 6 Click Apply when you are fin
284. lability See Planning Data Replication on page 214 for details about volume replication Note A hot spare cannot reside in a cluster which contains volumes that have a replication priority of redundancy Adding a Hot Spare To add a hot spare the cluster must contain sufficient DataFRAMEs to handle the volumes and snapshots that currently exist in that cluster See Adding a DataFRAME to an Existing Cluster on page 202 if you want to add a DataFRAME to become a hot spare 1 Click the Hot Spare drop down list and select the DataFRAME to designate as the hot spare 2 Click OK when you are finished Note A cluster must contain at least three DataFRAMEs to have one DataFRAME designated as a hot spare Removing a Hot Spare To remove a hot spare simply change the designation in the list to none The hot spare then becomes a DataFRAME in the cluster adding more space for storage 1 Click the Hot Spare drop down list and select None from the list 2 Click OK when you are finished The DataFRAME returns to the cluster as available storage Changing the Hot Spare Time Out The hot spare time out designates the amount of time before a hot spare is activated in the cluster You can change the value at any time The results of changing the time out value are listed below e Cluster operating normally changing hot spare time out has an effect only if a DataFRAME in the cluster becomes unavailable MPC DataFRAME 440 Stor
285. le 3 Run etc init d ebsd restart or service ebsd restart MPC DataFRAME 440 Storage System Software User Manual 317 Using the EBSD Driver for Linux 4 Cat proc ebsd client to verify driver version and devices online Installing the EBSD Driver with RPM Packages The RPM packages are located in the appropriate RPM directory on the Resource Driver CD Locate the appropriate RPM for your system Prerequisites for using RPM packages To use the RPM source kit you need either of the following RPMs e kernel source e gcc e glibcheader e glibckernelheader e rpmbuild e automake e glibc devel e libgcc or include the development and kernel hacking groups when building your system RPM Package Naming Convention RPM packages are named using the following convention ebsd Release Version Build Number Kernel Version 1386 rpm For example RPM name ebsd 6 x xxX xxxx 2 4 9 e 3enterprise i386 rpm e ebsd represents the driver name e 6 x xx represents the release version e xxxx represents the build number e 2 4 9 e 3enterprise represents the kernel version Installing A Binary RPM Package To install a binary RPM package on a system choose the appropriate RPM for the kernel version 1 Run uname r to determine what version of Linux you are running 318 MPC DataFRAME 440 Storage System Software User Manual Using the EBSD Driver for Linux 2 Run rpm ivh to install the RPM For example on Red Hat Adv
286. le 36 Progression of Increments in Manual Auto Grow Setting of 50 MB Soft Threshold MB Hard Threshold MB 95 100 145 160 195 210 245 260 995 1010 10241 1024 1 When the hard threshold reaches the volume length the soft threshold is increased to the same value to disable alerts 2 The hard threshold cannot exceed the volume length Manual Auto Grow N m Soft Threshold j Hi m Hard Threshold bi ee Ott des KE III 123 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Auto Grow Increments Figure 150 Example Manual Auto Grow Chart MPC DataFRAME 440 Storage System Software User Manual 223 Working with Volumes Automatic Auto Grow When auto grow is set for Auto the soft and hard thresholds increase automatically according to a preset algorithm The rate of increase begins small and increases to ever greater increments as illustrated belowin Table 37 and Figure 151 Table 37 Progression of Increments in Automatic Auto Grow Soft Threshold MB Hard Threshold MB 48 644 64 80 72 96 96 128 144 192 192 256 384 512 1024 1024 3 The ratio of the soft threshold to the hard threshold 3 4 in this example is maintained throughout the auto grow process 4 The first automatic auto grow is always to the nearest pre computed increment that is greater than the initial setting Automatic Auto Grow en Trams ee
287. le Name ki 20 0001_to_6 1 21 0001_u1 20041018 01 upgrade tar gz upgrade Files of Type AI Files e Open Install File Cancel Figure 31 Browsing for Upgrade or Patch File 6 Select the file and click Open Install File Focus returns to the Install Software window When the file name is present the Install button becomes enabled 7 Click Install The install status window opens shown in Figure 32 Status messages scroll on the window These messages can be saved to a file Optional Click Save To File and choose a name and location for the file MPC DataFRAME 440 Storage System Software User Manual 43 Working with DataFRAMEs Golden 110 0 33 17 Gathering the install destination information started Gathering the install destination information finished Opening the install conection started Opening the install conection finished Sending the install file started Sending the install file 0 complete updated every 10 seconds Sending the install file 100 complete updated every 10 seconds Sending the install file finished extracting returnd starting return return 2 sleeping 5 seconds finished sleeping ost qualification test V Automatically scroll Save to File Close Figure 32 Upgrade Status Messages After the installation completes the system reboots After the system comes back online it conducts a post i
288. lover for iSCSI initiators that do not support multiple IP addresses per target Io do not want to use a virtual IP address for this cluster ISNS Servers IP Add Edit Delete Cancel Figure 137 Configuring a Virtual IP Address for iSCSI 2 Add the IP address subnet mask and default gateway if required Adding an iSNS Server Optional Note If you use an iSNS server you may not need to add Target Portals in the Microsoft iSCSI Initiator 1 Click Add The Add iSNS Server window opens shown in Figure 138 Add iSNS Server x ISNS Server Figure 138 Adding an iSNS Server MPC DataFRAME 440 Storage System Software User Manual 197 Working with Clusters 2 Type the IP address of the iSNS server 3 Click OK The server is added to the list shown in Figure 139 xi New Cluster General iscsi r Virtual IP IP Address 192 168 0 10 A cluster of multiple DataFRAMEs requires a virtual IP to configure Subnet Mask 255 255 255 0 failover for iSCSI initiators that do not support multiple IP addresses per Default Gateway 192 168 0 2 target C I do not want to use a virtual IP address for this cluster iSNS Servers IP 192 168 0 List of iSNS servers Edit Delete Cancel Figure 139 List of iSNS Servers 4 Click OK when you have finished The cluster is created and displayed inside the management group shown in Figu
289. lume Parameters on page 229 for detailed requirements Changing the Size 1 In the size field change the number and change the units if necessary 2 Click OK when you are finished See Requirements for Changing Volume Parameters on page 229 for detailed requirements Warning Decreasing the volume size or hard threshold is not recommended without careful planning Changing the Hard Threshold 3 In the hard threshold field change the number and change the units if necessary 4 Click OK when you are finished See Requirements for Changing Volume Parameters on page 229 for detailed requirements Changing the Soft Threshold 1 In the soft threshold field change the number and change the units if necessary 2 Click OK when you are finished See Requirements for Changing Volume Parameters on page 229 for detailed requirements 232 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes Fixing a Replica challenged Redundant Volume If a DataFRAME goes offline and needs to be repaired or replaced anda replicated volume configured for redundancy becomes unavailable to clients the following procedure allows you to safely return the volume to fully operational status 1 Stop any clients from accessing the volume Select the volume in the Console Right click and select Edit Volume Change the data priority from data redundancy to data availability Remove the DataFRA
290. lumes Table 1 DataFRAME Configuration Tasks Complete This Task Find Detailed Information In Search for one or more DataFRAMEs on the network Finding Storage System Modules on the Network on page 19 Log in to the DataFRAMEs you want to work with DataFRAME Configuration Window on page 34 Configure individual DataFRAMEs DataFRAME Configuration Window on page 34 Create one or more management groups Chapter 9 Working with Management Groups Create one or more clusters Chapter 11 Working with Clusters Create one or more volumes Chapter 12 Working with Volumes Configure access to volumes Chapter 15 Controlling Client Access to Volumes 14 MPC DataFRAME 440 Storage System Software User Manual Getting Started Wizards The first time you open the Console the Wizard Launch Pad opens shown in Figure 1 izard Launch Pad EI Welcome to the Wizard Launch Pad The Vizard Launch Pad provides wizards for common tasks to get you started Find DataFRAMEs Wizard The Find DataFRAMEs Vizard finds DataFRAMEs on your network Create Volume Wizard The Create Volume Wizard creates a new volume Access Volume Wizard The Access Volume Vizard creates access privileges for a volume P amp C Show Launch Pad on startup This Wizard Launch Pad and the wizards are available via the Help menu Figure 1 Viewing the Wizar
291. ly_Administrator View only capability to all functions read only Adding Administrative Groups Administrative groups are listed on the DataFRAME Administration window on the Groups tab shown in Figure 81 Adding a Group 1 Select DataFRAME Administration from the configuration categories 2 Click Add on the Groups tab The Create Administrative Group window opens shown in Figure 81 120 MPC DataFRAME 440 Storage System Software User Manual Administrative Users and Groups Create Administrative Group General Group Permission Group Name 3 40 characters 0 9 a z _ Must begin with a lower case letter Description 3 40 characters 0 9 a z A Z _ space Must not begin with a space Users Add Ces Figure 81 Adding an Administrative Group 3 Type a Group Name and Description Both are required Table 27 Administrative Group Name Requirements Group Name Requirements Example e 3 to 40 characters e Software_Admins start with a letter e Region11_Managers e Use letters a z A Z numbers 0 9 or characters _ Adding a User to the Group 1 Click Add in the Users section The Add Users window opens with a list of administrative users shown in Figure 82 MPC DataFRAME 440 Storage System Software User Manual 121 Administrative Users and Groups Add Administrative Users to Group x Figure 82 Adding an
292. m eeeeeeeceeeeeeseeeeeeeeneeeeeees 74 Diagram of the Drive Bays in the MPC DataFRAME 440 Storage System ceeeseeeeseeeeeeeeteeeneeeeeens 76 Viewing the Network Configuration ccecceseceeeeeeeeeeeeeeeeeeeeeeaeeeeeeseaeseaeeseeeesaeeseaeeeeesiaeesneeesieeseaeseneetaas 79 Network Interface Ports and Open PCI Slots on the Back of the MPC DataFRAME 440 Storage System Figure 56 Distributing Bandwidth and Ensuring Fault Tolerance of Add on Ports Across PCI slots eeeeee 82 Figure 57 Viewing the WWN of a Fibre Channel Pont 83 Figure 58 Configuring the IP Address Manually cecceesceceeeeseeeeeeeeeneeeeaeesaeeseaeeeseescaeeseeeeaeeseeeeeaeeeeesnaeeeaeeneaten 83 Figure 59 Active Backup in a Two switch Topology with Server Failover ccceecceeseeeeeeeeeeeeeneeteeeeeeeeenaeeeaeeseaeens 89 Figure 60 Active Backup Failover in a Four Switch Topology 90 Figure 61 NIC Aggregation in a Partial mesh Topology with Server Failover ceccceesceseeeeeeeeeeeeeeeetneeeeeereeen 92 Figure 62 NIC Aggregation in a Single switch Topology cecccesceeseeeeneeceeeeeeeceaeeeaeeeeeeeaeeseeeseeeseaeeeieeenaeeeaeeseaeens 93 Figure 63 Selecting Motherboard PortO and Slot1 PortO for a New Bond 94 Figure 64 Creating a NIC Bond 95 Figure 65 Viewing a New Active Backup Bond 96 Figure 66 Viewing the Status of an Active Backup Bond 96 Figure 67 Viewing the Status of a NIC Aggregation Bond 97 Fig
293. manager icon appears in the management group as shown in Figure 132 MPC DataFRAME 440 Storage System Software User Manual 187 Disaster Recovery Using A Virtual Manager mec Computers DataFRAME Management Console lei E File Find Tasks View Help a Cluster _0 Virtual manager added Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Management Group Group_o managers 2of3 Local Bandwidth 32768 Kb sec Virtual Manager Yes Quorum 2 Status Normal Coordinating manager DF420 Figure 132 Management Group with Virtual Manager Added The virtual manager remains added to the management group until needed Starting a Virtual Manager to Regain Quorum 188 Only start a virtual manager when it is needed to regain quorum in a management group Figure 128 illustrates the correct way to start a virtual manager when necessary to regain quorum e Two site Scenario One Site Goes Down For example in the two site disaster recovery model one of the sites goes down On the site that is still up all managers must be running Select one of the DataFRAMEs at that site and start the virtual manager on it That site then regains quorum and can continue to operate until the other site is recovered Once the other site is recovered the managers in both sites reestablish communication and they ensure that the data in both sites is resynchronized When the data is re
294. mary and the backup application server becomes the production application server accessing data from the acting primary volume 368 MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations Application Workstations Clustered Production St Backup Production Application Application Application Application Servers Server Serer Server Read Only Read wWrite values eco Primary Remote Primary Snapshot Snapshot Snapshot Storage Subnet X Primary Remote LU A Figure 242 High Availability Configuration During Failover Data availability if the primary volume or production application server fails If either the primary volume or production application server in your production site fails only that data written to the volume since the last remote snapshot was created will be unavailable until the volume or production application server is restored Failover to the backup application server To maintain availability of the application and the remaining data the following process occurs 1 A script or other application monitoring the production application server discovers that primary volume is not available A script executes to fail over to the backup application server 2 The backup application server executes a script to convert the remote volume into a primary volume so that the volume can be accessed by the backup application server 3 Because the backup application server was
295. me Into a Primary Volume You can make a remote volume into a primary volume Changing the remote volume into a primary volume allows the backup application server to read and write to the volume This is useful in failover recovery if you want to use the failover site as the acting primary site Note You cannot make a remote volume into a primary volume while a remote snapshot is in progress Wait until the remote snapshot copy is complete before making the remote volume into a primary volume 358 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Designating Size and Threshold Values for the Converted Volume If the remote volume was originally created as a remote volume e you will need to designate a volume size and hard and soft thresholds If the remote volume was originally created as a primary volume that was then changed to remote e returning that volume to its primary state will automatically return the original size and threshold values You can change these values before completing the conversion 1 Log in to the management group containing the remote volume that you want to convert 2 Right click the volume in the network view and select Edit Volume The Edit Remote Volume window opens 3 Change the Type from Remote to Primary Notice that the window changes to the Edit Volume window and all the fields are greyed out as shown in Figure 237 Edit Remote Volume Edit Remote Volume Type
296. me in the DNS Suffixes list 1 Select the domain name to edit 2 Click Edit The Edit DNS Suffix window opens 3 Enter the change to the domain name 4 Click OK 5 Click Apply when you are finished Removing a DNS Server Remove a DNS server from the list 1 Select the server you want to remove from the DNS Servers list 2 Click Remove A confirmation message opens 3 Click OK to remove the DNS server from the list 4 Click Apply when you are finished MPC DataFRAME 440 Storage System Software User Manual 105 Managing the Network Removing a Domain Name from the DNS Suffixes List 1 Select the domain name you want to remove from the DNS Suffixes list 2 Click Remove A confirmation message opens 3 Click OK to remove the domain name from the list 4 Click Apply when you are finished Routing Overview The Routing tab displays the routing table You can specify static routes and or a default route If you specify a default route here it will not survive a reboot or shut down of the DataFRAME To create a route that will survive a DataFRAME reboot or shut down you must enter a default gateway on the TCP IP tab See Configuring the IP Address Manually on page 83 Information for each route listed includes the device the network gateway and mask and flags Adding Routing Information 1 On the Network View double click the DataFRAME and log in if necessary 2 The DataFRAME Configuration window opens S
297. me manager tool ebsdtool environment variable If the configuration file and the EBSD volume manager tool exist then the EBSD service e Loads the EBSD driver if needed e Starts all the devices listed in the EBSD configuration file Sebsdconf e Waits until all the devices become online or exceed the timeout value e Mount all the EBSD devices listed in etc fstab and not marked as noauto Stopping the EBSD Driver The EBSD stop service first checks for the EBSD configuration file ebsdconf environment variable and the EBSD volume manager tool ebsdtool environment variable If the configuration file and the EBSD volume manager tool exist then the EBSD service e Reports a system hang warning if a mounted device has lost connection e Stops all processes using mounted EBSD devices e Unmounts all mounted EBSD devices e Stops and removes all the EBSD devices e Unloads the EBSD driver Status of the EBSD Driver and Devices To display the status of the EBSD driver and the associated devices enter the following service ebsd status or MPC DataFRAME 440 Storage System Software User Manual 325 Using the EBSD Driver for Linux etc init d ebsd status This status command displays the information in the file proc ebsd client Disconnecting an EBSD Device Warning Disconnecting an EBSD device stops activity on the DataFRAME but preserves the data in the volume Prerequisite Stop all applications using the EB
298. ministration window on the Users tab along with their group membership and a description ER Configuration for DF140 192 168 0 7 ss Module 3 Storage Time Lie Toran Network User Member of Groups Description ES DataFRAME Adminis admin full_administrator Administrator Groups Users KH SNMP Reporting Figure 86 Adding Administrative Users 126 MPC DataFRAME 440 Storage System Software User Manual Administrative Users and Groups Adding an Administrative User 1 Select DataFRAME Administration from the configuration categories 2 Click the Users tab to bring it to the front shown in Figure 86 3 Click Add The Create Administrative User window opens shown in Figure 87 Create Administrative User EI User Name 3 40 characters 0 9 a z A Z _ space Must begin with a letter Description 3 40 characters 0 9 a z A Z _ space Must begin with a letter Member Groups Add Remove Password Confirm Password 5 40 characters Cancel Figure 87 Adding an Administrative User 4 Type a User Name and Description 5 Type a password and confirm that password Adding a Member Group 1 Click Add in the Member Groups section The Add Administration Groups window opens shown in Figure 88 MPC DataFRAME 440 Storage System Software User Manual 127 Administrative Users and Gro
299. mirroring the data from one disk onto a second disk RAID 10 RAID 10 combines mirroring data within pairs of disks with striping data across pairs RAID 10 combines data redundancy with the performance boost of RAID 0 Configuring RAID 1 or RAID 10 Whether the DataFRAME is configured in RAID 1 or RAID 10 depends on the number of disks in the module e If the DataFRAME contains only 2 disks configured for RAID then the mirrored disk pair is RAID 1 e Ifthe DataFRAME contains 4 or more disks configured for RAID then the 2 or more mirrored disk pairs are RAID 10 Storage Capacity in RAID 10 DataFRAME capacity in RAID 10 is the total capacity of all mirrored disk pairs in the module The capacity of a single disk pair is equal to the capacity of one of the disks as shown in Figure 43 ii aiid ad Bee Cen a capacity capacity capacity 250 GB 250 GB 250 GB Figure 43 Capacity of Disk Pairs in RAID 10 60 MPC DataFRAME 440 Storage System Software User Manual Storage RAID 5 and RAID 50 RAID 5 RAID 5 provides data redundancy by distributing data blocks across disks in a RAID set Redundant information is stored as parity distributed across the disks Figure 44 shows the distribution of parity across 4 disks in a RAID 5 set 0 P 0 0 0 parity 1 1 1 parity 1 2 2 parity 2 2 3 parity 3 3 3 Disk 1 Disk 2 Disk 3 Disk 4 Figure 44 Parity Distributed
300. mirrors are usually created for one time use and then discarded Creating a Split Mirror To create a split mirror e Create a remote snapshot e Create a volume list for that snapshot e Create an authentication group for client access e Configure client to access the remote snapshot MPC DataFRAME 440 Storage System Software User Manual 365 Remote Copy Disassociate Remote Management Groups Management groups become associated when linked by remote snapshots or remote snapshot schedules When you have management groups that no longer share remote snapshots or remote snapshot schedules you can disassociate those management groups Disassociating management groups destroys all the shared knowledge between those groups 1 Log in to both management groups that you want to disassociate 2 Right click the remote management group and select Edit Management Group The Edit Management Groups window opens shown in Figure 240 Edit Management Group Edit Management Group Group Name Local Bandwidth 32768 Kb Sec Remote Management Groups Management Gr Logged In Remote Bandwi Connected Transaction_D Yes 32768 Kbisec Yes Disassociate Edit Bandwidth Cancel Figure 240 Editing a Management Group 3 Select the management group or groups you want to disassociate 4 Click Disassociate A confirmation message opens describing the results of disassociating the management groups
301. mote Copy Table 63 Steps to Create Snapshots Volumes and Volumes and Snapshots on Snapshots on Action Activity Primary Remote if ree e Management Management p Group Group 1 Stop applications that are accessing the volumes 2 Make a snapshot of the OrigPrimaryVol_0 Creates a snapshot of original volume the original primary volume that includes the data from 1 00 1 30 OrigPrimarySS_0 p m 3 Make the acting primary RemoteVol_0 Returns the remote volume into the remote management group to volume This ActPrimarySS_0 automatically creates a snapshot of the acting primary volume its original configuration 2 Synchronize the Data Synchronize the snapshots OrigPrimarySS_0O and ActPrimarySS_0 created in Steps 2 and 3 of Table 63 as appropriate for the application Creating a New Primary Volume at the Original Production 362 Site If the original primary volume is not available designate a new primary volume synchronize the data from the acting primary volume and configure the remote snapshot schedule on the new primary volume 1 Stop the application that is accessing the acting primary volume 2 Create a remote snapshot of the acting primary volume and make a new primary volume on the original production site as part of creating that remote snapshot 3 Convert the remote volume into a primary volume 4 Make the acting primary volume into the remote volume This crea
