Home

Data Replication Manager HSG80 Version 8.7P Scripting User Manual

1. You will see a display similar to that in Example Display 1 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 11 1 Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode Example Display 1 BuildngATop gt show remote_copy_sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to this controller Not reserved Target state BUILDNGB D1 is NORMAL V N Open a command prompt window on the initiator host v Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 gt 4 Perform the following steps based on the initiator state of your remote copy sets that are in the failsafe error mode a If every initiator states is failsafe locked proceed to step 6 b If every initiator states is not failsafe locked proceed to step 6 c Ifthe initiator states are a mixture of failsafe locked or not being in failsafe locked proceed to step 5 gt 5 Forremote copy sets in failsafe error mode with mixed initiator states change all the remote copy sets in failsafe mode to normal mode Use the CLI command SET RemoteCopySetName E
2. Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 11 File Customization SerNum specifies the device serial number that Command Scripter uses to communicate with initiator controller tulsa As shown in Figure 4 1 this is the serial number of the CCL for Controller A tulsa that is communicating with Host A The maximum limit is three controllers per host One way to find the serial number of a controller CCL would be by using the following steps a Enter the command CMDSCRIPT F SUBSYSDATA A listing of available controller devices will display similar to the following Available devices Device Name Vendor Product ID Serial Num FW csi4 1 0 0 DEC HSG80CCL 2G91412410 V87P Scsi4 1 2 0 DEC HSG80CCL 2G91205687 V87P Scsi5 1 1 0 DEC HSG80CCL 2G91416136 V87P Sesib ls3 0 DEG HSG80CCL 2G91606296 V87P This output display shows CCLs that may be within the same controller subsystem pair For example the first and third device listed above may be tulsa_top and tulsa bot b Cross reference the device serial number to the controller name using the following command CMDSCRIPT F DeviceName SHOW THIS Example cmdscript f Scsi4 1 0 0 show this The resulting display shows the characteristics of the controller beginning with the controller name For example tulsa bot As shown in the first line of the listing in step 1a you now know that the serial number for the bottom controlle
3. From the Windows desktop click Start Click Settings Click Control Panel Double click System For Windows 2000 servers click Advanced then click Environment Variables For Windows NT Server click Environment For Windows 2000 in the System Variables section click New For Windows NT continue with step 5g In the dialog box type CLONE HOME in the Variable Name field In the Variable Value field enter the path to the script default directory for example C scripts For Windows 2000 click OK For Windows NT click Set Click OK until you reach the Control Panel Close the Control Panel Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 2 3 Installation Perl Interpreter A Perl interpreter is necessary to execute the Perl scripts and must be installed on each server that runs the scripts The Tru64 UNIX AIX and Solaris Version 8 platforms have a Perl interpreter included with their operating systems In the OpenVMS Windows NT 2000 and Solaris Versions 6 and 7 environments the interpreter is a component that must be separately obtained and installed For information on how to obtain and install a Perl interpreter for OpenVMS Windows NT 2000 and Sun Solaris 7 see the following sections HP OpenVMS Obtaining the OpenVMS Perl Interpreter The Perl interpreter for the OpenVMS platform can be downloaded from http www sidhe org vmsperl prebuilt html The interpreter
4. SET ERROR CONFIGURED OPERATION STEP 1 Background tulsa drmdispatch PLANNED failsafe revopl ALL NOTFORCED Background greenbay drmdispatch PLANNED failsafe revopl ALL NOTFORCED END SET ERROR CONFIGURED OPERATION STEP SET ERROR CONFIGURED OPERATION STEP2 Background tulsa drmdispatch PLANNED failsafe revop2 ALL NOTFORCED Background greenbay drmdispatch PLANNED failsafe revop2 ALL NOTFORCED END SET ERROR CONFIGURED OPERATION STEP2 C 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Structure of the Application Action List Action Commands Each action in the application action list begins with an action name on the first line and ends on a line with the action name preceded by the string END_ These names must be in capital letters For example SHORT_PLANNED_FAILOVER_STEP1 individual action lines go here END_SHORT_PLANNED_FAILOVER_STEP1 The character as the first character on a line indicates that the rest of that line is a comment Comment symbols placed anywhere other than the first character on a line are not allowed Empty lines are also not allowed An action line is constructed as follows Foreground Background ControllerName PerlScriptName Paraml Param2 Param3 Param4 Refer to Table C 1 for a description of this structure T
5. This section describes how the components work together to perform failover failback or resumption of operation by the use of scripts Perl Scripts The scripts are written in the Perl programming language and reside on the host s local hard drive For redundancy the scripts should reside on a server at both the initiator and target sites User Customized Script Support Files The failover failback and resumption of operation scripts use two user customized file types to provide variable information a configuration file and an application action list e The configuration file tells the failover failback scripts which devices are attached to an HSG80 controller and how the controller is configured with respect to devices and storagesets An example configuration generation file is provided in the DRM Scripting Kit to allow a configuration file to be created for the subsystem Once created customization is needed on the target controller configuration files One current configuration file for each HSG80 controller subsystem is stored at both the initiator and target sites Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 9 DRM Scripting Overview e The application action list is used by the hsg_control pl Perl script to perform failover failback and resumption of operation actions on the specified DRM initiator target controller pairs For example if you have four initiator controllers that need to f
6. To verify that the target hosts can connect to the LUNs use this command SHOW UNITS FULL In the Access field of the display all units that are used by remote copy sets should show that the target hosts connections are enabled This should also show the initiator controller connections 6 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Full Failback Procedure Window Edit Options Help n FULL FAILBACK STEP2 Status start time 10 Sep 2001 14 22 15 end time 10 Sep 2001 14 26 35 Figure 6 1 Operation completion status result display 3 The following step requires actions relative to each operating system in your configuration a HP OpenVMS Allow hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives 2 If the target site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 If the target hosts are not shut down use the following command to recognize the drives hwmgr scan scsi Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 6 3 Unplanned Site Fai
7. dev scp_scsi 6 dev kevm 44 dev disk floppy0c 3 5in floppy fdi0 unit 0 57 dev disk cdrom0c COMPAQ CD 224E bus 2 targ 0 lun 0 58 dev disk dsk0c COMPAQ BC03662379 bus 6 targ 3 lun 0 62 dev cport scpl HSG80CCL bus 1 targ 3 lun 0 63 dev cport scp2 HSG80CCL bus 0 targ 0 lun 0 70 dev disk dsk9c DEC HSG80 IDENTIFIER 1509 71 dev disk dsk10c DEC HSG80 IDENTIFIER 1529 The device names of the CCLs in this example are scpl and scp2 To verify which controller remote copy name uses a CCL enter the command cmdscript f DeviceName show this controller The resulting sample output describes the controller and its remote copy name Controller HSG80 2G02103566 Software V87P Hardware E12 NODE ID 5000 1FE1 0009 0ECO ALLOCATION CLASS 0 SCSI VERSION SCSI 3 Configured for MULTIBUS FAILOVER with 2G02103685 In dual redundant configuration Device Port SCSI address 6 Time 23 OCT 2001 10 33 10 Command Console LUN is lun 0 IDENTIFIER 1500 Host Connection Table is NOT locked Smart Error Eject Disabled Host PORT 1 Reported PORT ID PORT 1 TOPOLOGY 5000 1FE1 0009 0EC1 FABRIC fabric up Address 6F1900 Host PORT_2 Reported PORT_ID 5000 1FE1 0009 0EC2 PORT_2_TOPOLOGY FABRIC fabric up Address 6F1B00 REMOTE COPY HSGO15I Cache 256 megabyte write cache version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER DEFAULT 10 seconds Mirrored Cache 256 megabyte write cache version 0
8. gen ex bat Example file to create a configuration generation file for each controller pair hsg fb1 bat Initiates a Perl script that performs step 1 of a planned site failback on controllers identified in the application action list This file should not be modified hsg fb2 bat Initiates a Perl script that performs step 2 of a planned site failback on controllers identified in the application action list This file should not be modified A 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Kit Files Table A 1 Installed DRM Scripting Kit Files Continued Directory Filename Description hsg_fo bat Initiates a Perl script that performs a two step planned site failover on controllers identified in the application action list This file should not be modified hsg_op bat Runs a Perl script t0 change the remote copy set error mode on the initiator controller This file should not be modified BAT start perl job bat Launches drmdispatch pl from hsgcontrol pl This file should not be modified Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide A 5 DRM Scripting Kit Files Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide B Sample Controller Configuration File This appendix provides an example of an initiator controller configuration file for a DRM initiator target pair of tulsa initiator and fargo target Example of Control