302. mote snapshots that you want to retain You can retain primary snapshots to facilitate easy rollback of the primary volume Retention of snapshots will affect the amount of space that is used in the cluster of DataFRAMEs so balance the number of snapshots to retain with the amount of space you are willing to use To roll back to a snapshot that you did not retain you can still access remote snapshots or tape backups e Retain remote snapshots in the backup location to facilitate fast recovery of backed up data If you retain a number of remote snapshots after a tape backup is created you can access this data without going to the backup tape Example configuration e Retain 3 primary snapshots This enables you to roll the primary volume back yet it requires a relatively small amount of space on the primary cluster MPC DataFRAME 440 Storage System Software User Manual 377 Sample Remote Copy Configurations e Retain up to a week s worth of remote snapshots on the backup cluster e For snapshots older than 1 week go to the backup tape Achieving Non Destructive Rollback As discussed in Rolling Back a Primary Volume on page 363 rolling a snapshot back to a volume deletes any snapshots that were created since the snapshot that you roll back to For example suppose you created snapshots of a volume on Monday Tuesday and Wednesday On Thursday if you roll the volume back to Monday s snapshot then the snapshots from Tuesday and W
303. moving data and synchronizing hot spare DataFRAMEs The default rate is 32768 KB 4 MB per second You cannot set the range below 2048 KB 256 KB Local Bandwidth Settings The bandwidth setting is in KB kilobytes per second The industry standard for networking bandwidth is in bits per second bps Use the following table to convert megabits to kilobytes for setting the local bandwidth Table 32 Typical Network Types 168 Network Type Bits Per Second bps Kilobytes Per Second KB Fractional T1 256 Kbps 32 Fractional T1 1 2 768 Kbps 96 Ti 1 544 Mbps 197 Bonded T1 2 3 088 Mbps 395 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Table 32 Typical Network Types Network Type Bits Per Second bps Kilobytes Per Second KB Bonded T1 4 6 176 Mbps 790 Ethernet 10Base T 10 Mbps 1280 T3 44 736 Mbps 5726 Ethernet 100Base T 100 Mbps 12 800 OC 3 155 Mbps 19 840 OC 12 622 Mbps 79 616 Ethernet 1000Base T 1 Gbps 128 000 OC 192 10 Gbps 1 280 000 Set or Change Local Bandwidth 1 Log in to the management group 2 Click Edit Management Group The Edit Management Group window opens shown in Figure 120 Edit Management Group Edit Management Group Group Name Local Bandwidth 32768 KbiSec Remote Management Groups Management Gr Logged In Remote Bandwi Connected Disassociate Edit B
304. n On but not participating in RAID Status column Uninitialized other columns with information Not on Status column Off or Missing other columns with dashed DMA Off disk unavailable due to faulty hardware or improperly seated Verifying Disk Status Check the Disk Setup window to verify that all the disks in the DataFRAME are active and participating in RAID 74 MPC DataFRAME 440 Storage System Software User Manual Storage Replacing a Disk Prerequisite Always power the drive off before removing and replacing a disk See Powering Drives Off on page 78 Replacing Disks in RAID 0 If you lose a disk in RAID 0 you will lose all of the data on the DataFRAME and you will have to rebuild the DataFRAME to recover any replicated volumes within a Cluster If RAID 0 is configured but data redundancy is achieved by replication of data within a cluster of DataFRAMEs then data is not lost See Repairing a DataFRAME on page 206 In order to make the DataFRAME functional again you must replace the disk and reconfigure RAID 0 1 Remove the DataFRAME from the management group 2 See Removing a DataFRAME from a Management Group on page 174 3 Power off the drive See Powering Drives On or Off on page 78 4 Replace the disk in the DataFRAME 5 Power on the drive 6 Reconfigure RAID 0 Replacing Disks in RAID 1 10 or RAID 5 50 To replace a disk in a DataFRAME running RAID 1 10 or RAID 5 50
305. n Backing Up Multiple DataFRAMEs with the Same Configuration cceeceeceeseeeeeeeeneeeeeeeneeenaeeeneeteaees 45 Backing Up the DataFRAME Configuration File ccceeceeeseeeeeeeeeeeeeeeceaeeeaeeesaeeeaeeseeeesaeseaeeeneeenaeeeneeenaees 45 Restoring the DataFRAME Configuration from a File ceecseceeeeeeeeeeceneeeeeeeeneeseeeeeaeesaeseaeesaeeeeaeeseaeeeaeee 47 Rebooting the Data FRAME ssias2cc ccsiectavetesitedubacsecussecdagscbetasacer aeto oaa ORAA EEA EET NSE AEA Ea EPEE AEEA OEE ESLa EEE 48 Canceling a Reboot iioii sonir tiounen itinn aana en niiae eiaa isana se ieor aaa eie saa iEn 50 Powering Off the DataF RAME AE 50 MPC DataFRAME 440 Storage System Software User Manual Vv Contents Registering Features for a DataFRAME AAA Using the Feature Registration Tab AA Evaluating R6atures session dhei dace Ratna tied ered alee E Ae R Configuring Boot Devices 0 0 0 ee eeceeeeeteeeeeeeeneeeeaeeeeeeteaeesaeeeaaeeseeeecaeeseaeseaeeeaeeesaeeeaescaeeseeeseeeesaeeseeeseaeesieeeneeesaeees Modules with Two Boot DeViCeS ccceseccecseeeeseeeeeseeeeeeseeeeeseeeeensaeeseeeeesneneesneeeesnaeseseeeneneaeeeneeeeeneeness Checking Boot Device Status in a DataFRAME Replacing a Boot Device Only in Modules with Two Boot Devices DataFRAME 440 Storage System 54 Ch hapt r TT E 57 Storage Overview EE 57 Storage Requirement AAA 57 Getting RTE 57 Configuring and Managing RAID ccceeeceeeceseneeeeeeeeeeeseeeeeaeeea
306. n Figure 212 General lt Remove from management group gt Re ry if Figure 212 Removing the DataFRAME from a Management Group 2 Tab to Remove from management group and press Enter A window opens warning you that removing the DataFRAMEfrom the management group will delete all data on the DataFRAME and reboot the DataFRAME 3 Tab to Ok and press Enter 4 On the Configuration Management window tab to Done and press Enter Resetting the DataFRAME to Factory Defaults Resetting the DataFRAME to factory defaults deletes all data and erases the configuration of the DataFRAME including administrative users and network settings Warning Resetting the DataFRAME to factory defaults deletes all data on the DataFRAME 1 On the Configuration Interface main menu tab to Config Management and press Enter The Configuration Management window opens shown in Figure 213 Figure 213 Resetting to Factory Defaults 308 MPC DataFRAME 440 Storage System Software User Manual Using the Configuration Interface 2 Tab to Reset to factory defaults and press Enter A window opens warning you that resetting the DataFRAME configuration will delete all data on the DataFRAME and reboot the DataFRAME 3 Tab to Ok and press Enter 4 On the Configuration Management window tab to Done and press Enter 5 Use the default User Name admin and Password storage to log in to the DataFRAME MPC DataFRAME 440 Sto
307. n create a management group and a cluster of 3 DataFRAMEs e Create volumes on the cluster and set the replication level to 2 e Configure the production application server to access the primary volume See the EBSD User Manual for instructions about configuring EBSD clients e Create a second management group and cluster of 3 DataFRAMEs in the backup location e Create a schedule for making remote snapshots of the primary volume See Scheduling Remote Snapshots on page 351 Note Volume replication levels are set independently for primary and remote volumes How It Works If one of the DataFRAMEs in the primary location fails the primary volume will still be available If all of the DataFRAMEs fail or if the application server fails then failover to the backup application server occurs and the remote snapshot becomes available Application Workstations Clustered Production Get Backup Production Application X Application Application Application e Servers Server Server Server Read Only Read Write Primary Remote Volume Volume Primary Make Primary Volume Primary Remote Primary Snapshot Snapshot Snapshot Figure 244 High Availability During Failover Example Configuration MPC DataFRAME 440 Storage System Software User Manual 371 Sample Remote Copy Configurations Achieving Affordable Disaster Recovery Even if you do not have clustered application servers or network bandwidth required for conf
308. n Figure 157 EE MPC computers DataFRAME Management Console 5 x File Find Tasks View Help Available DataFRAMEs EIS _ a Cluster_0 d DF 420 1 Volume H Remote Snapshot Schedules Target information Volume List Memberships Details Snapshots i Snapshot Schedules Hosts Remote Snapshot Name Volume 0 Size 500 GB Replication Level None Created 11 23 2005 04 10 54 AM GMT Hard Threshold 500 GB Replication Priority Availability Cluster Cluster_0 Soft Threshold 500 GB Type Primary Checksum Data No Auto Grow Off Created by Script No Status Normal Description None Figure 157 Volume Tab View 3 Click the Snapshots tab to bring it to the front 4 From the Tasks menu select New Snapshot The New Snapshot window opens shown in Figure 158 240 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots New Snapshot New Snapshot Snapshot Name Description Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Snapshot_0 Figure 158 Creating a New Snapshot 5 Type a name for the snapshot Names are case sensitive They cannot be changed after the snapshot is created 6 Optional Type in a description of the snapshot 7 Optional Change the hard and soft thresholds for the snapshot Note The hard threshold of the snaps
309. n that group if you purchase the Scalability Pak upgrade To prevent a new DataFRAME from being included in this pool of storage you would put itin a separate management group To contain clustering managers Within a management group one or more of the DataFRAMEs will act as the managers that control data transfer and replication This chapter discusses Managers Quorum Setting the management group time Setting the local bandwidth Backing up the management group configuration Requirements for Creating Management Groups When creating a management group you must configure the following parameters Working with Management Groups Management Group What it means Requirement Configure Before you create a management group you should configure all the DataFRAMEs DataFRAMEs for that management group When planning your storage remember that all DataFRAMEs in a cluster must be configured alike See Configuration Tasks on page 14 Log in to You must be logged in to the DataFRAME to create a management group DataFRAMEs Starting a manager A management group must have at least one manager running So when you create a new management group the first DataFRAME added to the group has the manager started automatically You can add managers to other DataFRAMEs later Assigning manager The DataFRAMEs that are running managers must have static IP IP addresses addresses or reserved IP addresses if using DHCP
310. nagement group information and features such as viewing management group properties registering features and creating clusters and authentication groups Details Tab Details about the management group are listed along with the Task menu for taking action on the management group Information provided on the Details tab includes e The name of the management group e The number of managers operating within the group e How many of those managers are required for a quorum and are operational see Managers on page 158 for information about quorums e The synchronization bandwidth that is set when you edit the management group e The communication mode of the management group Managers Tab All the DataFRAMEs included in the management group are listed on the managers tab DataFRAMEs that are running managers display Yes in the Manager column Information provided on the Managers tab includes e Host name of the DataFRAME e Whether that DataFRAME is running a manager e Whether that DataFRAME is running a virtual manager Clusters Tab All the clusters created within the management group are listed on the clusters tab For more information see Chapter 11 Working with Clusters Information provided on the Clusters tab includes e Name of the cluster 166 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Authentication Groups Tab All the authentication groups that are associated wi
311. nager to Regain Quorum 184 MPC DataFRAME 440 Storage System Software User Manual Disaster Recovery Using A Virtual Manager Configuring a Cluster for Disaster Recovery In addition to using a virtual manager you must configure your cluster and volumes correctly for disaster recovery This section describes how to configure your system including the virtual manager Best Practice The following configuration steps ensure that you have all the data replicated at each site and the managers configured correctly to handle disaster recovery For the following example we are configuring two sites with two DataFRAMEs at each site for an even number of DataFRAMEs The management group contains one cluster The cluster contains four DataFRAMEs and one volume with 2 way replication that spans both sites That volume must contain all the data in each site Configuration Steps Name DataFRAMEs with Site Identifying Host Names To ensure that you can easily identify in the Console which DataFRAMEs reside at each site use host names that identify where each DataFRAME is located See Changing the DataFRAME Host Name on page 41 Management Group Name Transaction_Data e DataFRAME names Denver 1 Boulder 1 Denver 2 Boulder 2 Create Management Group Plan the Managers and Virtual Manager When you create the management group in the 2 site scenario plan to start two managers per site and add a virtual manager to
312. name or password the Log In window opens shown in Figure 29 38 MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs x Log in to DataFRAME DF140 1 192 168 0 9 a Trying the user name and password from the following DataFRAME s failed DataFRAME DF 140 admin user name or password was incorrect User Name Password Cancel Figure 29 Automatic Log In Fails because DataFRAME User Name and Password are Different 1 Type the correct User Name and Password 2 Click Log In Closing the DataFRAME Configuration Window Log out of the DataFRAME to close the DataFRAME configuration window or click Done to close without logging out Logging Out of the DataFRAME Log out to prevent access to a DataFRAME without closing the Console This provides security if you are leaving the management workstation but do not want to close the Console 1 When the DataFRAME Configuration window is open click Log Out as shown in Figure 27 and Figure 28 The Network view opens and the DataFRAME you logged out of no longer displays the logged in icon the pink square Note If you are logged in to multiple DataFRAMEs you need to log out of each DataFRAME individually MPC DataFRAME 440 Storage System Software User Manual 39 Working with DataFRAMEs Closing the DataFRAME Configuration Window without Logging Out Clicking Done on the DataFRAME configuration window returns the Console to th
313. named according to the card s slot and the port number such as Slot1 Port0 For information about how the interfaces are labeled on the back of the DataFRAME see Configuring the IP Address Manually on page 83 Once you have established a connection to the DataFRAME using a terminal emulation program you can configure an interface connection using the Configuration Interface 1 On the Configuration Interface main menu tab to Network TCP IP Settings and press Enter The Available Network Devices window opens shown in Figure 206 Available Network Devices lt Motherboar Figure 206 Selecting an Interface to Configure 2 Tab to select the network interface that you want to configure and press Enter The Network Settings window opens shown in Figure 207 304 MPC DataFRAME 440 Storage System Software User Manual Using the Configuration Interface If the interface you selected is a bond then the Logical Interface Device window displays first Click Change Settings to open the Network Settings window for the bond Figure 207 Entering the Host Name and Settings for an Interface 3 Enter the host name and tab to the next section to configure the network settings Note If you specify an IP address the Gateway is a required field If you do not have a Gateway enter 0 0 0 0 for the Gateway address 4 Tab to OK and press Enter to complete the network configuration A second window opens asking you to c
314. nclude the replication level and what factors contribute to the availability status Factors include but are not limited to the status of DataFRAMEs participating in any replication and RAID restriping due to auto grow or other factors 32 MPC DataFRAME 440 Storage System Software User Manual Getting Started Details Management Group Information Availability Management Group Group_0 Manager Yes Hot Spare NA Virtual Manager No Migrating Data NA Status RAID Normal Figure 22 Availability Tab Logging In to and Out of the DataFRAME After finding all the DataFRAMEs on the network you must log in to each DataFRAME individually to configure modify or monitor the functions of that DataFRAME 1 On the Network view double click the DataFRAME that you want to log in to The Log In window opens shown in Figure 23 k EdX Configuration for OF 140 192 168 0 7 Backup amp Restore Host Name DFi Appy IP Address 192 168 0 7 Model DF 100 Series SerialNumber OU JOARZEBASA Software Version 63 41 2505 0 Iestall Software Logged in User admin Change Password Figure 23 Logging in to a DataFRAME 2 Type the User Name and Password 3 Click Log In When you are successfully logged in the DataFRAME configuration window opens to the Module configuration category shown in Figure 25 Logging In to Additional DataFRAMEs Once you are logged in to a DataFRAME you can log in aut
315. ng with DataFRAMEs 56 MPC DataFRAME 440 Storage System Software User Manual 3 Storage Storage Overview For each DataFRAME you can select the RAID configuration the RAID rebuild options and monitor the RAID status You can also manage individual disks including powering them on or off and reviewing disk information Storage Requirement You must configure RAID before you can use a DataFRAME for data storage Getting There 1 On the Network View double click the DataFRAME and log in if necessary The Edit DataFRAME Configuration window opens 2 Select the Storage configuration category The Storage category opens shown in Figure 42 edit Configuration for DF440 1 192 168 0 40 Module RAID Setup Disk Setup SESCH RAID Status Normai TCP IP Network am DataFRAME Administ gt SNMP RAID Configuration RAIDO x CH 2 Start RAID K Reconfigure RAID IS Reporting The RAID Rebuild Rate is a percentage of RAID card throughput Percent ye fa Gore oa as SOI ape AE 1 10 20 30 40 50 60 70 80 90 100 Apply Device Name Device Type Subdevices dewscsi host0 busOftargetO unOidisc Disk D dewiscsiihostO busOtargett lun0 disc Disk D deviscsi host0 busOitarget2 unO disce Disk D dewscsishost0 busOitarget3lunOidisc Disk D dewscsi host0 busOitarget4 unO disc Disk 0 dewscsi host0 busOftarget5 unOidisc Disk D dewscsi host0 busOftarget6 lunO disc
316. ngs General Traps Active Monitoring Remote Log Files C Email Copy Configuration to DataFRAMEsS Copy A DataFRAME Management Group Cluster RW DF140 192 168 0 7 Available DataFRAMES RI DF 420 1 192 168 0 4 Available DataFRAMEs Oo DF420 192 168 0 3 Available DataFRAMEs Select All Clear All Copy Cancel Figure 26 DataFRAME Copy Configuration Window 1 In the Configuration Settings section select which configurations you want to copy For information about the configuration settings see the following Enabling the SNMP Agent on page 132 Using Active Monitoring on page 144 Remote Log Files on page 143 Setting Email Notification on page 151 2 In the Copy Configurations to DataFRAMEs section select the DataFRAMEs to which you want to copy the configurations 3 Click Copy The configuration settings are copied to the selected DataFRAMEs 4 Click OK on the confirmation window The Copy Configuration window closes 36 MPC DataFRAME 440 Storage System Software User Manual 2 Working with DataFRAMEs DataFRAME Configuration Window Overview The DataFRAME configuration window shown in Figure 27 opens when you log into an individual DataFRAME From the configuration window you have access to all the configuration tasks for individual DataFRAMEs edit configuration for DF140 192 168 0 7 SNMP Reportipg K 3 Storage Time Gei Network am DataFRAME Adminis Se H
317. ni wR A eee 128 Deleting Administrative Users AA 129 Chapter 7 Using SNMP cscs cascceitensawsesedcasnediinnstinsansdananseaesntascedacvesewetaceansstancedeuvertasscaeuccnnseendiye 131 Getting NICE 131 Enabling the SNMP Agent 2 3 2ecicccescesscessescagteacsazescteugeteccedscsh assed senasyenacdeeauneractuedeeaseeeeteaued dee SERA 132 Choosing Access Control ecceecceesceeeeeeeneeeeeeeeeeeceeeeeeeesaeceaeeeaeeeeaeeseeeeeaeesaeseaeeeneeeseeseaeeeaeesaeseaeeseeeesieeseeeeeaeesas 133 By Oe T A EERSTEN ENEE ergeet Ee 133 By lu 133 Editing Access Control Entries 2220 134 Deleting Access Control Entries A 135 Entering System Information Optional A 135 Using the SNMP MIB E 135 Installing the Storage System MIB EE 135 Disabling the SNMP Agent AAA 136 Disabling SNMP EE 136 Enabling and Disabling SNMP Traps cccceeeceeeeeseeeeeeeeeeeeeseeeeeaeeesaeeeaeesaeeseeeeseaeeeaeesaeseaeeseeeesieeseeeeaeeteeeneeenaaes 136 Enabling SNMP Traps naaa aea a aa aaa aan URE raaa PaE aaa aaa aa aaea Aaaa E AAEE 137 Disabling EI GN EE 138 Chapter 8 Reportmg xrsccccsccseccseseccnccecansczeeznccsnaacedtecsenaaseceeenuadecasuacteussanteccernaassieeceuneccseacnceie 139 Reporting Overview Using Passive Reports Saving the Report to a File eeececceeeeeeeeeeeeeeeeeeeaeeceeeeeaeessaeeeeceseaeeeaeeseeeeeseeseaeesaeesieeesaeeseaeesiaeseaeeeneeesas 140 Passive Reporting Detail s cisAiaues si Ani d e E init EELER Eden 140 Saving Log EE 142 Remote Log Files
318. ns for installing the Console configuring individual DataFRAMEs as well as creating volumes that span a cluster of multiple DataFRAMEs Topics in this manual include the following e Installing the Console e Configuring individual DataFRAMEs by Configuring monitoring and reporting e Creating volumes that span a cluster of DataFRAMEs by Creating management groups and clusters Creating volumes that span multiple DataFRAMEs Controlling client access to volumes Creating and using snapshots of volume Installing the Console 1 Before you install the Console software ensure that you have set up an IP address and password on your storage system Instructions for setting up an IP address and password are available on the MPC Storage System Quick Start User s Guide that shipped with your storage system 2 Insert the Drivers CD that shipped with your storage system into the system from which you will install the Console software Scroll down and click on the Agree to accept the license agreement Select Microsoft Windows Selection under Software Heading Select Run when prompted oa fk Ww Click on Next at the introduction screen Getting Started 7 Read the license agreement and click on I Accept to accept the terms of the license agreement Click on Next 8 Select the Complete option and click on Next E DFMC Installer Complete 2 All ofthe
319. nstall qualification After the system passes the post install qualification the upgrade process is complete 8 Click Close when the installation is completed Backup and Restore of DataFRAME Configuration Note 44 Backup and restore provides the capability to save the DataFRAME configuration file for use in case of a DataFRAME failure When you back up a DataFRAME configuration all of the configuration information about the DataFRAME is stored in a file on the computer where the Console is installed If a DataFRAME failure occurs you can restore the configuration file to a new DataFRAME The new DataFRAME will be configured identically to the DataFRAME when it was backed up Backing up the configuration file for a DataFRAME does not save information about the configuration of any management groups clusters volume lists or authentication groups that the DataFRAME belongs to It also does not back up license key entries for registered features To preserve a record of management group configuration information and license keys see Backing Up a Management Group Configuration on page 175 Back up the DataFRAME configuration every time you change DataFRAME settings This ensures that you can restore a DataFRAME to its most recent configuration MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs Backing Up Multiple DataFRAMEs with the Same Note Backing Configuration If you have mult