9. FAILOVER STEP1 app fol rc The script searches for a section named PLANNED SHORT FAILOVER STEPI in the app act file and performs all action entries between PLANNED SHORT FAILOVER STEPI and END PLANNED SHORT FAILOVER STEPI drmdispatch pl As hsgcontrol pl performs the actions in the application list it calls the drmdispatch pl script and passes the parameters specified in the application action list The following is the syntax for the Windows format for the script Its structure is shown in Table C 3 CLONE_HOME S BIN drmdispatchl pl ControllerName FailoverType ControlTable RCSProcessed Forced_NotForced ErrorLog Table C 3 Structure of the drmdispatch pl Script Command Variable Description CLONE_HOME BIN The variable pointing to the default directory of the script files and the BIN subdirectory This is the path to the drmdispatch pl script drmdispatch pl The Perl script name ControllerName The name of the controller receiving the action FailoverType Specifies either a PLANNED or DISASTER failover ControlTable Identifies the control table interpreted by the script These control tables have a tbl extension and should not be modified RCSProcessed A list of the remote copy sets to process This should be set to ALL Forced NotForced Indicates whether the script must clear conditions before deleting a unit This should be set to NOTFORCED which requires manually clearing conditi
10. Running the Resumption of Operations Program File Procedure This procedure is used when a connection is re established to the target site The remote copy sets are changed back to failsafe mode and host I O resumes with the target site v 1 Opena command prompt window on the initiator host v 2 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 v Change to the bat sh or vms subdirectory in the CLONE HOME directory V IN Run the hsg_op xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system gt 5 Youwillseea message that you are performing a DRM operation to modify error mode and asks whether normal or configured error mode is desired Enter c for configured error mode gt 6 You will see a confirmation message that asks you to confirm that a configured error mode is desired Enter y for yes gt 7 If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app op rc file from the config subdirectory of the CLONE HOME directory and run the program file gt 8 A normalization occurs When th
11. TMP None Contains step 2 pre full failover instructions read by the armdispatch pl Perl script This file should not be modified Installed as an empty directory but used as a repository for LOG and CHK files generated by scripts The LOG files trace all commands and responses sent to and from the controllers The CHK files contain error messages Installed as an empty directory but used as a repository for temporary files OpenVMS only files Default Hsgvms pm VMS gen_ex com hsg fbi com hsg fb2 com hsg fo com hsg op com Perl module containing library files used by Perl scripts This file should not be modified Example file to create a configuration generation file for each controller pair Initiates a Perl script that performs step 1 of a planned site failback on controllers identified in the application action list This file should not be modified Initiates a Perl script that performs step 2 of a planned site failback on controllers identified in the application action list This file should not be modified Initiates a Perl script that performs a two step planned site failover on controllers identified in the application action list This file should not be modified Runs a Perl script to change the remote copy set error mode on the initiator controller This file should not be modified start perl job com Launches drmdispatch pl from hsgcontrol pm This file should not be modified Data
12. Target Node Name BUILDNGB Target Unit Number 1 00000001 Number of Targets 1 00000001 Remote Copy Set Name RCS1 Association Set Name AS D1 Log Unit Number Not Available nstance Code 0E098901 EVL BuildngATop 06 JUN 2001 12 57 14 Instance Code 02908901 Template 81 51 Occurred on 06 JUN 2001 at 12 57 13 Power On Time 2 Years 72 Days 5 Hours 31 Minutes 56 Seconds Controller Model HSG80 Serial Number 2G84906303 Hardware Version E03 2B Software Version V86 1P Unit Number 1 0001 Unit Software Version 1 01 Unit Hardware Version 53 35 Retry Level 1 Retries 1 Port Te Target 0 LUN Q0 SCSI Device Type 0 00 Device ID BB00911CA0 Device Serial Number V00907W1 Device Software Revision Level 3B05 SCSI Command Opcode 42 2A Sense Data Qualifiers 64 40 SCSI Sense Data Error Code 112 70 current command execution Information field is valid Segment 0 00 Sense Key 2 02 NOT READY ILI 0 EOM 0 FM O Information 6AB80A00 Additional Sense Length 10 0A Command Specific Information 00000000 ASC 4 04 ASCQ 128 80 FRU 0 00 Sense Key Specific 000000 Instance Code 02908901 gt 2 Host I O to the remote copy sets will be paused Verify that the initiator state of the remote copy set is inoperative with unit failsafe locked with the following command SHOW REMOTE_COPY_SETS FULL You will see a display similar to that in Example Dis
13. This command sets both the read and write maximum cached transfer size Repeat this step for each remote copy set unit Enable Port 1 and Port 2 connections to the fabric with the following CLI commands E HIS_CONTROLL T R PORT_1_TOPOLOGY FABRIC ET OTHER_CONTROLLER PORT_1_TOPOLOGY FABRIC I R PORT 2 TOPOLOGY FABRIC E HIS CONTROLL nn n uo ET OTHER_CONTROLLER PORT_2_TOPOLOGY FABRIC Create connections to the remote target controllers Use the CLI commands ADD REMOTE_COPY_SETS RCS199 D199 InitiatorControllerName D199 Example add remote_copy_sets rcs199 d199 buildnga d199 NOTE This command will fail with the error message initiator unit specified not found However it will create and name the connections appropriately Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 13 Unplanned Site Failover with Failback to New Hardware Procedure gt 10 11 12 Set target access to all remote copy units by issuing the following CLI command SET UnitNumber ENABLE ACCESS PATH HostIDs Example set dl enable access path BUILDNGBa BUILDNGBb BUILDNGBc BUILDNGBd Compare the status of the controllers association sets remote copy sets units and connections Refer to the HP StorageWorks Data Replication Manager HSG80 ACS 8 7P Fa
14. When CLONE_HOME is used in a path name in this manual it refers to the name you assigned to the default directory of the script files So if you use C scripts as the default directory in Windows NT 2000 a path name of CLONE_HOME bin would be the same as C scripts bin Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 2 1 Installation HP OpenVMS 1 Create a directory for example SYS DEVICE SCRIPTS to be the default directory for the scripts 2 Copy the scripting kit self extracting file script_vms_v30 exe into the directory created in step 1 3 From the command line prompt enter the command RUN SCRIPT_VMS_V30 EXE The kit files will self extract into the default directory you created 4 Verify that the subdirectories bin config log tmp and vms were created in the default directory See Appendix A for a list and description of the installed files 5 Define the system logical name CLONE_HOME to point to the directory where you have installed the scripts For example S DEFINE SYSTEM EXEC CLONE_HOME SYSSSYSDEVICE SCRIPTS To ensure this logical name is defined after the next system reboot place this definition in the the system startup file SYSSMANAGER SYLOGICALS COM HP Tru64 UNIX IBM AIX and Sun Solaris 1 2 3 Create a directory for example scripts to be a default directory for the scripts Copy the scripting kit tar f
15. drm_create_remotecopyset cannot read configuration entry for remote copy set drm_create_remotecopyset cannot create the remote copy set drm_delete_remotecopyset the remote copy set does not exist drm_delete_remotecopyset the remote copy set is still part of an association set Table D 1 Scripting Error Codes Continued Attempted adding association set five times There is a controller problem with adding the association set Identify and correct controller problem Possible mismatch between configuration file and controller Verify that the configuration file accurately reflects configuration of controller Association set does not exist on controller but does exist in configuration file Update configuration file Name conflict exists Verify and correct the configuration filename Cannot find entry of remote copy set in configuration file Update the configuration file Cannot run add remote command Check for possible communications errors and repair The remote copy set does not exist on the controller Update the configuration file Mismatch between the configuration file and controller configuration Configuration file indicates remote copy set should be part of the association set but controller indicates it is not part of the association set Update configuration file Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Troubleshooting Table D 1 Script