320. nsures that if a DataFRAME in a cluster becomes unavailable clients using an initiator such as Novell that does not support multiple addresses per target can still access the volume through the other DataFRAMEs in the cluster If the initiator you are using does support multiple addresses you may not want to use a virtual IP Note If you are using Microsoft cluster services you must use a Virtual IP to ensure correct operation Requirements for a Virtual IP e DataFRAMEs must be in same subnet address range as the virtual IP e The virtual IP must be routable regardless of which DataFRAME it is assigned to e iSCSI clients must be able to ping the virtual IP e Must be unique to all DataFRAMEs on the network e Must be a specific IP reserved for this purpose If you use DHCP you must use a Static IP e All iSCSI initiators must be configured to connect to this IP for failover MPC DataFRAME 440 Storage System Software User Manual 193 Working with Clusters Using an iSNS Server An iSNS server simplifies the discovery of iSCSI targets on multiple clusters on a network You can have up to 3 iSNS servers 194 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters Creating a Cluster Creating a cluster is the first step in designating space for storage in a management group Note If you plan to have two clusters each with one DataFRAME the most reliable configuration is to create two managemen
321. ntication Group tab to see the new group displayed in the list shown in Figure 183 Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Name Volume List iSCSI Mode Initiator Node Na EBSD Mode Fibre Channel NamemwwWwPN AG_TransDataR olume_List_2 No access No Access Access Allowed bkupSS 1adfe23 AG_AcctRec Volume_List_1 No CHAP required iqn 1991 05 com No Access No Access AG_AcctPay Volume List D No CHAP required iqn 1991 05 com No Access No Access Figure 183 Viewing the Authentication Groups Editing an Authentication Group You can edit the following e Change the description e Add a volume list e Change the types of authentication in either of those modes Warning Depending on your configuration editing an authentication group may interrupt client access to volumes If necessary stop client access before editing an authentication group See Client Access and iSCSI on page 266 and Client Access and Fibre Channel MPC DataFRAME 440 Storage System only on page 271 before changing iSCSI or Fibre Channel authentication group parameters 1 Log into the management group and select that management group in the Network View The management group Tab View opens 2 Click the Authentication Groups tab to bring it to the front 3 Select from the list the group you want to edit 4 From the Tasks menu select Edit Authenticat
322. nual Setting the Date and Time 4 Click OK The NTP server is added to the list on the NTP tab shown in Figure 78 wer Modse ag Slorage o fr com Network Cater RANE Admnrisi g gt SNWP Hl Reporing NTP Mode NTP is cf To turn NTP on aac an NTP server to the NTP tab below Time Zone Date and Time NTP NTP Server Preferred tne mit edu Figure 78 Viewing the List of NTP Servers Editing NTP Servers You can change the properties of NTP servers To change the IP address of an NTP server you must remove the one no longer in use and add a new NTP server 1 Make certain that the NTP Mode is On 2 On the NTP tab select the NTP server you want to edit 3 Click Edit The Edit NTP Server window opens shown in Figure 79 O e NTP Server Preferred Not Preferred Figure 79 Editing an NTP Server 4 Change the preference of the NTP server 5 Click OK The list of NTP servers displays the changed NTP server in the list MPC DataFRAME 440 Storage System Software User Manual 117 Setting the Date and Time Note To change the IP address of an NTP server you must remove the server no longer in use and add a new NTP server 118 MPC DataFRAME 440 Storage System Software User Manual D Administrative Users and Groups The Storage System Software comes configured with two default administrative groups and one default administrative user You can add edit and delet
323. o access those remote snapshots Off site remote snapshots particularly when supplemented with synchronous volume replication within a cluster ensures high availability of critical data volumes Configuration for High Availability To use remote snapshots for high availability configure a backup application server to access remote snapshots in the event of a primary system failure Figure 241 illustrates this simple high availability configuration e Configure clustered application servers in both the primary and backup locations e During normal operation the production application server read writes to the primary volume Sample Remote Copy Configurations e Set up a schedule for copying remote snapshots to the backup location If your application server uses multiple volumes that must be in sync use a script to quiesce the application before creating remote snapshots Configuration Diagram Application Workstations Backup Application Serer Clustered SH Production Application Application Servers Read Only Primary ao Remote Volume H Volume Storage L i Subnet Primary Remote Snapshot Snapshot Figure 241 High Availability Example Configuration How This Configuration Works for High Availability If the production application server or volumes become unavailable application processing fails over to the backup application server As shown in Figure 242 the remote volume and remote snapshots become pri
324. o avoid potential connectivity and performance problems with other devices on your network keep the frame size at the default setting If you decide to change the frame size set the same frame size on all DataFRAMEs on the network and set compatible frame sizes on all clients MPC DataFRAME 440 Storage System Software User Manual 101 Managing the Network The frame size on the DataFRAME should correspond to the frame size on Windows and Linux application servers Table 20 shows recommended DataFRAME frame sizes and the corresponding frame sizes for Windows and Linux clients Table 20 Setting Corresponding Frame Sizes on DataFRAMEs and Windows or Linux Clients DataFRAME Windows Client Linux Client Frame Frame Size Frame Size Size 1500 Default 1542 Default 1500 Default 4046 4088 4046 8972 9014 8972 Frame sizes greater than 1500 bytes called jumbo frames can co exist with 1500 byte frames on the same subnet if the following conditions are met e Every device downstream of the DataFRAME on the subnet must support jumbo frames e If you are using 802 1q virtual LANs jumbo frames and non jumbo frames must be segregated into separate VLANs Best Practice Change the speed and duplex settings while the DataFRAME is in the Available mode and not in a management group Note The frame size for a bonded logical interface must be equal to the frame size of the NICs in the bond Editing the Frame Si
325. o remove variables to stop active monitoring You can return a variable to active monitoring at any time Permanent variables such as motherboard temperature cannot be removed See List of Monitored Variables on page 149 1 Select the variable you want to remove 2 Click Remove A confirmation message opens 3 Click OK The variable is removed Note Variables are not deleted when they are removed from active monitoring You can add them back to active monitoring at any time Adding Variables to Monitor You can only add variables that have been previously removed The variables that the DataFRAME is currently monitoring are listed in the box All variables in the list are configured and set for Console alerts 1 Click Add The Configure Variable wizard opens to Step 1 shown in Figure 102 Configure variable Step 1 of 3 Select Variable Fibre Channel Status Next gt Cancel Figure 102 Adding a Variable Step 1 2 Select the variable that you want to monitor and click Next The Configure Variable wizard Step 2 opens shown in Figure 103 146 MPC DataFRAME 440 Storage System Software User Manual Reporting Configure variable Step 2 of 3 x Specify Frequency Seconds Enter number of minutes Minutes booo si sdY lt Back Next gt Cancel Figure 103 Adding a Variable Step 2 3 Specify the frequency for monitoring the variable and click Next
326. ode Virtualization and Clustering in violation because of virtual DataF RAME DF 420 L B Not licensed for Multi Node Virtualization and Clustering in violation because of virtual DataFRAME DF 140 1 LB Not licensed for Multi Node Virtualization and Clustering in violation because of virtual We suggest you upgrade your license immediately by contacting your storage provider www mpccorp com 1 888 224 4247 KS Figure 195 Evaluation Period Countdown Message 290 MPC DataFRAME 440 Storage System Software User Manual Feature Registration Viewing Licensing Icons Icons indicate the status of licensing on individual modules Figure 196The figure below illustrates the icons for each state n license violation for Remote Copy evaluation period Cota ena rer Stongen West re re eremm LEI No license for Remote Copy Details Managers Clusters Authentication Groups Volume Lists Red Remote Copy DN Management group version is 6 1 licensed D Communications protocol version is 2 P E OctarRave Da Figure 196 Icons Indicating License Status for Features Evaluating the Scalability Pak The Scalability Pak includes the following features e Multiple nodes in a cluster e Hot spares e N way replication e Virtual manager Starting the License Evaluation Period If you put more than one DataFRAME into a cluster the 30 day license
327. odern disks A health test one or program inside the disk constantly more tracks a range of the vital parameters drives fails including driver disk heads surface health test state and electronics This information may be used to predict hard drive failures RAID controller Checks the status of the RAID BBU is Failed if BBU Status controller Battery Backup Unit BBU Normal Charging Faulty or Missing 154 MPC DataFRAME 440 Storage System Software User Manual Reporting Viewing Alerts Any time that an actively monitored variable causes an alert the alert is logged by the DataFRAME If the Console is open alerts display in the Alert Messages tab on the Console main window shown in Figure 108 Available DataFRAMEs Management Groups Alert Messages EBSD Hosts 7 0607 2005 12 52 58 PM MST Alert from Nightly FT 2 10 0 7 4 RAID Status normal Value changed 6 06 07 2005 12 52 17 PM MST Alert from Stavy 1 10 0 11 123 CPU Fan Speed 0 Value is below 3100 5 06 07 2005 12 52 13 PM MST Alert from Nightly FT 2 10 0 7 4 Device flash 1 Status Active Value changed 4 06 07 2005 12 52 11 PM MST Alert from bmastors 10 0 3 109 CPU Fan Speed 0 Value is below 3100 3 0607 2005 12 51 41 PM MST Alert from Nightly FT 10 0 7 2 RAID Status normal Value changed 2 06 07 2005 12 51 11 PM MST Alert from bmastors 10 0 3 109 CPU Fan Speed 0 Value is below 3000 1 06 07 2005 12 51 11 PM MST Alert from Niahtlv
328. ole alert if not normal Power Supply No Yes 1 minute Console alert if Status status changes RAID Status Yes Yes 15 Console alert if seconds changes Remote Copy No Yes 15 Console alert if true Complete minutes Remote Copy No Yes 15 Console alert if true Failovers minutes Remote Copy No Yes 15 Console alert if fails Status minutes Remote No Yes 1 minute Console alert if Management changes Group Status Snapshot No Yes 1 minute Console alert if Status snapshot status changes MPC DataFRAME 440 Storage System Software User Manual Reporting Table 31 List of Variables Available for Active Monitoring i z Perm Specify Default Default Action Variable Name units Variable Freq Freq Threshold Storage Server No Yes 1 minute Console alert if not Status up Volume No Yes 1 minute Console alert if Restripe completed Complete Volume Status No Yes 15 Console alert if minutes volume status changes Volume No Yes 15 Console alert if Thresholds minutes threshold exceeded for any volume or snapshot in the mgt group Setting Email Notification If you request email notification on the Active tab you set the email addresses to receive the notifications there You then use the Email tab to configure the SMTP settings for email communication For more information on configuring active monitoring see Using Active Monitoring on page 144
329. omatically to additional DataFRAMEs configured with the same user name and password by double clicking those DataFRAMEs in the Network view MPC DataFRAME 440 Storage System Software User Manual 33 Getting Started If you try to log in to a DataFRAME that uses a different user name or password the Log In window opens shown in Figure 24 x Log in to DataFRAME DF140 1 192 168 0 9 Trying the user name and password from the following DataFRAME s failed DataFRAME DF140 admin user name or password was incorrect User Name Password Cancel Figure 24 Automatic Log in Failure 1 Type the correct User Name and Password 2 Click Log In Logging Out of the DataFRAME Log out to prevent access to a DataFRAME without closing the Console This provides security if you are leaving the management workstation but do not want to close the Console 1 When the DataFRAME Configuration window is open click Log Out shown in Figure 25 The Network view opens and the DataFRAME you logged out of no longer displays the logged in icon the pink square Note If you are logged in to multiple DataFRAMEs you need to log out of each DataFRAME individually DataFRAME Configuration Window The DataFRAME Configuration window opens when you log into an individual DataFRAME From the configuration window you have access to all configuration tasks for individual DataFRAMEs To configure specific settings of an individual mod
330. on DataFRAMEs individually which is recommended If you are upgrading multiple DataFRAMEs that are not in a management group you can upgrade them simultaneously 1 Log in to the first DataFRAME you want to upgrade 2 On the Module Information tab click Install Software The Install Software window opens shown in Figure 30 42 MPC DataFRAME 440 Storage System Software User Manual E tustall software xj Working with DataFRAMEs File Name Version No file name Description No file name Browse Insta DataFRAME Version DF420 1 192 168 6 3 33 0007 0 LI DF140 1 192 168 6 3 41 2605 0 J DF420 192 168 0 3 6 3 33 0007 0 Management Group Cluster Available DataFRA Available DataFRA Available DataF RA Check All Install file on selected DataFRAMEs one at a time Recommended Install file on selected DataFRAMEs simultaneously Advanced Clear All Close Figure 30 Upgrading the DataFRAME Software 3 Select the Install check box next to the DataFRAME you want to upgrade You can select multiple DataFRAMEs to upgrade from the list 4 Select Install file on selected DataFRAMEs one at a time Recommended 5 Click Browse to navigate to the folder on the Console computer where you copied the upgrade or patch file as shown in Figure 31 x Look In Storage_System_Software cs et ce gz Fi
331. on on 2 DataFRAMEs this does not guarantee data availability in the event that one DataFRAME becomes unavailable See Managers on page 158 216 MPC DataFRAME 440 Storage System Software User Manual Working with Volumes If your volumes contain critical data configure them for 2 way replication and a priority of redundancy MPC DataFRAME 440 Storage System Software User Manual 217 Working with Volumes Requirements for Volumes When creating a volume you define the following parameters Table 35 Parameters for Volumes Volume Parameter Configurable in Volume Type What it Means Type Any Whether the volume is primary or remote Primary volumes are used for data storage Remote volumes are used for configuring Remote Copy for business continuance backup and recovery or data mining migration NOTE Remote Copy is a feature upgrade You must purchase a Remote Data Protection Pak license to use remote volumes past the 30 day trial period Volume Name Any The name of the volume that is displayed in the Console A volume name must be from 1 to 127 characters and is case sensitive Description Any Optional A description of the volume Cluster Any If the management group contains more than one cluster you must specify the cluster on which the volume resides Replication Level Any The number of copies of the data to create on DataFRAMEs in the cluster The
332. on the DataFRAME see Chapter 5 Setting the Date and Time Then refresh the management group time 1 Log in to the management group 2 Select the Times tab 3 From the Tasks menu click Refresh All Verify the time settings on the DataFRAMEs running managers 4 Click Reset Management Group Time A confirmation message opens 5 Click OK All the times listed on the Times tab should be the same Starting and Stopping Managers Start or stop managers on DataFRAMEs already in a management group 1 Log in to the management group 2 Click the Managers tab in the Tab View The Managers tab displays shown in Figure 122 MPC DataFRAME 440 Storage System Software User Manual 171 Working with Management Groups kl C Computers DataFRAME Management Console File Find Tasks View Help SS Available DataFRAMEs Group_0 Sr lee ee DF420 1 DF140 1 GEES Details 1 Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Name Manager Virtual Manager DF420 DF140 DF140 1 Figure 122 Starting a Manager 3 Select from the list the DataFRAME on which you want to start a manager 4 From the Tasks menu click Start Manager 5 Click OK 172 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups Adding Manager IP Addresses to Application Servers When you add a manager to a management group a window o
333. onfirm the changes 5 Press Enter Return to the Storage System Console and locate the DataFRAME using the Find menu to search by subnet and mask or search by entering the DataFRAME IP address Deleting a NIC Bond You can delete two types of NIC bonds using the Configuration Interface e Active backup bond e NIC aggregation bond For more information about creating and configuring NIC aggregation and active backup bonds see Configuring NIC Bonding on page 84 When you delete an active backup bond the primary interface assumes the IP address and configuration of the deleted logical interface The other NIC is disabled and its IP address is set to 0 0 0 0 When you delete a NIC aggregation bond one of the active interfaces in the bond retains the IP address of the deleted logical interface The other NIC is disabled and its IP address is set to 0 0 0 0 MPC DataFRAME 440 Storage System Software User Manual 305 Using the Configuration Interface 1 On the Configuration Interface main menu tab to Network TCP IP Settings and press Enter The Available Network Devices window opens shown in Figure 208 The logical bond is listed in the window Figure 208 Selecting a Bonded Interface in the Available Network Devices Window 2 Tab to select the bond and press Enter The Logical Failover Device window opens shown in Figure 209 Figure 209 Deleting a NIC Bond 3 Tab to Delete Bond and press Enter A window opens
334. or a Fibre Channel volume enter the LUN number Warning Before you associate a LUN to an authentication group the host server s Disk Management or equivalent window must be closed 7 Click OK when you are finished 8 The volume is listed on the Volumes tab Adding Authentication Groups to the Volume List 1 Next select the Authentication Groups tab shown in Figure 187 New Volume List d New Volume List Name TransDataBkUp Description Yolumes Authentication Groups Authentication EBSD GC Fibre Channel Remove Cancel Figure 187 Connecting Authentication Groups to a Volume List 2 Click Add A list of existing authentication groups opens 3 Select the authentication group that you want to give access to the volume or snapshot 4 Click OK The authentication group is listed on the Authentication Groups tab MPC DataFRAME 440 Storage System Software User Manual 283 Controlling Client Access to Volumes Completing the Volume List 1 Click OK on the New Volume List window when you are finished The Network View opens Go to the Volume Lists tab to see the new list displayed shown in Figure 188 Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Volume List Name Volume List 0 Volume_List_1 Volume_List_2 Volumes Authentication Groups Volume D Snapshot_0 AG_AcctPa
335. orking with Management Groups on page 157 for more information about removing DataFRAMEs from a management group Registering Available DataFRAMEs for License Keys DataFRAMEs that are not in a management group are licensed individually in the module configuration category see Registering Features for a DataFRAME on page 51 296 MPC DataFRAME 440 Storage System Software User Manual Feature Registration Registering DataFRAMEs in a Management Group DataFRAMEs that are in a management group are licensed through the management group Submitting DataFRAME Serial Numbers First you must submit the serial numbers of all the DataFRAMEs that you want to register 1 Select the management group for which you want to register features or applications 2 Select the Register tab shown in Figure 198 The Register tab lists licenses that have been purchased If you are evaluating features the time remaining in the evaluation period is listed on the tab as well Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Figure 198 Registering Features and Applications 3 From the Tasks menu select Feature Registration The Feature Registration window opens shown in Figure 199 Listed are all the DataFRAMEs in that management group MPC DataFRAME 440 Storage System Software User Manual 297 Feature Registration Feature Registration Feature Registration
336. ot client_name s The EBSD driver client s hostname ip_bind x x x x The IP address of the client that you want this driver to bind to This address identifies the interface over which the driver will communicate to this volume in a multihomed system management_group s The name of the management group that contains the volume auth_group s The authentication group assigned to this volume volume_name S The volume name that is used to create the local ebsd disk MPC DataFRAME 440 Storage System Software User Manual 321 Using the EBSD Driver for Linux 322 Table 56 Parameters in ebsd conf Parameter What It Is access_mode r ro rw r or ro read only rw read write The access mode the driver should use to access this volume use_unicast true false Whether the driver should use unicast discovery unicast_list X X X X X X X X Used if the use_unicast flag is set to true Coma separated list of ip addresses to be used for unicast discovery enabled true false Designates whether the device should come online at boot time If false then the device is disabled at boot time use_multicast true false see note Whether the driver should use multicast discovery NOTE Use either unicast or multicast Do not use both together Sample Device Entry in etc ebsd conf EHE HEHE EHE HEHE HEH HH HH HHHH HH HHHH Sam
337. ote snapshots Retaining 4 hour snapshots for 1 week can result in the accumulation of over 40 remote snapshots Another approach would be to create 2 remote snapshot schedules for the volume e One schedule to create remote snapshots every 4 hours but only retain the most recent 3 remote snapshots This will ensure that you do not lose more than 4 hours worth of data in an emergency e A second schedule to create remote snapshots every 24 hours and retain 7 remote snapshots 374 MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations Use remote snapshots in conjunction with local synchronous volume replication To prevent data loss reinforce Remote Copy with synchronous replication of the volume within the cluster of DataFRAMEs at the primary geographic site With synchronous replication a single DataFRAME can be off line and your primary volume will remain intact At the backup location you can also use synchronous replication to protect your remote volume against DataFRAME failure Example configuration e In the production location create a cluster of 3 DataFRAMEs all with managers e Create volumes on the cluster and set the replication level to 2 e Create a schedule for making remote snapshots of the primary volume Set the recurrence to every 4 hours and retention of remote snapshots to 2 days Note You can use the same volume replication configuration on the remote volume as wel
338. ote volume is created the license evaluation period begins on both the primary and remote DataFRAMEs For example suppose the primary volume is on Cluster 1 You create a remote snapshot of that primary volume to Cluster 2 DataFRAMEs in both clusters show the clock ticking for the license evaluation period Read the Remote IP Copy User Manual before working with these features Backing Out of the License Evaluation Period If you decide not to purchase the Remote Data Protection Pak you must delete any remote volumes you have configured The features you are evaluating dictate the steps required to safely back out of the evaluation configuration particularly if you want to save any volumes or snapshots in the test configuration 1 Back up any volumes you plan to retain Table 54 describes additional steps to safely back out of the Remote Data Protection Pak evaluation MPC DataFRAME 440 Storage System Software User Manual 293 Feature Registration Table 54 Safely Backing Out of Remote Data Protection Pak Evaluation Feature Being Evaluated Steps to Back Out Remote snapshots removing data e Delete any remote snapshots from the remote target Delete the remote volume Remote snapshots retaining the e Make the remote volume into a primary volume data on the remote target Disassociate the primary and remote management groups if the remote copy was between management groups Scripting Evaluation Application ba