16. example B 1 audience ix B benefits of scripts 1 8 button names convention defined xi C caution symbol and definition xi CLI commands 1 1 1 6 CLONE HOME 2 1 2 3 4 3 to 4 18 5 1 C 1 Command Console LUN 1 14 4 3 to 4 15 5 2 HP OpenVMS 3 1 4 4 HP Tru64 UNIX 3 2 4 6 IBM AIX 3 4 47 Microsoft Windows NT 2000 3 5 4 11 Sun Solaris 3 5 4 13 command names convention defined xi Index Command Scripter See SANworks Command Scripter compatible operating systems 1 15 condensed status display 5 3 configuration file See controller configuration file configuration generation files HP OpenVMS 4 3 4 18 HP Tru64 UNIX 4 6 4 18 IBMAIX 4 7 4 18 Microsoft Windows NT 2000 4 9 4 18 running 4 17 Sun Solaris 4 13 4 18 confirmation messages D 9 Connections Section customization 4 21 example B 1 control table 1 10 1 13 controller configuration file customization 4 19 defined 1 9 process flow 1 13 sample B 1 conventions application names defined xi button names defined xi command names defined xi dialog box names defined xi document xi file names defined xi keyboard keys defined xi menu items defined xi menu sequences defined xi Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Index 1 Index system responses defined xi user input defined xi variables xi website addresses xi customization 1 10 application action list 4 23 Association Set Section 4 19 config
17. part number EK G80TR SA The few instance codes that appear in the examples are described in Table D 2 Table D 2 Instance Code Legend Instance Code OEOF8B01 What It Means The copy was terminated due to a write failure on the target unit The write failure was due to the links being down target inaccessible The copy will restart when at least one link is restored The initiator unit is specified by the Initiator WWLID field 0E098901 The remote copy set specified by the Remote Copy Set Name field has gone inoperative due to a disaster tolerance failsafe locked condition 02908901 The host command failed because the remote copy set went failsafe locked prior to command completion The remote copy set is specified by the Remote Copy Name field The Information field of the Device Sense Data contains the block number of the first block error 07050064 Failover Control received a Last Gasp message from the other controller The other controller is expected to restart within a given time period If the other controller does not the other controller will be held reset with the Kill line 43010064 Host Port Protocol component has detected that the other controller has failed and that this controller has taken over the units specified in the extended sense data Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide D 9 Troubleshooting D 10 Data Replication Manager HS
18. 1 6 S SANworks Command Scripter 1 8 to 1 10 1 13 HP OpenVMS installation 2 6 HP Tru64 UNIX installation 2 7 IBM AIX installation 2 7 Microsoft Windows NT 2000 installation 2 8 obtaining 2 5 Sun Solaris installation 2 9 verifying communication with controllers 5 2 scripting procedures 1 6 Extended Planned Site Failover with Full Failback 10 1 Planned Site Role Reversal 13 1 Resumption of Operations After Unplanned Loss of Target Site Failsafe Mode 7 1 Resumption of Operations After Unplanned Loss of Target Site Normal Mode 8 1 Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode 11 1 Short Planned Site Failover with Fast Failback 9 1 Unplanned Site Failover with Failback to New Hardware 12 1 Unplanned Site Failover with Full Failback 6 1 scripts benefits 1 8 description 5 1 error codes D 2 failover and failback 1 10 how they work 1 9 1 10 information flow 1 11 kit files A 1 process flow 1 12 1 13 scenarios 1 6 termination 5 5 SCSI 2 mode HP Tru64 UNIX 1 15 3 2 4 6 IBM AIX 1 15 3 4 4 7 Index 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Index Microsoft Windows NT 2000 1 15 4 9 Sun Solaris 1 15 3 5 4 13 SCSI 3 mode HP OpenVMS 1 15 3 1 4 3 HP Tru64 UNIX 1 15 3 2 4 6 IBM AIX 1 15 3 4 4 7 Microsoft Windows NT 2000 1 15 3 5 4 11 Sun Solaris 1 15 3 5 4 13 sd conf 3 6 software required 1 16 status display condensed 5 3
19. 5 1 RC File Location Script program file RC file created hsg_fb1 xxx app fb1 rc hsg fb2 xxx app fb2 rc 5 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Scripting File Descriptions and Behaviors Terminating a Script IMPORTANT HP recommends that you not terminate a script unless absolutely necessary Terminating a script may leave the system in an unknown state If this occurs you are responsible for putting the system in a known state The way scripts are terminated after being invoked depends on the operating system on which they are running Procedures to terminate running scripts are discussed below for each operating system HP OpenVMS To terminate scripts while processing press Ctrl C A message appears that asks whether you want to terminate the batch job Pressing y yes ends a script while pressing n no cancels the request and display the command prompt Since some scripts are called in the background there may be other Perl scripts running To end these scripts Enter the command SHOW QUEUE Observe any background scripts running A resulting example display for a running script will look like the following Batch queue DRMSCRIPT available on VMS019 Entry Jobname Username Status 25 ROLE_REVERSE_FAILOVER_STEP1_FARGO DOE Executing 2 Delete any background jobs by deleting the specific entry For example enter the command delete entry 25 This deletes
20. 7 Initiator Site Cleanup Procedure eee eee 10 9 Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode Running the Resumption of Replication Program File Procedure 11 1 Continuing the Resumption of Replication Program File Procedure 11 4 vi Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Contents 12 Unplanned Site Failover with Failback to New Hardware Procedure Running the Unplanned Failover Program File Procedure 12 1 Target Host Setup Procedure 2 2 eee eee eens 12 4 Initiator Site Preparation Procedure 1 0 0 eee eee eee eee 12 6 Running the New Hardware Failback Program Files Procedure 12 14 Initiator Site Cleanup Procedure 0 0 cece cece 12 15 13 Planned Site Role Reversal Procedure Initiator Site Preparation Procedure 1 0 0 eee eee eee 13 1 Running the Role Reversal Failover Program File Procedure 13 3 Target Host Setup Procedure 2 0 eee cee eee eens 13 4 Running the Role Reversal Failback Program File Procedure 13 7 Initiator Site Cleanup Procedure 20 ee eects 13 8 A DRM Scripting Kit Files Sample Controller Configuration File C Structure of the Application Action List Default Application Action List 2 0 0 eee I C 1 Action Commands x cs ints bes ux n a e ve ec ad er Wk evt C 5 How the
21. ALLOCATION CLASS 0 SCSI VERSION SCSI 3 Configured for MULTIBUS FAILOVER with 2694115654 In dual redundant configuration Device Port SCSI address 6 Time 10 MAY 2001 16 43 12 Command Console LUN is lun 0 NOIDENTIFIER Host PORT 1 Reported PORT ID 5000 1FE1 0000 4251 PORT 1 TOPOLOGY FABRIC fabric up Address 200213 Host PORT 2 Reported PORT ID 5000 1FE1 0000 4252 PORT 2 TOPOLOGY FABRIC offline REMOTE COPY BUILDNGB Cache 256 megabyte write cache version 0012 Cache is GOOD No unflushed data in cache CACHE FLUSH TIMER DEFAULT 10 seconds 12 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure Mirrored Cache 256 megabyte write cache version 0012 Cache is GOOD No unflushed data in cache Battery NOUPS FULLY CHARGED Expires 01 MAY 2003 2 Open a command prompt window on the target host 3 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 4 Change to the bat sh or vms subdirectory in the CLONE_HOME directory 5 Runthe hsg_fo xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system 6 Youwillseea message asking what type of failover to run
22. EIA 423 maintenance port on the front bezel of the HS series array controllers Also called a maintenance terminal Logical Unit Number LUN A value that identifies a specific logical unit belonging to a SCSI target ID number a number associated with a physical device during a task s I O operations Each task in the system must establish its own correspondence between logical unit numbers and physical devices Glossary 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary LOG_UNIT A CLI command switch that when enabled assigns a single dedicated log unit for a particular association set The association set members must all be in the normal error mode not failsafe See also write history logging LUN See Logical Unit Number maintenance terminal See local terminal mini merge As applied to the Data Replication Manager the data transfers to be made whenever a target becomes inaccessible This happens when both links or both target controllers have gone down The transfers that would have been made are instead logged into the association set s assigned log unit to wait until the remote copy set subsystem comes back online See also fast failback write history logging mirroring Duplicating data onto another computer at another location Mirroring is performed for backup purposes or to be in closer proximity to the user mirrorset 1 A group of storage devices organized as duplicate copie
23. Enter u for an unplanned failover 7 You will see a confirmation message that asks you to confirm that an unplanned failover is desired Enter y for yes 8 If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app fo rc file from the config subdirectory of the CLONE HOME directory and run the program file IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide Q 29 whenan operation completion status result is displayed similar to Figure 6 1 continue the failover procedure at the target site with the Target Host Setup Procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 3 Unplanned Site Failover with Failback to New Hardware Procedure Target Host Setup Procedure 2 1 To verify that failover completed successfully issue this CLI command SHOW REMOTE_COPY_SETS FULL The output shows the status of remote copy sets NOTE Be sure that the units you see listed under Initiator State are at the target site 3 To verify that the targ
24. Example Display 2 Example Display 2 BuildngATop show remote copy sets full Name Uses Used by RCS1 remote copy D1 Reported LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0134 Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE NORMAL FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to the other controller No targets gt 12 The target site is now available for desired actions Continue with the Continuing the Resumption of Replication Program File Procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 11 3 Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode Continuing the Resumption of Replication Program File Procedure When the target site is operational and back on line continue resumption of operations with the following procedure for resuming remote copy set operation If you are connected to the initiator site controllers when the target site comes online you will see the following confirmation messages on your terminal as shown in Example Display 3 Refer to Appendix D for the meaning of the instance 07 JUN 2001 14 16 15 Instance Code 07050064 Years 73 Days 6 Hours 50 Minutes 57 Seconds Peer to Peer Remote Copy target controller HSG80 Serial Number 2694115654 Hardware Version E10 28 V86 1P 08090010 No Last Failure Parameters 07 JUN 2001 14 16 15 Instance Code 43010064 Years 73 Day
25. Failsafe Chapter 11 target site Resumption of Replication After Extended Planned Loss of Target Failsafe Mode Unplanned loss of Unknown Normal or Chapter 12 initiator site function failsafe Unplanned Site Failover Initiator site hardware with Failback to New will be new Hardware Planned change of Unknown Normal or Chapter 13 operations from initiator failsafe Planned Site Role site to alternate site Reversal Initiator site remains operational Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 7 DRM Scripting Overview Benefits of Scripts The use of scripts in a DRM environment simplifies procedures from the operator s perspective when performing failover failback and resumption of operation changes A program file consisting of a batch shell or command file can start an entire failover sequence Down time is shortened by eliminating the delay between command entries The use of scripts also ensures that the sequence of commands has been predetermined in a calm environment rather than during a crisis when mistakes are more common The result is a failover and failback process that is timely consistent and efficient Components for Scripting Scripting requires the following components e The HP DRM Scripting Kit Version 3 0 for your operating system e A Perl interpreter part of the AIX Tru64 and Solaris 8 operating systems but must be obtained separately for OpenVMS So
26. LUN is used then this chapter can be skipped The scripts reside on hosts in a zone having access to controllers at the initiator and target sites The following procedure shows you how to set up one CCL for each of the platforms that support scripting Set up another CCL for the other site controller initiator or target to allow communication from either site s host HP OpenVMS CCL and Job Queue Setup OpenVMS is only supported in SCSI 3 mode The following procedures describe OpenVMS CCL setup and job queue setup HP OpenVMS CCL Setup 1 From the initiator controller use the following command SET THIS_CONTROLLER IDENTIFIER Value The IDENTIFIER switch creates a CCL identifier that makes the controller and the CCL visible to the host Value is a number between 0 32766 2 From the target controller use the following command SET THIS_CONTROLLER IDENTIFIER Value Where the value can be a number between 0 32766 unique for each CCL Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 3 1 CCL Setup From the initiator and target site hosts enter the following command MCR SYSMAN IO AUTOCONFIGURE LOG From the host enter the following command SHOW DEVICE GG A listing of initiator and target CCLs appears in the format 1 GGAValue where Value is a number between 0 32766 Verify that Command Scripter is working by entering the following
27. Manager HSG80 ACS Version 8 7P Scripting User Guide Extended Planned Site Failover with Full Failback Procedure Target Host Setup Procedure 1 2 To verify that the target site host can connect to the LUNs use this command SHOW UNITS FULL In the Access field of the display all units that are used by remote copy sets should show that both the target host and the initiator controller connections are enabled The following step requires actions relative to each operating system in your configuration HP OpenVMS Allow the hosts to recognize new units 1 If you have shut down the host boot it now Booting the host enables OpenVMS to recognize the drive 2 If you did not shut down the host use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG HP Tru64 UNIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 If the target hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number IBM AIX Allow the hosts to recognize new units 1 If the target site hosts are shut d
28. Perl Scripts Use the Application Action List 0005 C 6 hsecontrol Pl arsos nrnna en kd ness eese neve WS aeta C 6 drmdispatch pl 22 22 he ee 2202 er aa le ert en C 7 D Troubleshooting Troubleshooting Recommendations 0 0 e eects D 1 Scripting Error Codes 22 2222 cece ence eens D 2 Confirmation Message Instance Codes 00 c cece eee eens D 9 E DRM Power Up and Power Down Power Up Data Replication Manager Systems 00 00 e eee eee eee E 1 Target Site Power Up Procedure 1 0 0 2 cee eee ee eee E 1 Initiator Site Power Up Procedure 1 0 2 0 00 eee E 2 Power Down Data Replication Manager Systems 2 2202 0 0 0 eese E 2 Initiator Site Power Down Procedure 00 eee eee ern E 2 Target Site Power Down Procedure 00 cece ee eee eee nee E 3 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide vii Contents Glossary Index Figures 1 1 Scripting information flow llsseeeeeee eee 1 11 1 2 Script PF Cessin g 4 s lur ats eae et RS Be ar 1 12 1 3 Data Replication Manager basic configuration 1 13 4 1 Generation file setup using a CCL eee eee 4 5 4 2 Generation file setup with anon RCSLUN 2 2222 ceenenen 4 10 4 3 Copying association set information 0 0 cee eee eee 4 20 44 Copying remote copy set information 0 0 0 ee 4 21 5 1 Verbose status di
29. READAHEAD_CACHE WRITEBACK_CACHE MAX_READ_CACHED_TRANSFER_SIZE 32 MAX_WRITE_CACHED_TRANSFER_SIZE 32 Access None State ONLINE to this controller Not reserved NOPREFERRED_PATH Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 11 Unplanned Site Failover with Failback to New Hardware Procedure D2 DISK20000 LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003F NOIDENTIFIER Switches RUN NOWRITE_PROTECT READ_CACHE READAHEAD_CACHE WRITEBACK_CACHE MAX_READ_CACHED_TRANSFER_SIZE 32 MAX_WRITE_CACHED_TRANSFER_SIZE 32 Access None State ONLINE to this controller Not reserved NO PREFERRED_PATH Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 gt 5 Set up new units for any additional remote copy sets that were added at the target site while failed over by using the following CLI command ADD UNIT UnitName ContainerName DISABLE_ACCESS_PATH ALL Example add unit dl disk10000 disable_access_path all gt 6 Atthe initiator site the units must be preferred to one controller or the other Check for preference with the following CLI command SHOW UNITS FULL You will see a display similar to that in Example Display 6 Example Display 6 D1 BuildngATop gt show units full DISK10000 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 013
30. Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Run the hsg_fo xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failover to run Enter p for a planned failover You will see a confirmation message that asks you to confirm that a planned failover is desired Enter y for yes You will see a message asking what type of planned failover to run Enter e for an extended planned failover If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Press v for verbose or c for condensed To change the type of display you must delete the app fo rc file from the config subdirectory of the CLONE HOME directory and run the program file IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide listed in Table 1 9 Continue with the planned failover at the target site with Target Host Setup Procedure 10 4 Data Replication