339. ou select Reporting from the DataFRAME configuration category list the Reporting category opens shown in Figure 98 Edit Configuration for DF140 192 168 0 7 xj Module d Passive Log les Active Email Diagnostics Alerts T Storage Time TCPIIP Network item Value 3 DataFRAME Administ Passive Report 11 18 2005 11 07 32 AM PST gi SNMP Host Name DF140 ki Reporting IP Number 192 168 0 7 DataF RAME Software Version 6 3 41 2605 0 GUI Software Version 6 3 41 Build 0005 Support Key unknown call support Boot Devices IC vance or flash 0 Active flash 1 Active more details NIC Data 2 cards Card 1 Description Intel Corp 82546GB Gigabit Ethernet Controller Copper Mac Address 00 30 48 2E BA 5A Address 192 168 0 7 Mask 255 255 255 0 Log Out Refresh Save to File Y Advanced Figure 98 Viewing the Reporting Window Reporting Using Passive Reports Passive reports display statistics about the performance of the DataFRAME its drives and configuration Statistics in the passive reports are point in time data gathered when you click the Refresh button on the Passive tab Select Advanced to see additional statistics 1 Select Reporting from the configuration categories The Reporting window opens 2 Click Refresh to display statistics on the Passive tab 3 Optional To view extended statistics about the DataFRAME click Advanced at the bottom of the window
340. ou want to stop the manager Right click and select Stop Manager A confirmation message opens oar Go M Click OK to confirm stopping the manager Removing Manager IP Addresses from Application Servers When you remove a manager from a management group a window opens that displays all the IP addresses of the managers in that management group and a reminder to reconfigure the application servers that are affected by the change Click Save Manager IP Addresses to save this list You can print the list and use it as a reference when reconfiguring application servers Removing a DataFRAME from a Management Group Remove a DataFRAME from an existing management group Prerequisites Stopping or removing the DataFRAME from data storage activities e Remove all snapshots and volumes from the cluster containing the DataFRAME See Deleting a Snapshot on page 253 and Deleting a Volume on page 233 e Remove the DataFRAME from any cluster to which it belongs See Removing a DataFRAME from a Cluster on page 204 e Stop the manager on the DataFRAME if it is running a manager You may want to start a manager on a different DataFRAME to maintain quorum and the best fault tolerance See Stopping a A Manager on page 173 Removing a DataFRAME With a License Key If you have registered the DataFRAME for add on features the license key is saved on the DataFRAME when you remove it from the management group That key
341. ountain Standard Time Start At 07 06 2005 06 39 43 PM Z Recur Every Thresholds for Scheduled Snapshots Size 8 0 Hard Threshold 2 0 Soft Threshold 170 Retention for Scheduled Snapshots Retain Maximum Of 5 Snapshots Retain Snapshots For Figure 165 Editing a Snapshot Schedule 5 Change the desired information 6 Click OK Note If you change the hard threshold be sure to review the information about snapshot thresholds and their effect on volume thresholds in Managing Capacity Using Volume and Snapshot Thresholds on page 237 Deleting Snapshot Schedules 1 Select the volume for which you want to delete the snapshot schedule The volume Tab View opens 2 Click the Snapshot Schedule tab to bring it to the front 3 Select the schedule you want to delete 4 From the Tasks menu select Delete Schedule 250 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Scripting Snapshots Application based scripting is available for taking snapshots Using application based scripts allows automatic snapshots of a volume For detailed information see Chapter 14 Working with Scripting Check your vendor s web site for specific applications for which sample scripts have been developed Rolling Back a Volume to a Snapshot Rolling back a volume to a snapshot replaces the original volume with a read write copy of the selected snapshot The new volume ha
342. oup Information Availability HostName DF140 RAID Status Normal IP Address 192 168 0 7 RAID Configuration RAID 0 Model DF 100 Series Storage Mode block lt Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Figure 16 Viewing individual DataFRAME information 30 MPC DataFRAME 440 Storage System Software User Manual Getting Started Details Tab Includes host name IP address the model serial number RAID status RAID configuration storage mode and software version RAID States RAID states are displayed on the DataFRAME Details tab e If RAID is normal a green circle displays in the DataFRAME configuration details tab when the DataFRAME is selected in the Network View as shown in Figure 17 J J Details Management Group Information Availability Host Name DF140 RAID Status Normal IP Address 192 168 0 7 RAID Configuration RAID O Model DF 100 Series Storage Mode block SerialNumber 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Figure 17 Icon Showing RAID is Normal e f RAID is off a red circle displays in the DataFRAME configuration details tab as shown in Figure 18 For information about turning RAID on see Starting RAID on page 68 Details Management Group Information Availability HostName DF 140 1 RAID Status or IP Address 192 168 0 9 RAID Configuration RAID 0 Model DF 100 Series Storage Mode block Serial Number 00 30 48 2E B9 D2 Software Vers
343. oup tab 2 Add or remove authentication groups as required 3 Click OK when you are finished The Tab View opens Select the Volume Lists tab shown in Figure 191 to review your changes Removing a Volume from a Volume List Remove volumes from a volume list in preparation for deleting the volume list Warning Before deleting volumes from the volume list stop any applications from accessing those volumes you are removing 1 In the Edit Volume List window shown in Figure 189 on the volumes tab select the volume or snapshot you want to remove from the volume list 2 Click Remove 3 Click OK when you are finished The Tab View opens Select the Volume Lists tab shown in Figure 191 to review your changes Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Volume ListName Volumes Authentication SC Volurne_List_1 Snapshot_O AG_AcctRec Volume_List_2 rem_test AG_TransDataRemote Figure 191 Viewing the Edited Volume List 286 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Deleting a Volume List Deleting a volume list removes that list from the management group Prerequisites e Remove all volumes from the volume list e Remove all authentication groups from the volume list Warning Before you delete access to a volume the host server s Disk Management or equivalent window must be closed Note To pre
344. ous volume replication on multiple DataFRAMEs within a cluster in combination with asynchronous Remote Copy on a different cluster of DataFRAMEs creates a robust high availability configuration Uses for Remote Copy Table 58 Uses for Remote Copy Use Remote Copy for How It Works Business continuance disaster Using Remote Copy store remote snapshots off site The recovery remote snapshots remain continuously available in the event of a site or system failure Off site backup and recovery Remote Copy eliminates the backup window on an application server by creating remote snapshots on a backup server either local or remote and backing up from that server Split mirror data migration Using Remote Copy make a complete copy of one or more content distribution volumes without interrupting access to the original volumes Move the copy of the volume to the location where it is needed Benefits of Remote Copy e Remote Copy maintains the primary volume s availability to application servers Snapshots on the primary volume are taken instantaneously and are then copied to remote snapshots in the off site location e Remote Copy operates at the block level moving large amounts of data much more quickly than file system copying e Snapshots are incremental that is snapshots save only those changes in the volume since the last snapshot was created Hence failover recovery may need to resynchronize only the latest c
345. out begins counting from the time that the DataFRAME begins blinking in the Console The default time is set to 0 seconds so that the hot spare takes over as soon as the system detects that the DataFRAME is unavailable For example if you set the time out to 60 seconds then the hot spare is activated 1 minute after the system detects that the DataFRAME is unavailable 192 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters Swap in Hot Spare You can manually force a DataFRAME designated as a hot spare to activate in the cluster if a DataFRAME in that cluster is not available and the cluster is blinking red in the Console Swapping in a hot spare overrides the hot spare time out setting However the setting remains intact in the cluster and continues to apply once the cluster configuration has returned to normal Clusters and iSCSI If you plan to use iSCSI with the Storage System Software there are iSCSI features you configure at the cluster level either when you create the cluster or by editing the cluster to configure these items e iSCSI Failover If you are using an initiator that does not support multiple addresses per target such as Novell to ensure iSCSI failover you must configure a virtual IP for the DataFRAMEs in a cluster s ISNS Server If you use an iSNS server configure your cluster to register the iSCSI target with the iSNS server iSCSI Failover and Virtual IP A virtual IP address e
346. ower Off Disk A confirmation message opens 5 Click OK Powering Drives On 1 When a new disk is inserted into a DataFRAME that is on the disk must be powered on 2 Select the disk in the list to power on If the disk is not powered on it is listed as Off or Missing in the Status column and the other columns display dotted lines like this 3 Click Power On Disk A confirmation message opens 4 Click OK 78 MPC DataFRAME 440 Storage System Software User Manual H Managing the Network Managing the Network Overview The DataFRAME has two integrated TCP IP network interfaces In addition the MPC DataFRAME 440 Storage System can include three add on cards each with 2 or 4 interfaces For each DataFRAME you can e Configure the DataFRAME s TCP IP interfaces e Set up and manage a DNS server e Manage the routing table e View and configure the TCP interface speed duplex and frame size e Update the list of managers running in a DataFRAME s management group e Bond NICs to ensure continuous network access or to improve bandwidth Getting There 1 On the Network View double click the DataFRAME and log in The DataFRAME Configuration window opens 2 Select TCP IP Network from the DataFRAME configuration categories The window opens with the TCP IP tab on top shown in Figure 54 edit Configuration for DF440 1 192 168 0 40 RAID Setup Disk Setup Time RAID Status Normal D TCP IP Netwo
347. p error gossip info Ihydra log kernel error Save Files Add Edit Delete Log Out gn Figure 99 Saving Log Files to a Local Machine Scroll down the Choose Logs to Save list Select the file or files you want to save To select multiple files use the Ctrl key Click Save Files The Save dialog opens Select a location for the file or files Click Save In Directory The file or files are saved to the designated location MPC DataFRAME 440 Storage System Software User Manual Reporting Remote Log Files Use remote log files to automatically write log files to a computer other than the DataFRAME For example you can direct the log files for one or more DataFRAMEs to a single log server in a remote location The computer that receives the log files is called the Remote Log Target You must also configure the target computer to receive the log files Adding a Remote Log 1 Select Reporting from the configuration categories The Reporting Window opens 2 Select the Log Files tab The Log Files window opens shown in Figure 99 3 Click Add below the list of remote logs The Add Remote Log Target window opens shown in Figure 100 Add Remote Log Target Log Type Alerts log local5 mark Destination Figure 100 Adding a Remote Log 4 In the Log Type list select the log that you want to direct to a remote comput
348. p on which to start a manager The DataFRAME Tab View opens MPC DataFRAME 440 Storage System Software User Manual 163 Working with Management Groups 2 On the Management Group Information tab select the Tasks menu and click Start Manager as shown in Figure 116 Add to New or Current Management Group Remove from Management Group Start Manager Stop Manager Start Virtual Manager gt GI Normal Stop Virtual Manager Repair DataFRAME Swap in Hot Spare Click Start Manager onfiguration RAID 0 ge Mode block are Version 6 3 41 2605 0 Edit Configuration Copy Configuration Log Out Figure 116 Starting a Manager 3 Repeat steps 1 and 2 to start managers on additional DataFRAMEs Logging In to a Management Group You must log in to a management group to administer the functions of that group Note Log in to a management group by logging in to a DataFRAME that is designated as a manager for that management group 1 Click Log In on the management group in the Network View shown in Figure 117 EE Mre computers DataFRAME Management Console ll x File Find Tasks View Help SS Available DataFRAMEs BG Group o pa DF140 1 DF140 Click Log In DF420 1 k ails nagers Clusters uthentication Groups oume Lists tegisten imes emote Snapshot Detail Manag Ciusti Authentication G Vol List Ri t iJ Ri tes hot Management Group Group_0 Manage
349. pModules snmpModules SNMP MIB Information snmpTargetMIB MIB Tree snmpNotificationMIB MIB Tree snmpProxyMIB MIB Tree snmpUsmMIB usm MIBObjects usmUser usm UserTable MIB Tree snmpVacmMIB vacm MIBObjects vacmContextTable MIB Tree snmpCommunityMIB MIB Tree MPC DataFRAME 440 Storage System Software User Manual 313 SNMP MIB Information 314 MPC DataFRAME 440 Storage System Software User Manual C Using the EBSD Driver for Linux The EBSD Driver for Linux provides access to and management of Storage System Software volumes from Linux systems Install and configure the EBSD Driver for Linux on the computer that accesses the DataFRAME Note You will need root privileges during installation and configuration Use the X11R6 window environment Installing the EBSD Driver for Linux You can install the EBSD Driver for Linux using either the Installation Wizard or the RPM packages available on the Resource CD See Installing the EBSD Driver with RPM Packages on page 318 Options Available When Using the Installation Wizard The EBSD Driver for Linux CD has a graphical installation wizard that allows you to install the EBSD Driver directly on your system or the EBSD Driver Bundles The EBSD Driver Bundles can be installed locally or on a network share The appropriate driver version can be installed later from the bundled files Copying Driver Bundle to a Network Share Optional Copying the driver bundle result
350. pens shown in Figure 123 which displays all the IP addresses of the managers in that management group and a reminder to reconfigure the application servers that are affected by the change Manager IP Addresses x For each host that is using EBSD enter the following manager IP addresses in the driver dialog box used to create and edit volumes 10 0 33 18 10 0 33 25 Save Manager IP Addresses Figure 123 Adding Manager IP Addresses to Application Servers Click Save Manager IP Addresses to save this list You can print the list and use it as a reference when reconfiguring application servers Stopping a A Manager Under normal circumstances you stop a manager when you are removing a DataFRAME from a management group You cannot stop the last manager in a management group Deleting the management group is the only way to stop the last manager Implications of Stopping Managers e Quorum may be decreased e Fewer copies of configuration data are available e Fault tolerance may be lost e Data integrity and security may be compromised Warning Stopping a manager can result in the loss of fault tolerance MPC DataFRAME 440 Storage System Software User Manual 173 Working with Management Groups Select the management group in the Network View Log in to the management group Select the Managers tab in the Tab View The Managers tab displays Select the DataFRAME on which y
351. ple device entries device0 type volume client_name myclient ip_bind 10 0 1 63 management_group my_mgtgroup auth_group public volume_name my_volume_0 access_mode rw use_unicast true unicast_list 10 0 0 12 10 0 0 13 enabled true use_multicast false EE HEE EHE HEHE HEH H H HH HHHH HH HHHH device1 type volume client_name myclient ip_bind 10 0 1 63 management_group my_mgtgroup auth_group public volume_name my_volume_1 access_mode rw use_unicast true unicast_list 10 0 0 12 10 0 0 13 enabled true use_multicast false MPC DataFRAME 440 Storage System Software User Manual Using the EBSD Driver for Linux Connecting the EBSD Devices to the DataFRAME EBSD Server 1 Run the startup script which will spawn a child process for the new device etc init d ebsd start loads the driver 2 Start and wait until all devices are online There is a timeout of two minutes Verifying EBSD Devices The EBSD driver creates a block device for each volume 1 Check the current status of the device entries Use cat proc ebsd client It should say online suzyl cat proc ebsd client Version 04 04 03 4 1 14 0004 Majors 176 177 178 Device 0 my_mgtgroup_0 my_volume_0 Status Online Active Read DP Requests 0 Write DP Requests 0 Ops o sync 0 Cycles 147 BSize 512 Capacity 52428800 kb Device 1 my_mgtgroup_0 my_volume_1
352. ple network configurations using NIC aggregation in high availability environments are illustrated Server re ve IP SAH Ethemet Switch am ee er mg IP SAH Ethemet Switch Storage Module Cluster Server B IP SAN Ethemet Switch Figure 61 NIC Aggregation in a Partial mesh Topology with Server 92 Failover MPC DataFRAME 440 Storage System Software User Manual Managing the Network ServerA ServerB ill IP SAH Ethemet Switch Storage Module Cluster Figure 62 NIC Aggregation in a Single switch Topology Creating a NIC Bond Follow these guidelines when creating NIC bonds You can create bonds of 2 or 4 interfaces You can create more than one bond on a DataFRAME An interface can only be in one bond To provide failover capability in the event of a PCI card failure bond interfaces located in the motherboard with interfaces in PCI slots This ensures that if an entire PCI card fails then the bonded interface will use an interface in the motherboard to continue operating Record the configuration information of each interface before you create the bond When you delete an active backup bond the preferred interface assumes the IP address and configuration of the deleted logical interface When you delete a NIC aggregation bond one of the interfaces retains the IP address of the deleted logical interface The IP address of the other interface is set to 0 0 0 0 Create a bond on a DataFR
353. plication without registering a 30 day evaluation period begins Throughout the evaluation period you receive reminders to register and purchase a license for the add on module and applications you want to continue using The Feature Registration tab lists the status of add on modules and applications on your DataFRAME An example of an unlicensed application in the evaluation period is shown in Figure 40 gt Power Of Feature Registration Boot Devices Module Information Backup amp Restore Registration is required for addibonal software features Current feature registr ation license keys are displayed betow For addtional feature registration please contact your storage provider www mpecorp com 1 080 224 4247 and be prepared to provide the serial number s indicated below u n licensed and i n Setat manner 003048 7E BASA the evaluation License Key period Figure 40 Using Remote Copy without a License For more detailed information about the evaluation process see Evaluating Features on page 52 Configuring Boot Devices The MPC DataFRAME 440 Storage System has two boot devices Modules with Two Boot Devices 52 Note When a DataFRAME with two boot devices powers on or reboots it references boot configuration information from one of two compact flash cards located on the front of the module The module boot configuration information is mirrored between the two compact flas
354. pull a disk from its drive bay and then push it back into that same drive bay e For RAID 1 10 or 5 50 you must first power on the disk on the Disk Setup tab and then select the disk and click Add to RAID e DataFRAMEs with BBU For RAID 0 you must first go to the Disk Setup tab and power on the disk Then go to the RAID Setup tab and click Reconfigure RAID Warning Reconfiguring RAID destroys all the data on the disks DataFRAMEs without BBU For RAID 0 you must first power on the disk on the Disk Setup tab and then select the disk and click Add to RAID For information about the rate at which RAID rebuilds on disks added to RAID see Setting RAID Rebuild Rate on page 68 Managing Disks Use the Disk Setup tab to monitor information about the disks in the selected DataFRAME to power on a disk that you have replaced or added to the DataFRAME and to add disks to RAID You can also power off disks on this tab Getting There 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens to the Module Information tab shown in Figure 42 2 Select the Storage configuration category 3 Click the Disk Setup tab to bring it to the front as shown in Figure 51 Any drive bays that do not contain disks are labeled Off or Missing in the Status column Disks that have been inserted in the DataFRAME but not yet added to RAID are labeled Uninitialized in the St
355. quired Figure 169 Open the MS iSCSI initiator and Copy the Initiator Node Name to the Initiator Node Name Field Figure 170 illustrates the configuration for a single host authentication with 1 way CHAP required 268 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes zix Sees TE General IPSec New Authentication Group Connect by using Name AuthGroup_0 Local adapter Default e Descrinti Source IP Default d escription Target Porta Default zl Volume List No Volume List iSCSI EBSD Fibre Channel M CRC Checksum T Data digest I Header digest vj Allow access via iSCSI Authentication CHAP logon information CHAP helps ensure data security by providing authentication between CHAP not required a target and an initiator trying to establish a connection To use it specify the same target CHAP secret that was configured on the target for this initiator EN ii User name ign 1991 05 com microsoft kwegner wxp CHAP required Teriga Jeseeseeeees CHAD Name qn 1991 05 com microsoft kwegner wxp EE JT Perform mutual authentication Target Secret To use mutual CHAP specify an initiator secret on the Initiator Settings Initiator Secret page and configure that secret on the target Figure 170 Configuring iSCSI shown in the MS iSCSI initiator for a Single Host with CHA
356. r Both interfaces have an active status If the interface link to one NIC goes down the other interface continues operating Using both NICs also increases network bandwidth Requirements for NIC Aggregation To configure NIC aggregation e Both NICs should be enabled e NICs must be configured to the same subnet e NICs must be connected to a single switch that is LACP capable and supports 802 3ad link aggregation If DataFRAME is directly connected to a server then the server must support 802 3ad link aggregation 90 MPC DataFRAME 440 Storage System Software User Manual Managing the Network Which Physical Interface is Preferred Because the logical interface uses both NICs simultaneously for data transfer neither of the NICs in an aggregation bond are designated as preferred Which Physical Interface is Active When the NIC aggregation bond is created if both NICs are plugged in both interfaces are active If one interface fails the other interface continues operating For example suppose Motherboard Port0 and Slot1 PortO are bonded in a NIC Aggregation bond If Motherboard Port0 fails then Slot1 PortO remains active Once the link is fixed and Motherboard PortO is operational it becomes active again Slot1 PortO remains active Table 16 DataFRAME NIC Aggregation Failover Scenario and Corresponding NIC Status Example Failover Scenario NIC Status 1 NIC aggregation bond0 is created Bondo is the master lo
357. r Checksum Data Status Description Snapshots f Volume_0 11 23 2005 04 10 54 AM GMT Cluster_0 No Normal None Snapshot Schedules 500 GB Hard Threshold 10 GB Soft Threshold 10 GB Auto Grow Off Size Hosts Replication Level Remote Snapshot None Replication Priority Availability Type Primary Created by Script No Figure 160 Snapshot Tab The tabs provide access to snapshot information and features such as editing snapshots rolling back a volume and associating authentication groups with the snapshot The Tasks button on each tab provides access to the actions you can take related to that tab 242 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Details Tab Displays information about the selected snapshot Use the Tasks menu to edit and delete snapshots and roll back volumes from this tab Hosts Tab Lists all EBSD hosts that are associated to the snapshot To retrieve the IP Mode Type and driver version for the snapshot put the IP address of the host in Find By Module IP or Host Name Remote Snapshot Tab Lists remote snapshots associated with a snapshot Buttons include creating a remote snapshot and canceling a remote snapshot that is in progress Target Information Tab Displays the Initiator name assigned by the Microsoft iSCSI Initiator and the masked Target Secret Volume List Memberships Tab Displays information about volume list memberships
358. r Changing Hard Threshold Increase the hard threshold to turn off an alert generated when the threshold is exceeded The hard threshold must be equal to or less than the size of the volume and there must be sufficient space on the cluster To decrease the hard threshold first decrease the size of the volume and then decrease the hard threshold to the same value as the size Warning Decreasing the volume size or hard threshold is not recommended If you shrink the volume in the Storage System Console before shrinking it from the client file system your data will be corrupted or lost Soft Threshold The soft threshold must be equal to or less than the hard threshold To decrease the soft threshold first decrease the hard threshold and decrease the soft threshold to a value less than the hard threshold Auto Grow Change the auto grow setting as desired You can turn auto grown on or off Warning Decreasing the volume size or hard threshold is not recommended If you shrink the volume in the Storage System Console before shrinking it from the client file system your data will be corrupted or lost Getting There 1 Select the volume you want to edit in the Network View The volume Tab View opens 2 Click Edit Volume The Edit Volume window opens as shown in Figure 156 See Table 38 for detailed information about making changes to the volume parameters 230 MPC DataFRAME 440 Storage System Software User Manual E
359. r business continuance disaster recovery and provide a consistent predictable update of data for remote backup and recovery Planning the Remote Copy Schedule When creating a remote snapshot schedule a number of considerations are important to plan All of these issues impact the amount of storage available in the system Recurrence How often do you want the snapshots created The recurrence frequency must account for the amount of time it takes to complete a remote snapshot For example if your recurrence schedule is set for a new snapshot every 4 hours you should ensure that the time to copy that snapshot to the remote location is less than 4 hours One way to check the time required to copy a snapshot is to run a test of the actual process In the test you take two remote snapshots of the primary volume Since the first remote snapshot copies the entire volume it will take longer to copy The second remote snapshot copies only changes made to the volume since the first remote snapshot Since you create the second remote snapshot after the time interval you intend to schedule the copy time for the second remote snapshot is more representative of the actual time required for copying subsequent remote snapshots 1 Create a remote snapshot of the primary volume 2 Wait for the copy to finish 3 Create another remote snapshot of the primary volume MPC DataFRAME 440 Storage System Software User Manual 335 Remote Copy 4 Track
360. rage System Software User Manual 109 Managing the Network Selecting the Interface Used by the Storage System Software Note Warning The Storage System Software uses one network interface for communication with other DataFRAMEs on the network In order for clustering to work correctly the Storage System Software communication interface must be designated on each DataFRAME The interface can be e a single NIC that is not part of a bond e a bonded interface consisting of 2 or 4 bonded NICs Only NICS that are in the Active or Passive Ready state can be designated as the communication interface You cannot make a disabled NIC the communication interface When you initially set up a DataFRAME using the Configuration Interface the first interface that you configure becomes the interface used for Storage System Software communication To change the communication interface first remove the DataFRAME from the management group To select a different communication interface 1 Make sure that the DataFRAME is not in a management group 2 Select TCP IP Network from the DataFRAME configuration categories 3 Click the Communication Mode tab to bring it to the front shown in Figure 74 4 Select an interface from the Storage System Software Interface drop down list 5 Click Apply Updating the List of Manager IP Addresses Update the list of manager IP addresses to ensure that a manager running on this DataFRAME is communica
361. rage System Software User Manual 309 Using the Configuration Interface 310 MPC DataFRAME 440 Storage System Software User Manual B SNMP MIB Information SNMP Agent The SNMP Agent resides in the DataFRAME The agent takes SNMP network requests for reading or writing configuration information and translates them into internal system requests Management Information Base MIB files are provided which can enable the system administrator to use their favorite SNMP tool to view or modify configuration information The SNMP Agent supports versions 1 2c and 3 of the protocol Security can be configured based on the host making the request and a password Note To ensure that all items display properly in your SNMP tool use version 2c or later of the protocol Supported MIBs MIB II e Host Resources MIB UCD Extensions MIB SNMPv3 MIB Exceptions MIB II system sysServices interfaces ifTable ifEntry ifLastChange interfaces ifTable ifEntry ifInNUcastPkts interfaces ifTable ifEntry ifInDiscards interfaces ifTable ifEntry ifInUnknownProtos interfaces ifTable ifEntry ifOutNUcastPkts ip ipAddrTable ipAddrEntry ipAdEntReasmMaxSize ip ipRouteTable ipRouteEntry ipRouteMetric2 ip ipRouteTable ipRouteEntry ipRouteMetric3 ip ipRouteTable ipRouteEntry ipRouteMetric4 ip ipRouteTable ipRouteEntry ipRouteAge ip ipRouteTable ipRouteEntry ipRouteMetric5 ip ipForward MIB Tree tcp tcpiInErrs tcp tcpOutRsts SNMP MIB Information tcep ipv T
362. re 140 5 Select the cluster to open the clusters Tab View also shown in Figure 140 198 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters mrc Computers DataFRAME Management Console File Find Tasks View Help Available DataFRAMEs T Weu a a Cluster H DF420 1 lt p am d DF140 1 DF140 DF420 GER Cluster Details Volumes Daaf Disk Usage Usage Graphs iSCSI Remote Snapshot Name E TN Cluster lab View Hot Spare Time Out 0 Seconds Hot Spare None Status Normal Description Figure 140 Viewing a Cluster and the Cluster Tab The Cluster Tab View The cluster Tab View is shown in Figure 140 In the Cluster Tab view tThe tabs provide access to cluster information and features along with a Tasks button for taking actions on features Details Tab Includes name status description hot spare and the DataFRAME to which the Virtual IP is assigned Volumes Tab Includes the name replication level size hard threshold and soft threshold for volumes in the cluster DataFRAMEs Tab For each DataFRAME in the cluster this tab includes the host name IP address and whether the module is a hot spare for the cluster MPC DataFRAME 440 Storage System Software User Manual 199 Working with Clusters Disk Usage Tab Displays usage statistics for the cluster and the modules volumes and snapshots contained in the cluster
363. reated and the volume becomes a remote volume The system creates a new primary snapshot of the primary volume and a remote snapshot of the remote volume It then copies the data from the primary snapshot to the remote snapshot This process will recur according to the schedule Editing a Remote Snapshot Schedule Note 354 When editing a remote snapshot schedule you can change the following items e Schedule description start date and time recurrence policy e Primary Setup primary snapshot thresholds retention policy e Remote Setup retention policy Plan threshold changes carefully See the Managing Capacity Using Volume and Snapshot Thresholds on page 237 for detailed information about threshold requirements 1 Select the primary volume that has the schedule you want to edit 2 Click the Remote Snapshot Schedules tab 3 Select from the list the schedule to edit A From the Tasks menu select Edit Schedule The Edit Remote Snapshot Schedule window opens shown in Figure 233 Wu Change the desired information 6 Click OK MPC DataFRAME 440 Storage System Software User Manual Remote Copy Edit Remote Snapshot Schedule Edit Remote Snapshot Schedule Name Description Remote Snapshot Schedule Time Zone Mountain Standard Time Start At 12 09 2004 04 45 02 PM V Recur Every 30 Primary Setup Remote Setup Management Group Transaction_Data Volume Name Credit
364. replication level must be at most the number of DataFRAMEs in the cluster or three whichever is smaller See Planning Data Replication on page 214 Replication Priority Any Availability Default setting These volumes will remain available as long as at least one DataFRAME out of every n n replication level remains active When the unavailable DataFRAME returns to active status in the cluster than the volume resynchronizes across the replicas Redundancy Choose this setting to ensure that the volume will go offline if it cannot maintain two replicas For example if 2 way replication is selected anda DataFRAME in the cluster becomes unavailable thereby preventing 2 way replication the volume goes offline until the DataFRAME is again available Size Primary The logical block storage size of the volume Hosts and file systems will operate as if storage space equal to the volume size is available in the cluster This volume size may exceed the true allocated disk space on the cluster for data storage which facilitates adding more DataFRAMEs to the cluster later for seamless storage growth However if the volume size does exceed true allocated disk space the ability to make snapshots may be impacted See Chapter 13 Working with Snapshots Remote volumes contain no data and therefore do not have a size The default value in the size field is equal to the available space on the cluster
365. rk Ki ki Start RAID DataFRAME Administ RAID Configuration RAID 0 Ge SNMP v Reconfigure RAID El Reporting The RAID Rebuild Rate is a percentage of RAID card throughput Percent P d Device Name Device Type Subdevices dewscsi host0 busOfargetO lunOidisc devscsifhost0 husOitargett lunO disc dewscsi host0 busOtarget2 unOldisc dewscsi hostO busMitarget3 unO dise devscsi host0 busOitarget4 unOidise dewscsi host0 busOfargetS lunO disc devscsihostO busOitarget6 lunOidisc deviscsi host0 busOitarget8 unOidisc Disk Disk Disk Disk Disk Disk Disk Disk Log Out deviscsi host busOtargetO unOidisc Figure 54 Viewing the Network Configuration Managing the Network The TCP IP Tab The TCP IP tab lists the network interfaces on the DataFRAME You can configure each of these interfaces Table 8 Network interfaces Displayed on the TCP IP Tab Name Description NICs Embedded in the DataFRAME Motherboard Motherboard Port1 1000BASE T interface Motherboard PortO 1000BASE T interface Add on NICs in PCI Slots Slot1 PortO Slot1 Port1 and so on Multiple add in PCI cards each containing up to 4 Ethernet or Fibre Channel interfaces Bonded Interfaces bondN Optional You can create multiple bonded interfaces each consisting of 2 or 4 physical interfaces Use the TCP IP tab to manage the network con
366. roll back 2 Select the snapshot to which you want to roll back 3 Review the snapshot Details tab to ensure you have selected the correct snapshot 4 From the Tasks menu on the Details tab select Roll Back Volume The Roll Back Volume window opens shown in Figure 166 Roll Back Yolume Roll Back Volume New Volume Name Volume_0 Description Cluster Replication Level Replication Priority Size Hard Threshold 2 0 Soft Threshold 2 0 Auto Grow Checksum Data Figure 166 Rolling Back a Volume 5 Type a new name for the rolled back volume You can also change the hard and soft thresholds if necessary 6 Click OK The Roll Back Volume confirmation message shown in Figure 167 explains that the original volume will be deleted 252 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Storage System Console xj A Rolling the volume back to this snapshot will delete the volume Before proceeding you must manually delete all snapshots of the volume that are newer than this snapshot To preserve the volume and snapshots that you need to delete create a copy of the volume and snapshots before rolling the volume back Are you sure you want to roll the volume back to this snapshot Cancel Figure 167 Verifying the Volume Roll Back 7 Click OK The snapshot version of the volume is restored as a read write volume 8 Add the r
367. rs NA Local Bandwidth NA Virtual Manager NA Quorum NA Status NA Coordinating manager NA Figure 117 Logging in to a Management Group 164 MPC DataFRAME 440 Storage System Software User Manual Working with Management Groups The Manager Log In List window opens shown in Figure 118 Any DataFRAMEs to which you are already logged in display Yes in the Logged In column Manager Log in List xi Select a manager to log in to Logged In Boulder 3 Cancel Figure 118 List of DataFRAMEs Running Managers 2 Select a DataFRAME and click Log In Whatever view of the Console is displayed when you log in to a management group that is the view that returns after logging in to that management group E Mre computers DataFRAME Management Console E ioj xj File Find Tasks View Help REES TA Completed management group Managers Clusters Authenticat S Volume Lists Register Times Remote Snapshot Management Group Group_0 Managers 1 of 1 Local idth 32768 Kbisec Management Virtual Manager None Quorum 1 Group Tab View Status Normal Coordinating manager DF140 Figure 119 Viewing a Management Group in the Console MPC DataFRAME 440 Storage System Software User Manual 165 Working with Management Groups Management Group Tab View When you have logged in to and selected a management group the management group Tab View opens The tabs provide access to ma
368. rting SNMP General SNMP Traps IP Host Name if Log Out Figure 91 Using SNMP Using SNMP Enabling the SNMP Agent 3 Click Enable SNMP Agent The Enable Agent fields become activated shown in Figure 92 edit Configuration for DF140 192 168 0 7 ZS Module SNMP General SNMP Traps W Storage Time Disable SNMP Agent TCP IP Network Sm DataFRAME Administ 1 e Enable SNMP Agent H Reporting Community String public Access Control IP Host Name 192 168 0 2 Single Host Add System Location System Contact Log out Apply Figure 92 Enabling the SNMP Agent 4 Type the Community String Note The community string identifies a group of hosts that are allowed read only access to the SNMP data The community public is typically used to denote a read access community This string is entered into the SNMP Management tool not included when attempting to access the system 132 MPC DataFRAME 440 Storage System Software User Manual Using SNMP Choosing Access Control 1 Click Add to add an SNMP client The Add SNMP Client window opens shown in Figure 93 You can add SNMP Client either by specifying IP addresses or host names x IP Network Address IP Netmask Single Host Figure 93 Adding an SNMP Client By Address 1 Click By Address 2 Type the IP Network Address 3 Sele
369. s DNS Routing Communication WW Storage Time DataFRAME Administ Name Description Speed Met DuplewMet Status Frame Size Preferred J SNMP bond0 Logical Fai Auto Auto Auto Auto Active Default Jl Reporting Motherb Intel Corp Unknown Unknown Passive F Default Yes Motherb Intel Corp 100Mbs A Half f Auto Active Default Preferred interface 4 Log Out Set Preferred Figure 66 Viewing the Status of an Active Backup Bond 96 MPC DataFRAME 440 Storage System Software User Manual Managing the Network k Bo Configuration for DF 140 192 168 0 7 gen Mode gt TCPAP TCP Status DNS Rowing Conmenunication 2 EE Admiris Name Descrig o Stabs e Preferred g HN Dev Logkal Fai Au Auto pf A Defaut e Ser eing Motherd intel Comp mt Unknowns Passie F Default P Motherb intel Corp 100 JA Halffmuto Achva Defaut Neither interface is preferred Figure 67 Viewing the Status of a NIC Aggregation Bond Note If the bonded NIC experiences rapid sequential Ethernet failures the Console may display the DataFRAME as failed flashing red and access to data on the DataFRAME fails However as soon as the Ethernet connection is reestablished the DataFRAME and the Console display the correct information Deleting a NIC Bond When you delete an active backup bond the preferred interface assumes the IP address and confi
370. s IP addresses in the DataFRAME configuration window in the TCP IP Network category on the DNS tab You can remove these DNS servers but the DataFRAME will not be able to resolve host names until you enter a new DNS server DNS and Static IP Addresses If you assigned a static IP address to the DataFRAME and you want the DataFRAME to recognize host names you must manually add a DNS server to the Network DNS tab Note If you initially set up the DataFRAME to use DHCP and then change the configuration to use a static IP address the DNS server provided by DHCP will remain on the DNS tab You can remove or change this DNS server 1 On the Network View double click the DataFRAME and log in if necessary 2 The DataFRAME Configuration window opens Select TCP IP Network from the DataFRAME configuration categories MPC DataFRAME 440 Storage System Software User Manual 103 Managing the Network 3 Click the DNS tab to bring it to the front shown in Figure 70 GBEdit Configuration for DF140 192 168 0 7 x Se Module TCPAP TCP Status DNS Routing Communication E W Storage i Time DNS Domain Name H D DataFRAME Administ DNS Servers in order gt SNMP IS Reporting feta a v Add Edit Remove DNS Suffixes in order lengineering mpccorp com A 4 D Log Out Apply Figure 70 Adding DNS Servers Adding the DNS Domain Name
371. s a different name than the original and the original volume is deleted Prerequisites e Stop applications from accessing the volume e Disable the volume if it is mounted by a host Requirements for Rolling Back a Volume Many of the parameters for the new volume must be configured as if you had created this volume for the first time Note After rolling back a volume to a snapshot you lose all data stored after the rolled back snapshot Table 44 Requirements for Rolling Back a Volume Parameter Requirements for Changing New Volume Name You must choose a new name for the volume The name must be from 1 to 127 characters Names are case sensitive New Hard Threshold Hard threshold size must be equal to or less than the size of the volume See Managing Capacity Using Volume and Snapshot Thresholds on page 237 New Soft Threshold Soft threshold size must be equal to or less than the hard threshold size Authentication Groups You must include the new volume in a volume list See Volume Lists Overview on page 281 Hosts You must reconfigure hosts to connect to the new volume Prerequisites e Stop applications from accessing the volume MPC DataFRAME 440 Storage System Software User Manual 251 Working with Snapshots e Delete all snapshots that are newer than the snapshot you are rolling back Rolling Back the Volume 1 Log in to the management group that contains the volume that you want to
372. s in separate tar gz files plus an install sh file copied to the location you specify The tar gz files correspond to the versions of Linux that are currently supported by the EBSD driver Running the install sh file from the directory containing all the tar gz files installs the appropriate driver for the Linux version on that system Skip this section if you only want to install the EBSD driver from the CD Go to Installing the EBSD Driver Using the CD or the Driver Bundles on page 316 Using the EBSD Driver for Linux With Autorun Enabled 1 Start the Windows Manager of your choice such as KDE or GNOME 2 Insert the driver CD into the CD drive of the EBSD host server Autorun should automatically start if properly configured A folder directory window also opens displaying the contents of the CDROM 3 The autorun window opens asking you to verify that you want to run Autorun The autorun window may be behind the CD folder directory 4 Click Yes to have the automatic install process run The automatic installation starts and steps you through the install process 5 On the Choose Product Component window of the automatic installation select EBSD Driver Bundles 6 Click Next The Choose Install Folder window opens 7 Choose to accept the default directory opt Storage_System Storage_System_Software 6 x Drivers EBSD or browse to the directory where you want the EBSD driver bundle installed This location can be anot
373. s of Passive Ready becomes active This configuration remains until the failed preferred interface is brought back online When the failed interface is brought back online it becomes Active The other NIC returns to the Passive Ready state MPC DataFRAME 440 Storage System Software User Manual Managing the Network Requirements for Active Backup To configure active backup e Both NICs should be enabled e NICs should be connected to separate switches Which Physical Interface is Preferred A preferred interface is an interface within an active backup bond that is used for data transfer during normal operation When you create an active backup bond one of the interfaces becomes the preferred interface in the bond You can change the preferred setting after creating the bond See Creating a NIC Bond on page 93 Which Physical Interface is Active Note When the active backup bond is created if both NICs are plugged in the preferred interface becomes the active interface The other interface is Passive Ready For example suppose you create an active backup bond consisting of 2 NICs Motherboard PortO and Slot1 PortO If Motherboard Port0 is the preferred interface it will be active and Slot1 PortO will be Passive Ready Then if Motherboard Port0 fails Slot1 PortO changes from Passive Ready to active Motherboard PortO changes to Passive Failed Once the link is fixed and Motherboard PortO is operational
374. s point the primary snapshot has been created 6 If you have already created the remote volume select the management group and existing remote volume in the Remote section of the New Remote Snapshot window Otherwise create a remote volume See Creating a Remote Volume on page 342 340 MPC DataFRAME 440 Storage System Software User Manual New Remote Snapshot New Remote Snapshot Primary Management Group Volume Name Snapshot Name Remote Management Group Volume Name Transaction_Data CreditData05 Snapshot_0 Transaction_Data Y New Snapshot Remote Copy xi a CreditData05 v New Volume Snapshot Name Snapshot Description Figure 219 New Primary Snapshot Created Completing the Remote Snapshot 1 Type a name for the remote snapshot 2 Optional Type a description for the snapshot The completed window is shown in Figure 220 3 Click OK New Remote Snapshot New Remote Snapshot Primary Management Group Transaction_Data Volume Name CreditData05 Snapshot Name Snapshot_0 New Snapshot Remote Management Group Transaction_Data Y v Volume Name Volume_0 New Volume Snapshot Name BkUpsSs_1 Snapshot Description Figure 220 Completing the New Remote Snapshot Dialog The system creates the remote snapshot in the cluster that contains the remote volume The system then copies the primary snapshot onto the remote snaps