31. Scripting Kit A self extracting kit that contains batch files Perl scripts Perl modules control tables and example files used to configure a DRM environment and perform failover and failback DT See disaster tolerance Glossary 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary dual redundant configuration A storage subsystem configuration consisting of two active controllers operating as a single controller If one controller fails the other controller assumes control of the failing controller s devices See also failover failback environmental variable Environmental information such as drive path or filename associated with a symbolic name With the DRM scripts the environmental variable CLONE_HOME defines the path to the default directory where the scripts reside fabric A network of Fibre Channel switches or hubs and other devices See also switch fabric failback In a DRM environment after failover occurs failback moves data operations back to the initiator after the initiator site has been brought back online See also failover failover The process that takes place when one controller in a dual redundant configuration assumes the workload of a failed companion controller Failover continues until the failed controller is repaired or replaced See also failback dual redundant configuration planned failover failsafe locked A condition a remote copy set assumes under cer
32. Table 4 1 lists the files created or customized on the initiator and target hosts while performing the procedures in this chapter A configuration generation file for each controller subsystem is created on both hosts and saved in the BAT SH or VMS subdirectory of CLONE_HOME the name of the directory where the scripts reside depending on your operating system The xxx file extension denotes a filename ending with com OpenVMS sh Tru64 UNIX AIX and Solaris or bat Windows Table 4 1 Created or Customized Host Files Initiator Host Target Host Subdirectory File Subdirectory File bat sh or vms bat sh or vms InitiatorName_gen xxx Example tulsa_gen bat TargetName_gen xxx Example fargo_gen bat bat sh or vms bat sh or vms InitiatorName_gen xxx Example tulsa_gen bat TargetName_gen xxx Example fargo_gen bat config InitiatorName cfg Example tulsa cfg config InitiatorName cfg Example tulsa cfg config TargetName cfg Example fargo cfg config TargetName cfg Example fargo cfg config app act config app act The generation files you create are run on each host to create initiator and target controller configuration files These will have a cfg extension and will be saved in the config subdirectory On each host customization occurs on two types of files the target controller configuration files TargetName cfg on the ini
33. Use a text editor and make the necessary modifications using the following syntax Perl I CLONE_HOME CLONE_HOME bin generate_cfg pl com cs RemoteCopyName Device where RemoteCopyName identifies the DRM controller remote copy name of the initiator subsystem for example tulsa The remote copy name can be obtained by running a SHOW THIS command from the controller Device specifies the non RCS LUN that Command Scripter uses to communicate with the initiator controller from the initiator host for example Q in Figure 4 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 9 File Customization Setup for One Controller Subsystem Initiator Site Target Site Controller A Setup for Maximum Controllers Per Host Initiator Site Target Site Host B Controller F CXO7732A Figure 4 2 Generation file setup with a non RCS LUN 4 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization 3 Save the edited initiator configuration generation batch file into the CLONE_HOME bat subdirectory with a meaningful name like tulsa_gen bat Modify the gen_ex bat file again to create a target configuration generation file for the initiator site The syntax is Perl I CLONE_HOME CLONE_HOME bin generate_cfg pl com cs RemoteCopyName Device where RemoteCopyName identifies the DRM controller remote copy
34. and restart of processing at both sites at both sites Loss of single disk in Failover not necessary Failover not necessary redundant storage Loss of single storageset Failover not necessary Failover not necessary Loss of single host of Failover not necessary Failover not necessary cluster If one host in a multi host environment fails you must decide whether or not a failover is the best course of action When you determine that a site failover is necessary identify which scenario best describes your situation planned unplanned or role reversal failover Use the planned failover procedure when failover is a scheduled event These are situations such as anticipated power disruption scheduled equipment maintenance at the local site or the need to transfer operations to another site Planned failovers are further characterized as short or extended A short planned failover also referred to as prefast in the scripts assumes the write history log will be able to accommodate the accumulated writes for the duration of the failover An extended planned failover also referred to as prefull in the scripts assumes the write history log will not accommodate the accumulated writes An unplanned failover involves situations such as multiple controller failures multiple host failures or an unplanned power outage at the local site A site role reversal failover transfers the initiator role to another site The original initiator site then
35. association sets remote copy sets units and connections Refer to the HP StorageWorks Data Replication Manager HSG80 ACS 8 7P Failover Failback Procedures Guide for the full status comparison procedure Open a command prompt window on the target host Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Run the hsg_fb xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter s for a fast failback You will see a confirmation message that asks you to confirm that a fast failback is desired Enter y for yes 9 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Short Planned Site Failover with Fast Failback Procedure 8 If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app_fb1 rc file from the config subdirectory of the CLONE HOME directory and run the program file 9 The failback script performs a mini merge to copy data from the wr
36. assumes the role of a target site 1 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Failback Procedure Choices During failover the remote copy sets at the target site are in a copy ready state waiting for the initiator site to become available When a new initiator site has been established or the original one has been restored site operation can resume after a failback procedure has been performed This involves synchronizing data on both the initiator and target subsystems so that operation can be returned to the initiator with minimal downtime IMPORTANT Verify that all components at both sites are operational before performing a failback The failback sequence is a scheduled event The HSG80 Array Controller requires that a viable dual redundant subsystem be available before a failback can take place IMPORTANT Failback to a single controller configuration is not supported Table 1 2 can help you determine which failback procedure to use in different circumstances Table 1 2 Types of Failover and Failback State ofthe initiator Failover Failback controller pair type used type used Example Initiator site Short Maintenance needs to be intact Planned performed at the initiator site The site is brought back up when maintenance is complete with a mini merge from the write history log Initiator site Extended Full Maintenance is performed at the intact Planned
37. controller Save the edited generation file into the CLONE_HOME sh subdirectory with a meaningful name like tulsa_gen sh Modify the gen ex sh file again to create a target configuration generation file for the initiator site The device identification number used in this file specifies the device identification number for the CCL of the controller at the target site Controller B communicating with the initiator host Host A Use the syntax perl I CLONE HOME S CLONE HOME bin generate cfg pl com cs RemoteCopyName idnum where RemoteCopyName identifies the target DRM controller remote copy name of the target subsystem for example fargo This name is case sensitive in IBM AIX and must match the case of the name used later when configuring the application action list idnum specifies the device number for the operating system that Command Scripter uses to communicate with controller fargo As shown in Figure 4 1 this is the device number of the CCL for the remote controller Controller B communicating with the initiator host Host A 4 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization 4 Save this program file in the sh subdirectory with a meaningful name like fargo_gen sh Repeat the above steps to create initiator and target configuration generation program files for the target site Use the correct device identification numbers of the CCLs communica
38. copied mirrored from the originating initiator site to a remote target site The result is an exact copy of the data remote copy set at the target site Used in disaster tolerance DT applications such as the Data Replication Manager See also disaster tolerance DT SAN See Storage Area Network script A program written in an interpreted programming language that specifies a set of actions to perform a specific task For DRM scripting the hsgcontrol pl script reads actions from the application action list and then calls the drmdispatch pl script The drmdispatch pl script interprets and executes instructions from the failover and failback control tables and initiates CLI commands to accomplish failover and failback See also failover failback scripting language A high level programming or command language that is interpreted translated on the fly rather than compiled ahead of time A scripting or script language may be a general purpose programming language or it may be limited to specific functions to augment the running of an application or system program Spreadsheet macros and communications scripts are examples of limited purpose scripting languages SHOW commands A set of CLI commands that display information about controllers storagesets devices partitions and units Storage Area Network SAN A back end network connecting storage devices via peripheral channels such as SCSI and Fibre Channel There are two way
39. during installation where CLONE_HOME is a variable indicating the default directory where the scripts reside Each batch shell or command file contains explicit Perl commands and parameters that perform specific actions based on customizations made to the controller configuration files and application action list discussed in Chapter 4 You should run the program file for the desired failover failback or resumption of operation scenario from a command prompt window to see the status results when the script finishes The program files listed here with a xxx extension denoting bat sh or com perform the following functions e hsg fo xxx performs a site failover on controllers identified in the application action list You are prompted to run a planned unplanned or role reversal failover e hsg fbl xxx performs a site failback step 1 on controllers identified in the application action list You are prompted to run a fast full role reversal or new hardware failback e hsg fb2 xxx performs a site failback step 2 on controllers identified in the application action list You are prompted to run a fast full role reversal or new hardware failback e hsg op xxx allows a resumption of operations by toggling the error mode of the remote copy sets between failsafe locked and normal mode Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 5 1 Scripting File Descriptions and Behaviors IMPORTANT Do no
40. ee ee 2 1 HP OpenVMS ieee Ha Ga E Re S Ue be 2 2 HP Tru64 UNIX IBM AIX and Sun Solaris 000000 2 2 Microsoft Windows NT 2000 222222020 eee eee eens 2 3 Perl Interpreter 22 8 4224 kenne LE QU dns 2 4 HP OpenVMS uree oe hele 22 2 22 Poe er cama ee en ede 2 4 Obtaining the OpenVMS Perl Interpreter llle esee 2 4 Installing the OpenVMS Perl Interpreter 0 00 00 0000s 2 4 Microsoft Windows NT 2000 ssseseeeeee eee eens 2 4 Obtaining the Windows Perl Interpreter ActivePerl 2 4 Installing Windows ActivePerl 0 00 ccc eee eee 2 4 Sun Solaris Versions 6 and 7 0 0 cee cece eee es 2 5 Obtaining the Sun Solaris Versions 6 and 7 Perl Interpreter 2 5 Installing the Sun Solaris Perl Interpreter eese 2 5 SANworks Command Scripter 22222 eee eee eee eee 2 5 Obtaining SANworks Command Scripter 0 0 c eee eee eee 2 5 Installing SANworks Command Scripter 2 0 0 0 e eee eee eee ee 2 5 HP OpenVMS ise Senye ee pipe ease distin ee tegi uds 2 6 HP Tru64 UNIX occore odour eee Mine ee ET i o NEE a E E hm he 2 7 BMAX LL se te be ete rn Pe Rare ehe el 2 1 Microsoft Windows NT 2000 lsseeeeeee I 2 8 Sun Solaris au aut ah es c ev e s ptt ese ug oet e ede eye e 2 9 3 CCL Setup Introduction uet uie ir sa ae tA e Ra ER ein 3 1 HP OpenVMS CCL and Job Queue Setup 3 1 HP Open VMS CCE Setwper usa ar
41. failover unit SITE_FAILOVER times Controller problem with site command failed failover of remote copy set Identify and correct problem Possible communication error Verify communication with controller and 2010 drm_move_initiator cannot Possible mismatch between read configuration entry for configuration file and controller Verify remote copy set that the configuration file accurately reflects configuration on controller correct problem as needed 2011 drm_move_initiator cannot Possible communications or controller move initiator role to target in problem Identify and correct problem remote copy set 2012 drm_move_ initiator initiator Possible mismatch between role for remote copy set is configuration file and controller Verify already specified to another that the configuration file accurately controller and or unit Mixed up reflects configuration on controller initiator and target 2021 add rcs_to_assocset Mismatch between configuration file configuration file not up to date and controller configuration Update configuration file D 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Troubleshooting add_rcs_to_assocset cannot add association set add_whl_to_assocset cannot add WHL to association set remove_rcs_from_assocset the association set does not exist drm_create_remotecopyset a storageset remote copy set or association set already exists
42. hsgcontrol pl Reads actions from the application action list and calls the drmdispatch pl Perl script to perform these named actions This file should not be modified BIN pchoice pl Handles user input for prompts that require a logical response This file should not be modified CONFIG app_ex act Example file used to create an application action list CONFIG failback_stepla tbl Contains step 1 failback instructions read by the drmdispaitch pl Perl script and performed on the initiator This file should not be modified CONFIG failback_stepib tbl Contains step 1 failback instructions read by the drmdispatch pl Perl script and performed on the target This file should not be modified Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide A 1 DRM Scripting Kit Files Table A 1 Installed DRM Scripting Kit Files Continued Directory Filename Description CONFIG failback_stepic tbl Contains step 1 failback instructions read by the drmdispatch pl Perl script and performed on the target This file should not be modified CONFIG failback_step2 tbl Contains step 2 failback instructions read by the armdispatch pl Perl script This file should not be modified CONFIG failover_step1 tbl Contains step 1 failover instructions read by the armdispatch pl Perl script This file should not be modified CONFIG failover_step2 tbl Contains step 2 failover instructions read by the armdispatch pl Perl script This file sho
43. initiator site for a length of time that would exceed the space on a write history log A failback to the initiator after maintenance requires a full disk copy Initiator site Unplanned Full Power goes off at initiator site intact Failover is performed to the target site Failback to the initiator is performed with a full disk copy once power is restored Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 5 DRM Scripting Overview Table 1 2 Types of Failover and Failback Continued State of the initiator controller pair Failover type used Failback type used Example Initiator site not intact Unplanned New Hardware Lightning strike damages equipment resulting in a disaster failover Once new equipment is installed a failback is performed Initiator site intact Role Reversal Role Reversal Another site is given the initiator role for an unspecified time At a later date the initiator role reverts back to the original site Summary of Failover Failback Procedure Choices Scripts are available for most of the failover failback and resumption of operation scenarios you are likely to encounter To perform these procedures with CLI commands and to find other failover and failback scenarios refer to the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide Table 1 3 lists the scripting sc
44. kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance 3 After the completion of maintenance work at the initiator site or whenever you are ready to failback to the initiator site continue with the Running the Role Reversal Failback Program File Procedure 13 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Planned Site Role Reversal Procedure Running the Role Reversal Failback Program File Procedure 1 Before performing the failback procedure locate your record of SHOW command output that details the initiator configuration Verify that your initiator controller configuration is the same as your target controller configuration Compare the status of the controllers association sets remote copy sets units and connections Refer to the HP StorageWorks Data Replication Manager HSG80 ACS 8 7P Failover Failback Procedures Guide for the full status comparison procedure Open a command prompt window on the target host Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on pa
45. lines is further detailed in Figure 1 2 on page 1 12 to show the interaction of specific failover and failback Perl scripts IMPORTANT The names of remote copy sets stripesets mirrorsets RAlDsets association sets and connections may not contain a hyphen This is a Perl language restriction Underscores _ are allowed 1 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Application Control action list table B a pe 5E ao c 26 ES eto ES tO 28 8 Pon CLI commands an Perl scripts device identifiers and verification Subsystem configuration data Configuration file interpreter SHOW command Server d Command Scripter Figure 1 1 Scripting information flow Fibre Channel SHOW command CLI commands verification Fibre Channel HSG80 controller CX7537B Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Batch shell or command file Perl interpreter app act file hsgcontrol pl Perl script c Control Called processes table Configuration drmdispatch pl armdispatch pl Perl script Perl script Perl modules SHOW command CLI commands SHOW command responses responses CXO7657A Figure 1 2 Script processing The Scripting Process Flow Figure 1 2 shows a high level view of the