375. s synchronized and running No action is required e Rebuild A new disk has been inserted in a drive bay and RAID is currently rebuilding No action is required e Degraded RAID is degraded Either a disk needs to be replaced or a replacement disk has been inserted in a drive You must add a disk to RAID on Disk Setup if you are inserting a replacement disk e Off Data cannot be stored on the DataFRAME The DataFRAME is down and flashes red in the Network view Replacing Disks and RAID Disk failure in a DataF RAME affects RAID for that module First replace the failed disk Then reestablish RAID on the DataFRAME e When using RAID 0 you must reconfigure RAID 0 If the DataFRAME is ina cluster you must first remove the DataFRAME from the management group and then reconfigure RAID 0 e When using RAID 1 10 or RAID 5 50 RAID must be rebuilt As long as RAID quorum was not lost you can replace disks in a DataFRAME and rebuild RAID while the DataFRAME remains in the cluster See RAID Quorum on page 69 You can view the status of the disks in the DataFRAME on the Disk Setup tab shown in Figure 51 on page 73 The RAID states are reported on the RAID Setup tab as shown in Figure 45 on page 61 MPC DataFRAME 440 Storage System Software User Manual 71 Storage Removing and Reinserting the Same Disk Warning If your DataFRAME has a BBU and you pull a disk from the drive bay you will lose all data on that disk If you
376. schedule New Remote Snapshot Schedule New Remote Snapshot Schedule Name pc Sen d emote Description snapshot Remote Snapshot Schedule SC hed u le P rima ry Time Zone Mountain Standard Time setup Start At 05 11 2004 02 23 58 PM Z Recur Every 1 d Setup Remote Setup Management Group Transaction_Data Volume Name Mastercard Snapshot Name Schedule Name Number Thresholds for Primary Snapshots Size 2 0 GB Hard Threshold 2 0 SB Y ard and soft thresholds Soft Threshold 2 0 Je x will be applied to the volume Retention for Primary Snapshots Retain Maximum Of 2 Snapshots O Retain Snapshots For Cancel Figure 231 Creating a New Remote Snapshot Schedule 352 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Remote Snapshot Schedule The time zone displayed in the Remote Snapshot Schedule area is the time zone set on the DataFRAME through which you are logged in to the management group Best Practice Set all DataFRAMEs in the management group to the same time zone Reset the management group time before creating a remote snapshot schedule For detailed information see Resetting the Management Group Time on page 171 1 Select a start date and time for the schedule 2 Optional Select a recurrence interval for the schedule Configuring the Primary Volume and Snapshots 1 On the Primary Setup tab
377. scriptions of Group Permissions Management Area Activities Controlled by This Area Network Choose type of network connection set the time and time zone for the DataFRAMEs identify the Domain Name Server and use SNMP Management Groups RAID Drive Hot Swap Set the RAID configuration for the DataFRAME Shut down disks restart RAID and hot swap disks Create management groups System and Disk Report View reports about the status of the DataFRAME Change Change administrative users passwords Password DataFRAME Add administrators and upgrade the Storage System Software Administration and Upgrade MPC DataFRAME 440 Storage System Software User Manual 123 Administrative Users and Groups What the Permission Levels Mean e Read Only User can only view the information about these functions e Read Modify User can view and modify existing settings for these functions e Full Users can perform all actions view modify add new delete in all functions Sorting Columns in the Administrative Group Window The columns in the Administrative Group window can be sorted in ascending or descending order e Click on the column header to sort e Click again to reverse the sort The arrow next to the column title indicates which column is the sorted column and whether the sorting order is ascending up arrow or descending down arrow pit contguraton orori40 192 16007 x Module
378. seasanlagivacedd deck esdcathadepatsagdiusseehtbscessaaneeadsasien Eaa a aa EAEE 337 Scheduled Remote Copy Planning Checklist AA 338 Working with Remote Snapshots AA 338 Creating a Remote Snapshot cecceeccceeeeeeeeeeeeeeeeeeneeeeeeeeeeeseaeeeaeeeeeeeaeeseeeseaeescaeseaeesieeeseeseeeesieeseaeesneeees 339 Et WEE 339 Creating the Primary Snapshot AA 340 Completing the Remote Snapshot AAA 341 Creating a Remote Volume ceeceeeceeeeeeeeeeneeeneeeeeeeeaeseaeseaeeseeeeaeeeseeseaeeeaeeseeeesaeeeaeeeeeseaeeeaeeeseeeseaeeeneeseaeeeaeeee 342 Making an Existing Volume into a Remote Volume AA 342 Creating a New Remote Volume oe ceeeceecceeseeceeeeeeeeeneeeeaeeeeeeeeaeesaeeeaeeesaeeeaeeseaesesaeeeaeenaeeseeeeeaeseaeeeaeesieeees 343 Viewing a List of Remote Snapshots AAA 345 Setting the Remote Bandwidth AAA 345 Canceling a Remote Snapshot AA 346 Editing a lee DT dE 347 Deleting a Remote Snapshot A 347 Monitoring Remote Snapshots A 348 Configuring Active Monitoring Alerts for Remote Copy cecceeeceeeeeeeeeeeeeeeeeeeteaeeeeesneeeseeeeeaeeseaeeeaeeseetes 348 Monitoring Remote Snapshot Details from the Console Tab VieW ccceeceesceeeeeeeeeeteeeeeeeeeeeetteeeeeeteas 348 Viewing Information in the Remote Snapshot Tab AA 348 Viewing Status in the Remote Copy Details WiINdOW ecccecceeeeeeeeeeeeeeeeeeseeeeeaeeeeeeseaeesaeeseaeesiaeeeaeenaeees 349 Scheduling Remote Snapshots A 351 Creating the Schedule ze egeeuareguerg reggggee Eege j
379. sed scripting is available for volume and snapshot features as part of the Configurable Snapshot Pak and the Remote Data Protection Pak Features that can be scripted include e Creating snapshots and setting hard and soft snapshot thresholds e Increasing volume hard and soft thresholds e Scripting automatic threshold increases e Creating remote volumes and snapshots Because using scripts with add on features and applications starts the 30 day evaluation period without requiring you to use the Console you must first verify that you are aware of starting the 30 day evaluation clock when using scripting If you do not enable the scripting evaluation period any scripts you have running licensed or not will fail Turn On Scripting Evaluation To use scripting while evaluating add on features or applications enable the scripting evaluation period 1 Select the management group 2 Select the Register tab 3 From the Tasks menu select Feature Registration 4 Select the Scripting Evaluation tab shown in Figure 197 294 MPC DataFRAME 440 Storage System Software User Manual Feature Registration Feature Registration Feature Registration Registration is required for additional software features Current feature registration license keys are displayed below For additional feature registration please contact your storage provider www mpccorp com 1 888 224 4247 and be prepared to provide the serial number s indicated below
380. seeeeeaeecseeseaeeeaeeseaeseeessaeesieesseeesaes 360 Resuming Production After Failover A 360 Synchronizing Data After Failover cecceecceseseeeeeeseeeeeeeeeeeseaeeeeaeeeaeessaeeeaeeseaeeeaeeeeeeeaeesaeeeeaeeseaeseaeeeeeesas 361 Returning Operations to Original Primary Site e eceecceeeceseeeeeneeeeeeeeeeeeeaeeeseeseaeeseeeesaeeeeeeeeaeesieeseaeesaes 361 Synchronizing the Data Between the Acting Primary Volume and the Original Primary Volume 361 Creating a New Primary Volume at the Original Production Site A 362 Setting Up a New Production Site 2 0 ees eeceeeeeeeeeeceeeeeeeeneeeeaeeeeeseaeeseaeeaeeaeeseeeeeaeeseaeeeaeesaeesaeeseaeescaeseaeseneeeeas 363 Making the Backup Site into the New Production Site 0 2 0 ecceeeeeeeeeeeeeeeeeeeeeeeeeeeesaeseeesseeeseeeeeeeenees 363 Rolling Back Primary and Remote VoOlUMES AA 363 Rolling Back a Primary Volume 000 00 ee eeeeceeeeeneeeeeneeeteneeeeeaeeeeeaneeesaeeeeenaeeeeeaeeseaeeeeeaeeeneaeeeseaeeessaeeesenaeeeeeeeesenaees 363 Rolling Back a Remote Volume o eeeeceeeesneeeeeeneeteneeeeeeneeeeeeneeseaaeeeseneeeeeaneeesaaeeseaeeeneaaeeseneeeeneaeeeseneeeneaes 365 Using Remote Snapshots for Data Migration and Data Mining ceecceeeeseeeeeeeeeteeeeeeeeeeseaeeeeeeeeaeeteeeeeeensaes 365 Creating a Split lge 365 Disassociate Remote Management Group eecceeeceeeeeeeeeeeeeeeeaeeceeeeeaeeeeaeeeseeeeaeeeaeeseaeseeeeeeaeesieeeeaeesiaes 366 Appendix E Sample Remote Copy Con
381. si host0 busOftargett flunO disc dewscsi host0 busOstarget2 unO disc dewscsi host0 busOfarget3 lunO disc deviscsi host0 busOltarget4 lunOldisc deviscsi host0 busOltargetS lunOldisc deviscsihost0 busOitarget lunOidisc deviscsihost0 busOitarget lunOidisc dewscsi hostt busOMargetO lunO disc Disk Disk Disk Disk Disk Disk Disk Disk Disk a Figure 51 Viewing the Disk Setup Tab in the MPC DataFRAME 440 Storage System MPC DataFRAME 440 Storage System Software User Manual 73 Storage Figure 52 Diagram of the Drive Bays in the MPC DataFRAME 440 Storage System Using the Disk Report The Disk Setup tab lists the individual disks in the module and provides information about them Table 6 Description of items on the disk report This Item Describes This Disk Lists the numbers of the disk drives The disks are numbered sequentially from left to right as you view the front of the DataFRAME Model The model of the disk in the DataFRAME Serial The serial number of the disk Number Class The class type of disk The DataFRAME uses SATA disks Capacity The data storage capacity of the disk Status Whether the disk is Active and participating in RAID Status column Active other columns with informatio
382. signated off site locations If your application server uses multiple volumes that must be in sync use a script to quiesce the application before creating remote snapshots e Create routine tape backups of the remote snapshots See the example configuration illustrated in Figure 248 376 MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations Configuration Diagram Application Workstations Production EH Library Production Application Application Sener Server Application Servers Read Only Read Only Primary ES 4 Remote E d Remote Primary Volume i Volume gi i Volume 2 Volume Storage ku Kat od d Subnet Primary Remote Remote Primary Snapshot Snapshot 1 Snapshot 2 Snapshot Production Site 1 Backup Site Production Site 2 Figure 248 Off site Backup and Recovery Example Configuration How This Configuration Works for Off site Tape Backup Depending on how long you retain the copies of the remote snapshots you can retrieve data directly from recent remote snapshots rather than going to tape backups Otherwise retrieve data as you normally would from the tape backup Best Practices Retain the most recent primary snapshots in the primary cluster By keeping snapshots on your primary volume you can quickly roll back a volume to a previous snapshot without accessing off site backups e When you create a schedule for Remote Copy you specify a number of primary and re
383. size and cannot exceed available space in the cluster Soft Threshold The amount of space actually used on the writable volume that triggers a warning alert This alert notifies the storage administrator that the writable volume is approaching the hard threshold The soft threshold must be less than or equal to the hard threshold 236 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Managing Capacity Using Volume and Snapshot Thresholds How Snapshots are Created When you create a snapshot of a volume the original volume is actually saved as the snapshot and a new volume the writable volume with the original name is created to record any changes made to the volume s data after the snapshot was created Subsequent snapshots record only changes made to the volume since the previous snapshot Hard Thresholds and Snapshots One implication of the relationship between volumes and snapshots is that the space used by the writable volume can become very small when it records only the changes that have occurred since the last snapshot was taken This means that less space or a smaller hard threshold may be required for the writable volume You can save space on your cluster of DataFRAMEs by estimating the size required for the changes in data between snapshots and decreasing the hard threshold of each snapshot accordingly This planning is particularly important if you plan to use a series of
384. size of the thresholds for the snapshot is a 100 MB hard threshold and a 98 MB soft threshold So Joe s use of java commandline CommandLine will look as follows java commandline CommandLine jjones trumpet 10 0 111 212 volume_snapshot X Rays XRayReview 98 100 review volume for xray storage 10 Deleting a Snapshot Delete a snapshot using java commandline CommandLine java commandline CommandLine lt admin name gt lt admin password gt lt manager ip gt volume_delete lt snapshot name gt lt failure timeout seconds gt Example Joe Jones plans to retain the snapshot for a review period so he writes a script to delete the snapshot after 5 weeks java commandline CommandLine jjones trumpet 10 0 111 212 volume_delete XRayReview 45 Assigning a LUN Number to a Fibre Channel Volume or Snapshot MPC DataFRAME 440 Storage System only A Fibre Channel volume or snapshot must have a LUN number assigned before it can be mounted Assign a LUN number as follows java commandline CommandLine java commandline CommandLine lt admin name gt lt admin password gt lt manager ip gt FC_LUN_set lt volume name gt lt auth group gt lt LUN gt lt failure timeout seconds gt Example Joe Jones is using Fibre Channel for a volume and snapshots for MRI images and he wants to mount the snapshots for backing up So first he assigns a LUN number java commandline CommandLine jjones trumpet 10 0 111 224 FC LUN set MRI604 FC auth 2 lt f
385. stem _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp sysconfig opt Storage System Storage System _Software 6 1 Drivers EBSD 2 4 21 4 ELsmp sysconfig ebsd opt Storage System Storage System Software 6 1 Drivers EBSD install sh MPC DataFRAME 440 Storage System Software User Manual 319 Using the EBSD Driver for Linux Uninstalling the RPM To uninstall the ebsd rpm package from the system run rpm e ebsd For example rpm e ebsd To verify that the package has been uninstalled successfully run rpm q ebsd This returns the message package ebsd 6 x xx xxxx is not installed Using the Source RPM Package to Build a Driver for a New Kernel Version If the kernel you have is not currently supported in the binary RPM package you can create the EBSD driver using a source RPM package Prerequisites e Have previously installed the binary RPM package e Have the kernel source installed on the system e Go to usr src and make a link to the kernel source directory called linux for example cd usr src ln s linux linux Determining the Appropriate Source RPM Package 1 Navigate to the SRPM directory on the CD 2 Find the closest lower version to the target kernel If you have multi processors select the smp version For example For kernel version 2 4 20 select source RPM 2 4 19 For kernel version 2 4 28 select source RPM 2 4 21 Installing the Source RPM Package 3 Run rpm ivh ebsd Release
386. synchronized stop the virtual manager to return to the disaster recovery configuration Note If the downed site is not recoverable you can create a new site with new DataFRAMEs and reconstruct the cluster Call your technical support representative Two site Scenario Communication Between the Sites is Lost In this scenario the sites are both operating independently On the appropriate site depending upon your configuration select one of the DataFRAMEs and start the virtual manager on it That site then recovers quorum and operates as the primary site Once communication between the MPC DataFRAME 440 Storage System Software User Manual Disaster Recovery Using A Virtual Manager sites is restored the managers in both sites reestablish communication and they ensure that the data in both sites is resynchronized When the data is resynchronized stop the virtual manager to return to the disaster recovery configuration Starting a Virtual Manager A virtual manager must be started on a DataFRAME ideally one that is not already running a manager However if necessary you can start a virtual manager on a DataFRAME that is already running a manager Figure 133 shows a management group with a down manager 1 Click the DataFRAME on which you want to start the virtual manager 2 From the Tasks menu on either the Details tab or the Management Group Information tab select Start Virtual Manager shown in Figure 133 EE MPC Computers
387. t groups with one DataFRAME in each group 1 Log in to the management group for which you want to create a cluster The management group Tab View opens 2 Click the Clusters tab The Clusters tab opens shown in Figure 135 GH rc computers DataFRAME Management Console loj x File Find Tasks View Help Available DataFRAMEs Wo Group 0 DF420 1 DF140 DF140 1 DF420 Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Name Clusters tab Figure 135 Viewing the Clusters Tab 3 From the Tasks menu click New Cluster The New Cluster window opens with the General tab on top shown in Figure 136 MPC DataFRAME 440 Storage System Software User Manual 195 Working with Clusters x New Cluster General iSCSI Cluster Name Cluster 0 Description DataFRAMEs Available DataFRAMEs in Cluster DataFRAME IP_ RAD DataFRAME IP RAD DF140 1 192 168 RAID O DF140 192 168 RAID O DF420 192 168 RAID O Q Add ma Q Add all A Remove gt v GC Remove All Hot Spare None Hot Spare Time Out 0 Seconds Cancel Figure 136 Creating a New Cluster 4 Type a meaningful name for the cluster A cluster name is case sensitive and must be from 1 to 127 characters 5 Optional Type a description of the cluster 6 Select one or more DataFRAMEs from the
388. t the power supplies Fibre Channel Items MPC DataFRAME 440 Storage System only Lists Fibre Channel device model channel number and status Cache Battery Items Status information about the batteries IDE Statistics Lists the drive number and capacity MPC DataFRAME 440 Storage System Software User Manual 141 Reporting Saving Log Files 142 If Technical Support requests that you send a copy of a log file the Log Files tab is where you can save that log file as a text file The Log Files tab lists two types of logs e Log files that are stored locally on the DataFRAME displayed on the left side of the tab e Log files that are written to a remote log server displayed on the right side of the tab Note Save the log files that are stored locally on the DataFRAME For more information about remote log files see Remote Log Files on page 143 1 Select Reporting from the configuration categories The Reporting Window opens Select the Log Files tab The Log Files window opens shown in Figure 99 E Edit Configuration for DF140 192 168 0 7 x Module j Passive LogFiles Active Email Diagnostics Alerts W Storage Time ce TCP IP Network Choose logs to save Remote logs R A i Greg E Alerts log Type Destination L A Reporting auth error auth info boot info cron error cron info daemons error daemons info dmesg gossi
389. tart with the first 64 bit slot Adding Fibre Channel Ports MPC DataFRAME 440 Storage System only When you add a card containing Fibre Channel ports to a PCI slot the Fibre Channel ports do not display on the TCP IP tab of the Network configuration window You can view the status of the Fibre Channel ports and the unique World Wide Name WWN of each port in the Passive Report 1 Select Reporting from the configuration categories The Reporting window opens 2 Click Refresh to display statistics on the Passive tab 3 Scroll down to the Fibre Channel statistics The Node WWN is the same for all ports in a management group The Port WWN is unique for each port 82 MPC DataFRAME 440 Storage System Software User Manual Managing the Network edit Configuration for DF440 1 192 168 0 40 xj Module T Storage Time TCP IP Network an DataFRAME Adminis g SNMP oqe IS d Passive LogFiles Active Email Diagnostics Alerts Hem Value Passive Report Host Name IP Number DataFRAME Software GUI Software the same for all ports in the management group e Figure 57 Viewing the WWN of a Fibre Channel Port upport Key Boot Devices flash 0 flash 1 NIC Data Card Description Mac Address Address Mask 01 03 2002 04 00 08 AM MST DF 440 1 192 168 0 40 Version 6 3 41 0009 0 Version 6 3 42 Build 2605 unknown call support Active Acti
390. tations of Remote Copy Copying the Primary Snapshot to the Remote Snapshot When the primary snapshot is copying to the remote snapshot the Console depicts the process with a moving graphic of pages from the primary to the remote snapshot as illustrated in Figure 215 The pages move in the direction of the data flow from primary to remote snapshot u Remote_Backup a a Transaction_Data a a Transaction Data Backup a Transaction_Data Boulder 1 Boulder 2 Rem VolCDO04 Copying from the primary snapshot to rimary Snapshot the remote snapshot Remote snapshot Figure 215 Icons Depicting the Primary Snapshot Copying to the Remote Snapshot Graphical Legend for Remote Copy Icons The graphical legend available from the Help menu depicts the icons associated with Remote Copy The following illustration displays the Remote Copy states icons from the graphical legend Remote Copy Snapshot is primary A Remote snapshot pair the Console graphics from the fer Graphical Legend in God Figure 216 Icons for Remote Copy as Displayed in the Graphical Legends Window 332 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Remote Copy and Volume Replication Remote Copy is asynchronous replication of data Volume replication is synchro nous replication Volume replication is described in detail in the Chapter 12 Working with Volumes Using synchron
391. tem Console Load the Storage System MIB in the Network Management System as outlined below 1 Load STORAGE SYSTEMS GLOBAL REG 2 Load STORAGE SYSTEMS DataFRAME COMMON MIB 3 The following MIB files can be loaded in any sequence STORAGE SYSTEMS DataF RAME COMMON DNS MIB STORAGE SYSTEMS DataF RAME COMMON CLUSTERING MIB STORAGE SYSTEMS DataF RAME COMMON INFO MIB STORAGE SYSTEMS DataF RAME COMMON NETWORK MIB MPC DataFRAME 440 Storage System Software User Manual 135 Using SNMP STORAGE SYSTEMS DataF RAME COMMON NIS MIB STORAGE SYSTEMS DataFRAME COMMON NOTIFICATION MIB STORAGE SYSTEMS DataF RAME COMMON NTP MIB STORAGE SYSTEMS DataF RAME COMMON STATUS MIB STORAGE SYSTEMS DataF RAME COMMON STORAGE MIB Note Any variable that is labeled Counter64 in the MIB requires version 2c or later of the protocol Note Other standard version 2c compliant MIB files are also provided on the resource CD Load these MIB files in the Network Management System if required Disabling the SNMP Agent Disable the SNMP Agent if you do not plan to use SNMP applications to monitor your network of DataFRAMEs 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select SNMP from the configuration categories The SNMP General window opens shown in Figure 91 Disabling SNMP 1 On the SNMP Genera
392. tem Software User Manual Getting Started e The States tab shown in Figure 4 displays the icons used to depict states that the items are in For example when you are logged into a DataFRAME a pink square displays underneath the DataFRAME When an item such as a DataFRAME or a cluster is selected it displays a yellow outline RIK x Items States Hardware Big Icons Small Icons DataFRAME is manager DataFRAME is virtual manager DataFRAME is logged in d DataFRAME is hot spare S Virtual manager not started EI E Volume has schedule KI Volume schedule s has failed bi Snapshot is primary D Snapshot is remote fo Remote snapshot pair AB Figure 4 Graphical Legend States Tab from the Help Menu MPC DataFRAME 440 Storage System Software User Manual Getting Started e The Hardware tab shown in Figure 5 displays the icons that represent the physical storage units x Items States Hardware Big Icons Small Icons DF 100 Series m e DF 400 Series Wuer KE Unknown P Ka Figure 5 Graphical Legend Hardware Tab from the Help Menu Finding Storage System Modules on the Network After opening the Console you must find the DataFRAMEs you want to manage Find these modules by one of two methods e Use a mask to search subnets to find all available DataFRAMEs on a network See Finding by Subnet and Mask on page 21 for more information about completing