46. not installed correctly you will see each unit twice 2 If you have changed the UNIT OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You should be able to see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 5 Unplanned Site Failover with Failback to New Hardware Procedure f Sun Solaris Allow the hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges You should be able to see all of the units by using the format command 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance Q 4 Oncethe problem that disabled the initi
47. number of the CCL for the local controller Controller A communicating with the initiator host Host A In OpenVMS the device identification number can be obtained by entering the command SHOW DEVICE GG A listing of CCLs appears in the format 1 GGAValue where Value is a number between 1 9999 An example of a valid device identification number would be 1 GGA3001 NOTE The disk identifier or Value can be any decimal number from 0 to 32766 except Mass Storage Control Protocol MSCP served devices If the Fibre Channel disk device is MSCP served the device identifier is limited to 9999 2 Save the edited generation file into the CLONE_HOME vns subdirectory with a meaningful name like tulsa_gen com 3 Modify the gen_ex com file again to create a target configuration generation file for the initiator site The device identification number used in this file specifies the device identification number for the CCL of the controller at the target site Controller B communicating with the initiator host Host A Use the syntax perl I CLONE HOME BlN generate cfg pl com cs RemoteCopyName idnum RemoteCopyName identifies the target DRM controller remote copy name of the subsystem for example fargo idnum specifies the device identification number for the operating system that Command Scripter uses to communicate with target controller fargo As shown in Figure 4 1 this is the identification number of the CCL for the remo
48. operating system in your configuration a HP OpenVMS Allow hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives 2 If the target site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 If the target hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters is located with the following command hwmgr scan scsi bus x where x is the SCSI bus number c IBM AIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives 2 If the target hosts are not shut down use the following commands to recognize the drives and verify that they are present cfgmgr v lsdev Cc disk Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Short Planned Site Failover with Fast Failback Procedure Use the following commands to access file systems on the failed over LUNs importvg y
49. reboot the target site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice 6 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Full Failback Procedure 2 If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You should be able to see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice f Sun Solaris Allow the hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges You should be able to see all of the units by using the format command 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After edi
50. shut down boot them now Booting the hosts enables AIX to recognize the drives If the initiator site hosts are not shut down use the following commands to recognize the drives and mount the file systems cfgmgr v mount all 9 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Short Planned Site Failover with Fast Failback Procedure gt 4 Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the initiator site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice gt 5 Microsoft Windows 2000 Allow hosts to recognize new units a If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the initiator site hosts 1 On each host log in using an account that has administrative privileges 2 Open Computer Management and click Disk Management 3 After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice b If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You see all of the
51. site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives b If the initiator hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number gt 3 IBM AIX Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 6 7 Unplanned Site Failover with Full Failback Procedure b Ifthe initiator site hosts are not shut down use the following commands to recognize the drives and mount the file systems cfgmgr v mount all gt 4 Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the initiator site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice gt 5 Microsoft Windows 2000 Allow hosts to recognize new units a If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the initiator site hosts 1 On each host log in u
52. the script running in the background 3 Confirm that all background jobs are deleted by entering the SHOW QUEUE command again 4 The resulting display will look like the following Batch queue DRMSCRIPT idle on VMS019 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 5 5 Scripting File Descriptions and Behaviors HP Tru64 UNIX IBM AIX and Sun Solaris Use the following procedure to terminate a script and background processes in Tru64 UNIX AIX or Solaris 1 2 3 Press the Ctrl C keys to terminate the script Type the following commands to find scripting processes still running ps ef grep cmdscript ps ef grep perl Terminate these processes using the kill command Microsoft Windows NT 2000 To terminate scripts while processing press the Ctrl C keys A message appears that asks whether you want to terminate the batch job Pressing y yes ends a script while pressing n no cancels the request and displays the command prompt Since some scripts are called in the background there may be other Perl scripts running To end these scripts 1 Open the Windows Task Manager by pressing Ctrl Alt Delete and clicking Task Manager Click Processes tab Search for all appearances of Perl exe in the list Highlight each occurrence and click End Process Close the Windows Task Manager 5 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 6 Unplanned
53. verbose 5 3 StorageWorks Command Console 1 16 Sun Solaris CCL setup 3 5 Command Scripter installation 2 9 creating configuration generation files 4 13 device number 4 13 4 14 DRM Scripting Kit description 1 8 installation 2 2 Perl interpreter 1 16 platform requirements 1 14 program file 1 1 running configuration generation files 4 18 script termination 5 6 SCSI 2 mode 1 15 3 5 SCSI 3 mode 1 15 3 5 using the CCL 4 13 supported platforms 1 14 switch zoning 1 16 4 13 4 13 symbols in text xi system responses convention defined xi T target controller configuration file 4 19 target site 1 1 target site symbol and definition xi terminating a script HP OpenVMS 5 5 HP Tru64 UNIX 5 6 IBM AIX 5 6 Microsoft Windows NT 2000 5 6 Sun Solaris 5 6 text symbols xi U user input convention defined xi V variables convention defined xi verbose status display 5 3 W warning symbol and definition xi Website addresses convention defined xi websites ActivePerl interpreter 2 4 Command Scripter updates 2 5 DRM updates 2 1 Perl interpreter for OpenVMS 2 4 Perl interpreter for Sun Solaris 2 5 Z zoning 1 16 4 9 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Index 5 Index Index 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide
54. which site should continue processing continue at inititator site or failover to target site Manually continue processing at initiator site Loss of target fabric Failover not necessary Manually continue processing at initiator site Loss of target controller pair Loss of single initiator controller Failover not necessary Failover not necessary Manually continue processing at initiator and target sites Failover not necessary Loss of both initiator switches Manual intervention to fail over data to target site and restart of processing at both sites Manual intervention to fail over data to target site and restart of processing at both sites Loss of single initiator switch Failover not necessary Failover not necessary Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Table 1 1 Possible Failover Situations Continued Recommended Action Remote Copy Set Remote Copy Set Type of Failure Error_Mode Normal Error_Mode Failsafe Extended power outage Manual intervention to fail Manual intervention to fail at initiator site over data and processing over data and processing to target site to target site Loss of both host bus Manual intervention to fail Manual intervention to fail adapters non clustered over data to target site over data to target site hosts and restart of processing
55. xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failover to run Enter u for an unplanned failover You will see a confirmation message that asks you to confirm that an unplanned failover is desired Enter y for yes If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app_fo rc file from the config subdirectory of the CLONE_HOME directory and run the program file When an operation completion status result is displayed similar to Figure 6 1 continue the failover procedure at the target site with the Target Host Setup Procedure IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide Target Host Setup Procedure 1 2 To verify that failover completed successfully issue this CLI command SHOW REMOTE_COPY_SETS FULL The output shows the status of remote copy sets Be sure the units listed under Initiator State are at the target site