393. tes a snapshot of that volume 5 Configure a new snapshot schedule on the new primary volume 6 Reconfigure scripts for failover on the application servers MPC DataFRAME 440 Storage System Software User Manual Remote Copy Setting Up a New Production Site Setting up a new production site involves creating a new primary volume and syncing up the acting primary volume before returning it to its original state as a remote volume The steps are the same as those for creating a new primary volume at the original production site Making the Backup Site into the New Production Site Turn the backup site into the new production site and designate a different backup site The steps are similar to those for initially configuring Remote Copy 1 Create a remote snapshot or a remote snapshot schedule on the acting primary volume 2 Make a new remote volume on the new backup site as part of creating that remote snapshot or remote snapshot schedule 3 Reconfigure scripts for failover on the application servers Rolling Back Primary and Remote Volumes Rolling back a volume from a snapshot is the method for reverting to an earlier copy of the data on a volume Rolling back destroys any snapshots that were created after the snapshot that is rolled back to Rolling Back a Primary Volume Rolling back a primary volume to a primary snapshot replaces the original primary volume with a read write copy of the selected primary snapshot The new
394. text Name of the management group containing the source volume snapshot name Value text Name of the snapshot or volume created using commandline CommandLine auth group name Value text Name of the authentication group associated with the volume in the Console local ip Value IP address The IP address of the machine the script is running on number of managers Value number The number of managers in the management group that contains the source volume managers ip Value IP address Separator space The IP addresses of the DataFRAME managers in the management group lock_mode Value ro rw The attribute of the volume read only or read write Scripted Commands for Volumes and Snapshots Below are examples of the Storage System Software functions that can be accomplished using application based scripts Creating a Snapshot Create a snapshot using commandline CommandLine commandline CommandLine lt admin name gt lt admin password gt lt manager ip gt volume_snapshot lt source volume name gt lt snapshot name gt lt soft threshold 258 MPC DataFRAME 440 Storage System Software User Manual Working with Scripting Megabytes gt lt hard threshold Megabytes gt lt description gt lt failure timeout seconds gt Example Joe Jones is creating a snapshot for his management group Images volume named X Rays and he wants the snapshot name to be XRayReview The
395. th volumes in a management group are created and managed from the management group See Chapter 15 Controlling Client Access to Volumes for more information Information provided on the Authentication Groups tab includes e Name of the authentication group e Authentication mode of the group e The subnet mask of the authentication group if the authentication is set up for subnet and mask e Volume lists associated with the group Volume Lists Tab Volume lists are created at the management group level and provide the connection between authentication groups and volumes Information provided on the Volume Lists tab includes e Name of the volume lists e Name of all volumes within the volume lists e Authentication group associated with the volume list Register Tab Register to use add on modules available for specialized storage features See Chapter 16 Feature Registration for more information about registering add on modules Information available on the Register tab includes e Version information about the software components of the system The version information is provided for customer support should you ever have a support issue e License information for all the DataFRAMEs in the management group Times Tab Resynchronize the management group time any time you change the time ona DataFRAME in the management group that is running a manager Note Use NTP to ensure closely synchronized times on the DataFRAM
396. the List Select a management group from the list and click OK 180 MPC DataFRAME 440 Storage System Software User Manual 10 Disaster Recovery Using A Virtual Manager A virtual manager provides disaster recovery for two specific system configurations A virtual manager is a manager that is added to a management group but is not started on a DataFRAME until it is needed to regain quorum See Managers and Quorum on page 158 for detailed information about quorum fault tolerance and the number of managers Virtual manager is part of the add on module Scalability Pak See Chapter 16 Feature Registration for information about add on modules and registering features The following are definitions of the terms used in this chapter e Virtual Manager A manager which is added to a management group but is not started on a DataFRAME until a failure in the system causes a loss of quorum The virtual manager is designed to be used in specific system configurations which are at risk for a loss of quorum e Regular Manager A manager which is started on a DataFRAME and operates according to the description of managers found in Managers on page 158 e Manager Either a virtual manager or a regular manager When to Use a Virtual Manager Use a virtual manager in the following configurations e A management group across two sites Using a virtual manager allows continuing operation by one site if the other site fails The
397. the file system takes over The file system reviews its available blocks list and reuses blocks that have been freed up Planning Hard Thresholds The hard threshold is the amount of application data that can actually be written to the volume This size is the actual physical space reserved for data on the disks in the cluster Therefore it is the limit beyond which data can no longer be written to the volume The hard threshold can be increased up to the volume size if they are not set as equal Best Practice if Not Using Snapshots For volumes that will not be used with snapshots hard thresholds should be set equal to the volume size This setting ensures that the hard threshold cannot be exceeded which prevents clients from accessing the volume If you intend to use snapshots see Managing Capacity Using Volume and Snapshot Thresholds on page 237 Best Practice if Using Snapshots For volumes that will be used with snapshots set the hard threshold size less than the volume size Next set the soft threshold less than the hard threshold MPC DataFRAME 440 Storage System Software User Manual 213 Working with Volumes Planning Snapshots Snapshots take up space on the cluster Planning how much space and planning the use and scheduling of snapshots impacts the hard threshold you should set for the volume Note Volume size volume thresholds and using snapshots should be planned in conjunction If you intend to use snaps
398. the management group You then have five managers for fault tolerance See Managers on page 158 MPC DataFRAME 440 Storage System Software User Manual 185 Disaster Recovery Using A Virtual Manager Add DataFRAMEs to the Cluster in Alternating Order Warning Create the cluster When adding DataFRAMEs to the cluster add them in alternating order as shown in Figure 130 The order in which the DataFRAMEs are added to the cluster determines the order in which copies of data are written to the volume Alternating the addition of DataFRAMEs by site location ensures that data is written to each site as part of the 2 way replication you configure when you create the volume See Creating a Cluster on page 195 Add DataFRAMEs to the cluster in the following order 1 1st DataFRAME Denver 1 2 2nd DataFRAME Boulder 1 3 3rd DataFRAME Denver 2 4 4th DataFRAME Boulder 2 If DataFRAMEs are added to the cluster in any order different than alternating order by site you will not have a complete copy of data on each site xi New Cluster if General iSCSI Cluster Name Cluster_ Description DataFRAMEs Available DataFRAMEs in Cluster DataFRAME Ip RAD DataFRAME IP RAID DF140 1 192 168 RAID 0 DF140 192 168 RAID 0 DF 420 192 168 RAID 0 Q Add lt Add All a Remove gt Ly Z Remove All gt gt Hot Spare None r Hot Spare Time Out 0 Seconds Cancel
399. the management group and moves to the Available group A placeholder or ghost DataFRAME remains in the cluster shown in Figure 145 It is labeled with an IP address instead of a host name Gmc computers DataFRAME Management t Console 5 x File Find Tasks View Help Available DataFRAMEs Group_0 DF 420 1 Figure 145 Viewing the Ghost DataFRAME Replace the disk in the DataFRAME After you replace the disk you must power the disk on and reconfigure RAID See Replacing a Disk on page 75 Add the repaired DataFRAME to the management group The DataFRAME returns to the management group and the ghost DataFRAME is still in the cluster shown in Figure 145 Figure 146 Returning the DataFRAME to the Management Group MPC DataFRAME 440 Storage System Software User Manual 207 Working with Clusters 7 Edit the cluster and add the repaired DataFRAME to the cluster Warning The repaired DataFRAME must be returned to the cluster in the same place it originally occupied to have the cluster resync rather than restripe To return the repaired DataFRAME to the cluster in the original order 8 In the Edit Cluster window shown in Figure 147 remove any DataFRAMEs in the list that are below the ghost DataFRAME The removed DataFRAMEs return to the DataFRAMEs Available column xi Ee ost DataFRAME as place Repaired holder DataFRAME
400. the original settings such as those in Figure 37 Powering Off the DataFRAME Powering off the DataFRAME through the Console physically powers off the DataFRAME The Console controls the power down process so that data is protected 1 Log in to the DataFRAME 2 Select the Power Off tab The Power Off window opens shown in Figure 37 3 Select Power Off The button changes to Power Off 4 In the minutes field type the number of minutes before the powering off should begin You can enter any whole number greater than or equal to 0 If you enter 0 the DataFRAME will power off as soon as you complete step 6 5 Click Power Off A confirmation message appears 50 MPC DataFRAME 440 Storage System Software User Manual Working with DataFRAMEs 6 Click OK The DataFRAME will power down in the specified number of minutes Note For information about powering off the module manually see the Technical Product Specification TPS provided with the DataFRAME Registering Features for a DataFRAME Using the Feature Registration tab you can register individual DataFRAMEs for add on modules and applications such as Remote Copy You can also register DataFRAMEs when they are in a management group For detailed information about registering see Registering Features and Applications on page 296 Using the Feature Registration Tab The Feature Registration tab as shown in Figure 39 displays the following information
401. there is a 30 second delay and then Motherboard PortO becomes the active interface Slot1 PortO returns to the Passive Ready state When the preferred interface comes back up there is a 30 second delay before it becomes active Table 14 DataFRAME Active Backup Failover Scenario and Corresponding NIC Status Example Failover Scenario NIC Status 1 Active backup bond is created The Bondo is the master logical active preferred interface is interface Motherboard Porto e Motherboard Port0 is Active e Slot1 PortO is connected and is Passive Ready MPC DataFRAME 440 Storage System Software User Manual 87 Managing the Network Table 14 DataFRAME Active Backup Failover Scenario and Corresponding NIC Status Example Failover Scenario NIC Status 2 Active interface fails BondO detects the ge Motherboard PortO status becomes failure and Slot1 PortO takes over Passive Failed Sloti PortO status changes to Active 3 The Motherboard Port0 link is restored Motherboard PortO status changes to Active after a 30 second delay e Sloti PortO status changes to Passive Ready Summary of NIC Status During Failover Table 15 shows the states of Motherboard PortO and Slot1 PortO when configured for Active Backup Table 15 NIC Status During Failover with Active Backup Failover Status Status er Status of Slot1 Port Normal Operation Preferred Yes Preferred No Status Active
402. thresholds e Planning data replication and data priority e Creating and managing volumes Prerequisite Before you create a volume you must have created a management group and at least one cluster See Chapter 9 Working with Management Groups and Chapter 11 Working with Clusters Planning Volumes Planning volumes takes into account multiple factors e How many volumes do you need e What type of volume are you creating primary or remote e What size do you want the volume to be e Do you plan to use snapshots e Do you plan to use data replication e Do you plan to grow the volume or to keep it the same size Note If you plan to mount file systems create a volume for each file system you plan to mount You can then grow each file system independently Working with Volumes Planning Volume Type e Primary volumes are volumes used for data storage e Remote volumes are used with Remote Copy for business continuance backup and recovery and data mining migration configurations See the Remote IP Copy User Manual for detailed information about remote volumes Planning Volume Size Volume size is the size of the virtual device communicated to the operating system and the applications Volume size falls into one of three categories e Volumes that are smaller than the storage capacity of the cluster e Volumes that are equal in size to the storage capacity of the cluster e Volumes that are larger than the storage
403. time stamp on volumes and snapshots You must set the DataFRAME time zone whether you set the time of day manually or use NTP e Using Network Time Protocol NTP Configure the DataFRAME to use an external time service e Setting Date and Time Set the date and time on the DataFRAME if not using an external time service Reset Management Group Time If you change the time on a DataFRAME that is running a manager you must reset the management group time If the management group time is different than a manager DataFRAME you run the risk of inconsistent or unexpected creation time stamps on volumes and snapshots and also that scheduled snapshots will not start at the intended time See Resetting the Management Group Time on page 171 Getting There 1 On the Network View double click the DataFRAME and log in if necessary The DataFRAME Configuration window opens 2 Select Time from the DataFRAME configuration categories The Time window opens shown in Figure 75 Setting the Date and Time fe Edit Configuration for DF140 192 168 0 7 es Module 20 Storage gg TCP IP Network DataFRAME Administ x SNMP Reporting if Log Out NTP Mode NTP is off To turn NTP on add an NTP server to the NTP tab below Time Zone Date and Time NTP Set the time zone for the physical location of the DataFRAME DataFRAME Time Zone GMT Ss Figure 75 Setting the Time Zone
404. ting Active Monitoring Variables Setting Notification Methods for Monitored Variables Use Edit to configure notification methods and change the frequency that the variable is monitored if allowed Configuring Notification Methods for All Variables You can configure alert actions for one variable and then apply those settings to all the variables in the list Use the Apply Threshold Actions button to globally apply the settings Then you can customize alerting actions for a particular variable by editing that variable 1 Select from the list the variable you want to edit 2 Click Edit The Configure Variable wizard opens to Step 2 seen in Figure 103 Note For some variables only the notification method can be changed For example the frequency for the motherboard temperature variable is set to 15 seconds and cannot be changed 3 Optional If allowed change the frequency for the variable and click Next 4 Optional Change the alert notification method 5 Optional To apply the alert actions including the email addresses that you selected in step 4 to all variables that are monitored on the DataFRAME select the Apply Threshold Actions to All Monitored Variables checkbox 6 Click Finish Note If you are requesting email notification be sure to set up the SMTP settings on the Email tab MPC DataFRAME 440 Storage System Software User Manual 145 Reporting Removing a Variable from Active Monitoring Use Remove t
405. ting correctly with all managers in the management group Requirements 110 Each time you update the list of managers you must reconfigure application servers that use the management group to which this DataFRAME belongs Only update the list mode if you have reason to believe that there is a problem with the communication between the other managers in the group and the manager on this DataFRAME 1 Select TCP IP Network from the DataFRAME configuration categories 2 Click the Communication Mode tab to bring it to the front shown in Figure 74 3 Click Update MPC DataFRAME 440 Storage System Software User Manual Managing the Network The list is updated with the current DataFRAME in the management group and a list of IPs with every manager s enabled network interfaces A window opens which displays the IP addresses in the management group and a reminder to reconfigure the application servers that are affected by the update Note For more information on unicast see Communication Mode on page 159 MPC DataFRAME 440 Storage System Software User Manual 111 Managing the Network 112 MPC DataFRAME 440 Storage System Software User Manual 5 Setting the Date and Time The DataFRAME DataFRAME uses the date and time settings to create a time stamp when data is stored You must set the date and time on each DataFRAME e Setting the Time Zone Set the time zone where the DataFRAME is located This time zone controls the
406. tion based scripting for taking snapshots Using application based scripts allows automatic snapshots of a volume and automatic increases in the volume thresholds Scripting also provides access to Remote Copy the ability to maintain multiple copies of data across multiple facilities See Chapter 13 Working with Snapshots for detailed information about snapshot requirements Information about Remote Copy can be found in the Remote IP Copy User Manual The tasks supported by scripting includes e Taking a snapshot of the volume e Mounting the snapshot es Optional Unmounting or deleting the snapshot e Increasing volume thresholds Two tools named commandline CommandLine and ebsdvm may be provided by your vendor to access the Console functionality Tools for Scripting Two software tools are available to use in scripts The first one java commandline CommandLine is used to create and delete snapshots and to automatically increase volume thresholds The second one ebsdvn is used to mount the snapshot Java commandline CommandLine Java commandline CommandLine is the program that actually invokes the snapshot function in the Console for creating and deleting snapshots In addition the program can respond when a soft threshold is reached on a volume and automatically increase the hard and soft thresholds on that volume 1 Set the environment Table 45 Setting the Environment for Using Scripting Tools Operating
407. to a Primary Volume and Back to a Remote Volume Convert a remote volume into a primary volume to gain read write access to the most recently completed Remote Copy snapshot However if that remote volume is the target for scheduled remote snapshots those snapshots cannot take place if the remote volume is not present Therefore you use the operation for returning the primary volume back to its remote status to allow the scheduled remote snapshots to continue Make Remote Volume into Primary Volume java commandline CommandLine lt remote admin name gt lt remote admin password gt 262 MPC DataFRAME 440 Storage System Software User Manual Working with Scripting lt remote manager ip gt volume_make primary lt remote volume name gt lt soft quota Megabytes gt lt hard quota Megabytes gt lt failure timeout seconds gt Make Primary Volume into Remote Volume java commandline CommandLine lt primary admin name gt lt primary admin password gt lt primary manager ip gt volume_make remote lt primary volume name gt lt snapshot name gt lt snapshot description gt lt failure timeout seconds gt Example Joe has scripted an operation to make his remote volume into a primary volume once a week so that he can access the data from the most recently completed scheduled snapshot Since he is running scheduled remote snapshots to that volume he then needs to convert that primary volume back into a remote volume so that the
408. ule The recurrence schedule can be in minutes hours days or weeks 9 Set a retention schedule The retention schedule can be for specified number of snapshots or for a designated period of time 10 Click OK The New Snapshot Schedule window closes and the new snapshot schedule appears on the tab shown in Figure 164 248 MPC DataFRAME 440 Storage System Software User Manual Working with Snapshots Remote Snapshot Schedules Target Information Volume List Memberships Details Snapshots Snapshot Schedules Hosts Remote Snapshot Snapshot Sched Hard Threshold Soft Threshold StartAt RecurEvery Retain Snap_Sch_0O 1 GB 1 GB 04 30 2005 11 0 1 Days 5 Max Weekly_SS 2 GB 2 GB 05 01 2005 06 0 1 Weeks 3 Max Figure 164 List of Scheduled Snapshots Editing Snapshot Schedules You can edit everything in the snapshot schedule except for the name 1 Select the volume for which you want to edit the snapshot schedule The volume Tab View opens 2 Click the Snapshot Schedules tab to bring it to the front 3 Select the schedule you want to edit 4 From the Tasks menu select Edit Schedule The Edit Snapshot Schedule window opens shown in Figure 165 MPC DataFRAME 440 Storage System Software User Manual 249 Working with Snapshots Edit Snapshot Schedule Edit Snapshot Schedule Name Description Snapshots Schedule Name Number Snapshot Schedule Time Zone M
409. ule you use the DataFRAME Configuration window shown in Figure 25 34 MPC DataFRAME 440 Storage System Software User Manual Getting Started Ei Edit Configuration for DF140 192 168 0 7 xj e A Power Off Feature Registration Boot Devices Time Module Information Backup amp Restore TCP IP Network S DataFRAME Adminis A Configuration Tabs fe SNMP Wm Reporting a Host Name bag Apply IP Address 192 168 0 7 on igura ion Model DF 100 Series Categories Serial Number 00 30 48 2E BA 5A Software Version 6 3 41 2605 0 Install Software Logged In User admin Change Password Figure 25 DataFRAME Configuration Window The left pane of the configuration window lists the DataFRAME configuration categories Within each category is a set of tabs used to configure different functions Module Upgrade the Storage System Software change the password or host name perform backup and restore of the DataFRAME configuration reboot or shut down the DataFRAME and manage boot devices See Chapter 2 Working with DataFRAMEs on page 37 Storage Manage RAID and manage individual drives including powering them on or off and reviewing drive information See Chapter 3 Storage on page 57 Time Use NTP or manually set the time zone date and time for the DataFRAME See Chapter 5 Setting the Date and Time on page 113 TCP IP Network Spe
410. ule snapshots The volume Tab View opens 2 Click the Snapshot Schedules tab to bring it to the front 3 From the Tasks menu select New Schedule The New Snapshot Schedule window opens shown in Figure 163 MPC DataFRAME 440 Storage System Software User Manual 247 Working with Snapshots New Snapshot Schedule New Snapshot Schedule Name Snap_Sch_0 Description Snapshots Schedule Name Number Snapshot Schedule Time Zone Greenwich Mean Time Start At 07 07 2005 12 19 20 AM V Recur Every 1 Thresholds for Scheduled Snapshots Size 8 0 Hard Threshold 3 Soft Threshold 3 0 Retention for Scheduled Snapshots Retain Maximum of 5 Snapshots Retain Snapshots For Figure 163 Creating a Snapshot Schedule 4 Type a name for the snapshots The name will be used with sequential numbering For example if the snapshot name is Backup the list of scheduled snapshots will be named Backup1 Backup2 Backup3 5 Optional Enter a snapshot description 6 Optional Change the hard and soft thresholds for the snapshots Note Setting the hard threshold smaller than the size of the original volume allows you to create snapshots that require less space on the cluster See Managing Capacity Using Volume and Snapshot Thresholds on page 237 7 Enter a start date and time The date and time must be valid but they can occur in the past 8 Select a recurrence sched
411. ume Primary Storage Volume Remote Primary Snapshot Snapshot Snapshot Figure 243 High Availability Configuration During Failback Best Practices Use remote snapshots in conjunction with local synchronous volume 370 replication Using remote snapshots alone any data written to the primary volume since the most recent remote snapshot was created will be unavailable if the primary volume is unavailable However you can lessen the impact of primary volume failure by using synchronous volume replication Volume replication allows you to create up to 3 copies of a volume on the same cluster of DataFRAMEs as the primary volume The only limitation is that the cluster must contain at least as many DataFRAMEs as replicas of the volume Replicating the volume within the cluster ensures that if MPC DataFRAME 440 Storage System Software User Manual Sample Remote Copy Configurations a DataFRAME in the cluster goes down replicas of the volume elsewhere in the cluster will still be available For 3 way replication up to 2 DataFRAMEs can fail For detailed information about volume replication see the chapter on volumes in the LeftHand SAN User Manual for details Example configuration This example illustrated in Figure 244 uses 3 DataFRAMEs per cluster However this scenario can use any number of DataFRAMEs Information about creating clusters and volumes can be found in the LeftHand SAN User Manual e In the production locatio
412. umes Authentication Groups Volume Name Permission Level FC LUN Edit Permission Level Remove Cancel Figure 185 Creating a New Volume List 4 Type aname and description for the volume list The volume list name is case sensitive Adding Volumes to the Volume List 1 Select the Volumes tab shown in Figure 185 2 Click Add to add a volume to the volume list 3 The Add Volume to Volume List window opens shown in Figure 186 Add Yolume to Yolume List xj Add Volume to Volume List Yolume_List_0 Volume Name CreditData05 Permission Level O No Access Read Access Read and Write Access Fibre Channel LUN Figure 186 Adding a Volume to a Volume List 4 Select the volume or snapshot to add to the list 5 Select the access permission level for the volume Characteristics of the permission levels are described below inTable 51 282 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Table 51 Characteristics of Permission Levels Type of Access Allows This No Access Prevents the authentication group from accessing the volume or snapshot Read Access Restricts the authentication group to read only access to the data on the volume or snapshot Read Write Access not Allows the authentication group read and write permissions to the available for snapshots volume 6 F
413. ups Add Administration Groups xj pal mirrata iew_onhy_administrator Cancel Figure 88 Adding a Group to an Administrative User 2 Select one or more groups you want to add 3 Click OK The Create Administrative User window opens 4 Click OK to finish adding the administrative user Sorting Columns in the Administrative Users Window The columns in the Administrative Users window can be sorted in ascending or descending order e Click on the column header to sort e Click again to reverse the sort e The arrow next to the column title indicates which column is the sorted column and whether the sorting order is ascending up arrow or descending down arrow Edit Configuration for DF140 192 168 0 7 x Member of Groups Description full_administrator Administrator Figure 89 Sorting Administrative Users Editing Administrative Users 1 Select DataFRAME Administration from the configuration categories 2 Click the Users tab to bring it to the front 3 Select the user to edit from the list of users 4 Click Edit The Edit Administrative User window opens shown in Figure 90 128 MPC DataFRAME 440 Storage System Software User Manual Administrative Users and Groups User Name 3 40 characters 0 9 a z _ Must begin with a lower case letter Description Administrator 3 40 characters 0 9 a z A Z _ space Must begin with a letter Member Groups ful
414. ure 139 Figure 140 Figure 141 Figure 142 Figure 143 Figure 144 Figure 145 Figure 146 Figure 147 Figure 148 Figure 149 Figure 150 Figure 151 Figure 152 Figure 153 Figure 154 Figure 155 Figure 156 Figure 157 Figure 158 Figure 159 Figure 160 Figure 161 List of Figures Configuring Email Settings for Email Alert Notifications eeecececeeeeeeeeeeeeeeeeeeeeeeeteeeeeaeeenaeeeeeeeees 151 Viewing the List Of Diagnostics 152 Alert Messages Tab on Console Main Wimdow 155 VIEWING E CC 155 Viewing DataFRAMEs Before Creating a Management Group 160 Data RAME Wah TEE 161 Management Group Information Tab 161 Creating a New Management Group 162 List of Manager IP Addresses for Management Group 162 New Management Group with One DatatpAME A 163 elle ENEE EEN 164 Logging in to a Management Group 164 List of DataFRAMEs Running Managers AA 165 Viewing a Management Group in the Conscole nennt 165 Editing a Management Group 169 Adding a DataFRAME to Existing Management Group 170 Starting a Manager 172 Adding Manager IP Addresses to Application Gervers 173 Backing up the Management Group Configuration ccccesceeseeeeeeeeeeeeeneeseaeeeeeeseeeeseaeseeetiaeenaeeteeeen 176 Save Window for Backing up the Management Group Configuration ccceeceseceeeeeeeereeteeeeteeteeee 177 Opening the Configuration Binary File 179 Verifying the Management Group Confiouration 179 Correct Two site Fail
415. ure 68 Viewing the TCP Gate 99 Figure 69 Editing TCP Speed Duplex and Frame Glze eeceesceeeceeeneeeneeseeeeeeeeeeaeeeaeeeeaeeeeeeeeaeeseeeseaeesieeseeeesaaes 101 Figure 70 Adding DNS Servers 0 ccecceeseeeseeeeeeeeeeeeeeeeaeeeeeeeaeecaeeeeaeeceaesaeesaeecsaeeseeeseaeesaaeseeesaeeseeesnaeeeeeseaeeeaeeseaee 104 Figure 71 Adding Network Routing Information 106 Figure 72 Adding Routing Information cceeecceeeseeeeseneeeseneeeeeneeeeeaeeeenaeeeseneeeeeaeessaeeeeeaeeeseaaeeseaeeesesaeeseeneeneaeees 107 Figure 73 Editing Routing Information 0 0 0 eeecceeeeeeeeeeneeeeeaeeeeeneeeeeeneeeeaeeeseneeeeeaaeeseaaeeeseeeeeseaaeeseaaeesenaeeeeeneeeneeees 107 Figure 74 Selecting the Storage System Software Network Interface and Updating the List of Managers 109 Figure 75 Setting the Time Zone E 114 Figure 76 Setting the DataFRAME Date and Time 115 Figure 7 7 Adding an NI P SQrver TTT 116 Figure 78 Viewing the List Of NTP Geners AE 117 Figure 79 Editing an NTP Geer 117 Figure 80 DataFRAME Administration Groups Tab 119 Figure 81 Adding an Administrative Group 121 Figure 82 Adding an Administrative User to a Group 122 Figure 83 Adding Permissions to Administrative Group eccceecceeseeeneeeeeeseeeeeaeeeeeeeaeesaeeeeaeeseaeesaeesaeesieeseeeensaes 123 Figure 84 Sorting Administrative Groupe 124 Figure 85 Editing an Administrative Group 125 Figure 86 Adding Administrative USers A 126 Figure 8
416. ure Scenarios Using Virtual Managers 183 Incorrect Uses of Virtual Manager to Regain Quorum 00 eeceeeeeeeeeeeeeeeeeteeeeseeeseaeeeseeseaeeseeeeteeeenetaaes 184 Adding DataFRAMEs to Cluster in Alternating Site Order 186 Cluster with DataFRAMEs Added in Alternating Order 187 Management Group with Virtual Manager Added AA 188 Starting a Virtual Manager 189 Indicator of the Virtual Manage sssini 189 Viewing the Clusters Tab 195 Creating a New Cluster 25 640 5 cactaiitiisien eee A ML ed 196 Configuring a Virtual IP Address for GC 197 Adding an iSNS Server cccsccessecseceeseeeeseceseeeeecenececaeesaenenaesacesanecsaeseaeseneecsaaeeaeensecesneeseaeeeneeseaneeaees 197 Ee ENEE 198 Viewing a Cluster and the Cluster Tab 199 Statistics Tora CUSTOM EE 200 Disk Space Allocated and Used for the Cluster and its Volumes and Gnapehots A 201 Editing a e UE EE 202 DataFRAME with Failed Disk 207 Viewing the Ghost DataFRAME AAA 207 Returning the DataFRAME to the Management Group 207 Returning the Repaired DataFRAME to the Cluster 208 Write Patterns in 2 way Replication A 216 Up to Ten Automatic Increments for Auto Grow 221 Example Manual Auto Grow Chart 223 Example Automatic Auto Grow Chart 224 Viewing the Volumes Tab ras22icsiseeicdscebensensavantchaadccavaeeceteseeauedeudeve seus lies vanaatsasiseagascts Eai EESE EE Ei ia 225 Creating a New Primary Volume 226 Setting Replication to None AAA 226 Viewing a Volume in a Cluster
417. used on the DataFRAME Volume Total raw space ever used in the volume Snapshot Total raw space ever used in the snapshot and in the writable snapshot Max Full Highest level of block level space ever used e Cluster Percent of raw space allocated to volumes and snapshots DataFRAME Percent of DataFRAME raw space to which data has been written Volume Percent of volume hard threshold raw space ever used e Snapshot Percent of snapshot hard threshold raw space ever used Misc Other information that impacts space usage 200 MPC DataFRAME 440 Storage System Software User Manual Working with Clusters Table 33 Disk Space use Reported on Disk Usage Tab Column Heading Information Reported e Cluster Percent of raw space in the cluster that has been allocated for volumes and snapshots e DataFRAME Whether the DataFRAME has been designated as a hot spare Volume e Replication level for the volume e Snapshot e Replication level for the snapshot Usage Graphs The usage graphs shown in Figure 142 provide a visual display of the cluster usage The display updates in real time so you can easily see changes in the usage in the space allocated for the cluster and in the space used for volumes and snapshots Details Volumes NSMs DiskUsage Usage Graphs iSCSI Remote Snapshot TransData05 Total Alloc 157 45 GB Volumes vs Snapshots Free 92 11 GB Allocated GB Avail 58 a 10 15 20 25
418. valuatton 294 Safely Backing Out of Scripting Evaluation ccccecceceeceeseeeseeeeeeeseeeeeseeeeaeeeeeeeaeeeeeseaeseseeseaeesieeeeeeensaes 296 Parameters in CDSC CONP ssssisisscsvcsstsesescusccasscssatevscancesesasdsssseecancucdsaccucesnseasdueesesucnavenaeecsaeusesraccssensernccenians 321 Parameters for Jproclebsdiclent A 323 Uses f r Remote COPY as ifdescaicttecevessticschaschescda decades eeneg deeg Aaaa REENEN ENEE 333 Remote Copy and Management Groups Clusters Volumes Snapshots and DataFRAME6 334 Snapshot Retention Policy and Maximum Number of Snapshots Hetained nee 337 Remote Copy Planning Checker 338 Values for Remote Copy Details Window nenea 350 Steps to Create Gnapshots ec ceceeeeeeseeeeneeseeeeeeeseeeeeaeeseeseaeeeaeeeeaeeseaeeeaeesaeseaeeseeeseaeeseeeseaeesieeseaeensaes 362 Requirements for Rolling Back a Primary Volume 364 MPC DataFRAME 440 Storage System Software User Manual 1 Getting Started Welcome to the MPC Storage System Console Console The Console is used to configure and manage storage volumes spanning clustered DataFRAMEs After you have installed your DataFRAMEs and have installed the Console on the system administrator s PC you must take certain steps to prepare for creating storage clusters and volumes The Console is the storage administrator s tool for e Configuring and managing the DataFRAME e Creating and managing clusters and volumes This user manual provides instructio
419. ve le is unique for 2 cards 1 each Fibre Channel port Intel Corp 825464 Controller Coppe 00 07 E9 2F AD 74 192 168 0 40 255 255 255 0 E Refresh Save to File C Advanced Configuring the IP Address Manually Use the TCP IP Network category in the DataFRAME configuration window to configure the IP address for an interface Note Any time you change an IP address of a DataFRAME that is running a manager the volumes on the DataFRAME may become inaccessible to hosts configured to access the volume You must reconfigure all hosts that are using that IP address 1 Select TCP IP Network from the DataFRAME configuration categories The window opens with the TCP IP tab on top 2 On the TCP IP tab select the interface from the list for which you want to configure or change the IP address 3 Click Edit The Edit TCP IP Configuration window opens shown in Figure 58 Name Type Description Mac Address IP Address Subnet Mask Edit TCP IP Configuration Motherboard Port1 Intel Corp 82546GB Gigabit Ethernet Controller Copper 00 30 48 2D 6E DA Disable Interface Obtain an IP address automatically using the DHCP BOOTP protocol Sk 10 0 6 133 255 255 240 0 Default Gateway 10 0 1 254 Cancel Figure 58 Configuring the IP Address Manually MPC DataFRAME 440 Storage System Software User Manual 83 Managing
420. vent a group from accessing a volume without deleting the volume list change the volume permissions to No Access See Editing Volume Permission Levels on page 285 1 Log into the appropriate management group 2 Select the Volume Lists tab shown in Figure 192 Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot volume List Name Volumes Authentication es Volume_List_1 Snapshot_O AG_AcctRec Volume_List_2 rem_test AG_TransDataRemote Figure 192 Volume Lists Tab 3 Select the volume list you want to remove 4 From the Tasks menu select Delete Volume List A confirmation window opens 5 Click OK to confirm the deletion The Volume Lists tab no longer displays the volume list Selecting an Authentication Group from the List Select an authentication group and click OK MPC DataFRAME 440 Storage System Software User Manual 287 Controlling Client Access to Volumes Selecting a Volume List from the List Select a volume list and click OK 288 MPC DataFRAME 440 Storage System Software User Manual 16 Feature Registration Add On Features and Applications Registration Overview Add on features and applications expand the capabilities of the Storage System Software Add on features and applications include the following e Scalability Pak e Configurable Snapshot Pak e Remote Data Protection Pak e Client Server Clustering Pak All add
421. virtual manager provides the ability to regain quorum in the operating site if one site goes down or in one selected site if communication between the sites is lost Such capability is necessary if volumes in the management group reside on DataFRAMEs in both locations e A management group in a single location with two DataFRAMEs If you create a management group with two managers in the same location that management group is in a non fault tolerant configuration One manager provides no fault tolerance Two managers also provide no fault tolerance due to loss of quorum if one manager goes down See Managers and Quorum on page 158 for more information Running two managers and adding a virtual manager to this management group provides the capability of regaining quorum if one manager goes down Disaster Recovery Using A Virtual Manager Benefits of a Virtual Manager Running a virtual manager supports implementation of disaster recovery configurations to support full site failover The virtual manager ensures that in the event of either a failure of a DataFRAME running a manager or of communication breakdown between managers as described in the two site scenario quorum can be recovered and hence data remains accessible Requirements for Using a Virtual Manager 182 It is critical to use a virtual manager correctly A virtual manager is added to the management group but not started on a DataFRAME until the management group
422. volume has a different name than the original and the original volume is deleted Prerequisites e Stop applications from accessing the volume e Delete all snapshots that are newer than the snapshot you are rolling back from Warning After rolling back a volume to a snapshot you lose all data that was stored since the rolled back snapshot was created Warning Any uncompleted remote copy snapshot that is newer than the snapshot that you are rolling back to will be cancelled MPC DataFRAME 440 Storage System Software User Manual 363 Remote Copy 1 Log in to the management group that contains the primary volume that you want to roll back 2 Select the snapshot that you want to roll back to 3 Review the snapshot Details tab to ensure you have selected the correct snapshot 4 From the Tasks menu select Roll Back Volume 5 The Roll Back Volume window opens shown in Figure 238 Roll Back Yolume Roll Back Volume New Volume Name Molume 0 Description Cluster Replication Level Replication Priority Size Hard Threshold Soft Threshold Auto Grow Checksum Data Figure 238 Rolling Back a Primary Volume 6 Type anew name for the rolled back primary volume You can also change the hard threshold and soft threshold if necessary Table 64 Requirements for Rolling Back a Primary Volume Item Requirements for Changing New Primary Volume
423. w Authentication Group for iSCSI ACCESS eeceeceeseneeeeeeeneeeeaeeeeeeteaeeteeeenneetaeeeeaeensaes 266 Open the MS iSCSI initiator and Copy the Initiator Node Name to the Initiator Node Name Field 268 Configuring iSCSI shown in the MS iSCSI initiator for a Single Host with CHAP sssssssssseeseessseesseen 269 Adding an Initiator Secret for 2 way CHAP Shown in the MS iSCSI Initiator 269 Creating a New Authentication Group for EBSD ACCESS 0 cecceeceeeseeeeeeeneeeeeeeeeeeeaeeteaeeeaeeteeeeaeensaes 270 Creating a New Authentication Group for Fibre Channel Access 271 Example Configuration for Assigning LUN Numbers AA 272 LUN Numbering Configuration that is NOT Allowed AAA 272 LUN Numbering Configuration with one LUN Shared Among Three Hosts eeeseeeeeeeeeeeeereeeeeeeene 273 Creating a New Authentication Group 274 Creating iSCSI Access in New Authentication Group 275 Configuring EBSD for New Authentication Group 276 Adding a Subnet and Mask for EBSD Host Authentication ceccceeeeeeeseeeeeeeeeeeneeeaeeeeeeeseaeeeeeeeaeenas 277 Configuring Fibre Channel for New Authentication Group ceeceeceeeeeeeeeeeeeeeeeeeeseaeeeeeeenaeetaeeeeaeensaes 278 Adding a Name and WWPN for Fibre Channel Authentication ceececeeceeeeeeeeeeeeeeeeeeeeeaeeeeeeneeees 278 Viewing the Authentication Groups eecceecceeseeeeeeeseeeeeaeeeseeeeaeesaeeseaeesaeesaeseaeeseaeeseeeseaeeeaeesieeeeaeensaes 279
424. which are inherited from the parent volume See Creating Access to Volumes on page 265 For information about authentication groups and volume lists see Chapter 15 Controlling Client Access to Volumes Mounting or Accessing a Snapshot A snapshot is a point in time picture of a volume In order to mount the snapshot for backing up or making the data available for other uses such as data mining or testing you can configure the snapshot as a read write volume Snapshot Writable Space When you configure a snapshot as read write additional space is created in the cluster for use by applications and operating systems that need to write to the snapshot when they access it For example MS Windows performs a write when the snapshot is mounted MS VSS writes to a volume that it is backing up You can see how much writable space is being used for a snapshot on the Disk Usage tab in the Cluster Tab View as shown in Figure 161 MPC DataFRAME 440 Storage System Software User Manual 243 Working with Snapshots Details Volumes DataFRAMEs Disk Usage p Type 1 Name J Space 1 MaxUsed Max Full we Cluster Cluster D 2 61 TB 0 75 0 19 86 Allocated a DataF RAME DF420 1 3 26 TB 0 25 MB DataF RAME DF140 1 889 65 GB 0 25 MB DataF RAME DF420 3 48 TB 0 25 MB Volume Volume 0 OMB Replication Level N Snapshot Snapshot_1 10GB 0MB MB sO MB Max Snapshot Snapshot_2 10 GB 0 MB Ha D Snapshot D Jot Replication Level N
425. y Snapshot_0 AG_AcctRec rem_test AG_TransDataRemoate Figure 188 Viewing the New Volume List Editing a Volume List Edit the volumes in a volume list to e Add a volume e Edit the permissions for a volume e Remove a volume Edit the authentication groups in a volume list to e Add a group to the list e Remove a group from the list Warning Before editing the volume list stop any applications from accessing volumes for which you are restricting permissions or removing authentication groups Opening the Volume List to Edit 1 Log in to the management group that contains the volume list you want to edit 2 Select the Volume Lists tab 3 Select the volume list to edit 4 From the Tasks menu select Edit Volume List The Edit Volume List window opens shown in Figure 189 284 MPC DataFRAME 440 Storage System Software User Manual Controlling Client Access to Volumes Edit Yolume List Edit Volume List Name Description Volumes Authentication Groups _VolumeName Permission Level FCLUN Volume D Snapshot_O Read Edit Permission Levet Remove Cancel Figure 189 Opening a Volume List to Edit Editing Volume Permission Levels Changing the permission level for a volume changes the access rights for the authentication groups that connect to that volume Warning Before editing the volume list stop any applications from accessing volumes for
426. y volume The Console displays the progress of this synchronization MPC DataFRAME 440 Storage System Software User Manual Remote Copy How Remote Copy Works Replicating data using Remote Copy follows a three step process 1 At the production location you create a snapshot of the primary volume this is called the primary snapshot 2 You create a remote volume at the remote location and then create a remote snapshot The remote snapshot is a snapshot of the empty remote volume and it is linked to the primary snapshot 3 The system copies data from the primary snapshot to the remote snapshot 1 System takes a snapshot 2 System creates an empty snapshot of the primary volume of the remote volume Primary Volume es Remote Volume Read Write H Pointer to remote Accessed by q 3 snapshots application server Source for primary snapshots Primary Snapshot Remote Snapshot Primary Location Remote Location 3 System copies data from the primary snapshotto the remote snapshot Figure 214 Basic Flow of Remote Copy Note Both primary and completed remote snapshots are the same as regular snapshots Chapter 13 Working with Snapshots Note Remote Copy can be used on the same site even in the same management group and cluster MPC DataFRAME 440 Storage System Software User Manual 331 Remote Copy Graphical Representations of Remote Copy The Storage System Console displays special graphical represen
427. y is complete e Remote Copy failovers an alert is generated when a remote volume is made primary e Remote management group status an alert is generated if the connection to a remote management group changes disconnects and or reconnects For detailed information about configuring Active Monitoring see Using Active Monitoring on page 144 Monitoring Remote Snapshot Details from the Console Tab View View information about each remote snapshot in both the Remote Snapshot tab and in the Remote Copy Details panel Viewing Information in the Remote Snapshot Tab The Remote Snapshot tab displays a list of remote snapshots connected with a selected item in the Network view For example if you select a management group the Remote Snapshot tab displays the list of remote snapshots associated with that management group You can view lists of remote snapshots by manage ment group cluster volume and snapshot levels 1 Select the appropriate item in the Network view 2 Click the Remote Snapshot tab to bring it to the front shown in Figure 228 348 MPC DataFRAME 440 Storage System Software User Manual Remote Copy Details Managers Clusters Authentication Groups Volume Lists Register Times Remote Snapshot Primary Man Primary Sna Remote Man Rernote Sna Complete Elapsed Time Data Copied Rate State Transaction CredDataRS1 TransData_ May05RemSS 100 3m 14s
428. ze To edit the frame size 1 On the TCP Status tab select the interface you want to edit 2 Click Edit The Edit Speed Duplex and Frame Size window opens shown in Figure 21 x Name Motherboard PortO Speed Duplex Speed Auto Auto Duplex Frame Size Defautt Setto 102 MPC DataFRAME 440 Storage System Software User Manual Managing the Network Table 21 Editing TCP Speed Duplex and Frame Size 3 Select Set To in the Frame Size section 4 Enter a value between 1500 and 9000 bytes in the Set To field 5 Click OK A series of status messages display Then the changed setting displays in the TCP status report Note You can also use the Configuration Interface to edit the frame size See Setting the TCP Speed Duplex and Frame Size on page 306 Using a DNS Server The DataFRAME can use a DNS server to resolve host names For example if you enter a host name to specify an NTP time server the DataFRAME will use DNS to resolve the host name to its IP address For example the time server in Boulder Colorado has a host name of time nist gov DNS resolves this host name to its IP address of 192 43 244 18 DNS and DHCP If you configure the DataFRAME to use DHCP to obtain an IP address and if the DHCP server is configured to provide the IP addresses of the DNS servers then a maximum of three DNS servers will automatically be added to the DataFRAME These DNS servers are listed a

EMA001082-00 - MPCdrivers.com

Contents

Download Pdf Manuals

Related Search

Related Contents