56. 0000 SK50300 SK60000 SK60300 E AO LA AE D EAE ee YS OG Fe YS AS UN A POA YAE YA OA A A A A DNDUOPBWWWWNNNHNEFERE PR WOWO COO CO PO I2 CO CO PO F2 CO CO P2 ES CO oooooooooooo0000000 Normal Termination Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 9 Unplanned Site Failover with Failback to New Hardware Procedure gt i Create and initialize all storagesets and units This includes all that had existed at the initiator site as well as those that were created at the target site since failover The units that will be part of remote copy sets must be identical to the corresponding units at the target site Wait to create any units that will be used for log disks until later in this procedure See the HP StorageWorks HSG80 ACS Version 8 7P Configuration Guide for information on creating storagesets and units gt j Verify the creation of the storagesets and units with the following CLI command SHOW UNITS FULL You will see a display similar to that in Example Display 4 Example Display 4 BuildngATop gt show units full LUN Uses Used by D1 DISK10000 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0134 NOIDENTIFIER Switches RUN NOWRITE PROTECT READ CACHE READAHEAD CACHE WRITEBACK CACHE MAX READ CACHED TRANSFER SIZE 32 MAX WRITE CACHED TRANSFER SIZE 32 Access All State ONLINE to this controller Not reserved NOPREFERRED_PATH Host Based Logging NOT Specified Size 17
57. 022 Cache is GOOD No unflushed data in cache Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 3 3 CCL Setup Battery NOUPS FULLY CHARGED Expires 29 AUG 2003 IBM AIX CCL Setup AIX is supported in SCSI 2 and SCSI 3 mode The following steps enable the CCL and configure it on the IBM AIX platform 1 To enable the controller CCL in SCSI 2 mode enter the following command from each controller subsystem SE HIS_CONTROLLER SCSI_VERSION SCSI 2 SE HIS_CONTROLLER COMMAND_CONSOLE_LUN 2 To enable the CCL in SCSI 3 mode enter the following command from each controller subsystem SET THIS_CONTROLLER SCSI_VERSION SCSI 3 3 Verify that the CCL is enabled by entering the following command from each controller subsystem SHOW THIS_CONTROLLER The following is a sample output showing that the CCL is enabled Controller HSG80 2605103470 Software V87P Hardware E12 NODE_ID 5000 1FE1 0009 0A40 ALLOCATION CLASS 0 SCSI VERSION SCSI 3 Configured for MULTIBUS FAILOVER with ZG05103641 In dual redundant configuration Device Port SCSI address 6 Time 13 DEC 2001 08 57 54 Command Console LUN is lun 0 IDENTIFIER 10 4 Verify that CCLs can be seen by the host by entering the command from the host lsdev Cc disk The following sample output verifies the host recognizes the CCL hdisk0 Available 10 60 00 6 0 16 Bit LVD SCSI Disk Drive hdisk1 A
58. 1 Verify that the intersite connections are severed with the following CLI command SHOW THIS_CONTROL ER SHOW OTHER_CONTROLLE R You will see a display similar to that in Example Display 1 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 1 Unplanned Site Failover with Failback to New Hardware Procedure Example Display 1 BuildngBTop gt show this_controller Controller HSG80 2G94115654 Software V87P Hardware E10 NODE_ID 5000 1FE1 0000 4250 ALLOCATION_CLASS 0 SCSI_VERSION SCSI 3 Configured for MULTIBUS FAILOVER with 2694319198 In dual redundant configuration Device Port SCSI address 7 Time 10 MAY 2001 16 41 11 Command Console LUN is lun 0 NOIDENTIFIER Host Connection Table is NOT locked Smart Error Eject Disabled Host PORT_1 Reported PORT_ID 5000 1FE1 0000 4253 PORT_1_TOPOLOGY FABRIC fabric up Address 260213 Host PORT_2 Reported PORT_ID PORT_2_TOPOLOGY FABRIC offline Address 260413 REMOTE_COPY BUILDNGB 5000 1FE1 0000 4254 Cache 256 megabyte write cache version 0012 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER DEFAULT 10 seconds Mirrored Cache 256 megabyte write cache version 0012 Cache is GOOD No unflushed data in cache Battery NOUPS FULLY CHARGED Expires 01 MAY 2003 BuildngBTop show other controller Controller HSG80 2G94319198 Software V86 1P Hardware E10 NODE ID 5000 1FE1 0000 4250
59. 2000 If the operating system is running shut it down and power off the hosts Sun Solaris If the operating system is running shut it down and power off the hosts Continue the planned failover process with the Running the Short Planned Failover Program File Procedure Running the Short Planned Failover Program File Procedure 1 2 Open a command prompt window on the target host Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 9 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Short Planned Site Failover with Fast Failback Procedure o 0 Q9 Change to the bat sh or vms subdirectory in the CLONE HOME directory Run the hsg_fo xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failover to run Enter p for a planned failover You will see a confirmation message that asks you to confirm that a planned failover is desired Enter y for yes You will see a message asking what type of planned failover to run Enter s for a short planned failover If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed re
60. 4 IDENTIFIER 1 Switches RUN NOWRITE_PROTECT READ_CACHE READAHEAD_CACHE WRITEBACK_CACHE MAX_READ_CACHED_TRANSFER_SIZE 32 MAX_WRITE_CACHED_TRANSFER_SIZE 32 Access None State ONLINE to this controller Not reserved PREFERRED_PATH THIS_CONTROLLER Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 12 12 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure 7 D2 DISK20000 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0135 NOIDENTIFIER Switches RUN NOWRITE_PROTECT READ_CACHE READAHEAD_CACHE WRITEBACK_CACHE MAX_READ_CACHED_TRANSFER_SIZE 32 MAX_WRITE_CACHED_TRANSFER_SIZE 32 Access None State ONLINE to this controller Not reserved PREFERRED_PATH OTHER_CONTROLLER Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 If the units need to be preferred use the following command SET UnitName PREFERRED_PATH THIS_CONTROLLER Example set dl preferred_path this_controller or SET UnitName PREFERRED_PATH OTHER_CONTROLLER Example set d2 preferred_path other_controller Set the maximum cached transfer size to 128 with the following CLI command SET UnitName MAXIMUM_CACHED_TRANSFER_SIZE 128 Example set dl maximum_cached_transfer_size 128 NOTE
61. 769177 blocks Geometry C H S 5258 20 169 D2 DISK20000 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0135 NOIDENTIFIER Switches RUN NOWRITE PROTECT READ CACHE READAHEAD CACHE WRITEBACK CACHE MAX READ CACHED TRANSFER SIZE 32 MAX WRITE CACHED TRANSFER SIZE 32 Access All State ONLINE to this controller Not reserved NOPREFERRED_PATH Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 12 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure gt k HP Tru64 UNIX IBM AIX Microsoft Windows and Sun Solaris Go to Step 3 HP OpenVMS only Use the following command to set each unit s device identifier to the value it was prior to hardware replacement SET UnitName IDENTIFIER Value Example set dl identifier 1 This becomes the VMS device identifier for DGx1 gt 3 Disable all access to the units with the following CLI command SET UnitName DISABLE ALL Example set dl disable all Repeat this step for each unit gt a Verify the disabled access with the following CLI command SHOW UNITS FULL You will see a display similar to that in Example Display 5 Example Display 5 BuildngATop gt show units full LUN Uses Used by D1 DISK10000 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0134 IDENTIFIER 1 Switches RUN NOWRITE_PROTECT READ_CACHE
62. 80 ACS Version 8 7P Scripting User Guide Glossary This glossary defines terms associated with the use of scripts to perform failover failback and resumption of operations in a Data Replication Manager DRM environment It is not a comprehensive glossary of computer terms ACS See array controller software ActivePerl The Perl interpreter used by the DRM Perl scripts in the Windows platform application action list A file that controls multiple instances of an action across DRM pairs initiator target pairs An application action list is specific for one failover failback application and specifies the actions that are to be performed on an initiator target pair array controller See controller array controller software ACS Software that is contained on a removable PCMCIA program card that provides the operating environment for the array controller Also known by the acronym ACS association set A group of remote copy sets that share common attributes Members of an association set can be configured to transition to the same state at the same time An association set e Shares the same log unit e Has its host access removed from all members when one member fails e Keeps I O order across all members Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary 1 Glossary asynchronous mode A mode of operation of the remote copy set whereby the write operation provides command completion to the ho
63. CLONE HOME directory and run the program file Continue with the new hardware failback procedure at the initiator site with Initiator Site Cleanup Procedure Initiator Site Cleanup Procedure gt 1 After you have completed this action verify the connections with the following CLI command SHOW CONNECTIONS You will see a display similar to that in Example Display 7 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 15 Unplanned Site Failover with Failback to New Hardware Procedure Unit Example Display 7 BuildngATop gt show connections Connection Name Operating system Controller Port Address Status Offset BUILDNGBA PPRC_TARGET THIS 2 260413 OL this HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4254 BUILDNGBB PPRC_TARGET OTHER 2 200413 OL other HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4252 BUILDNGBC PPRC_INITIATOR THIS 2 offline HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4254 BUILDNGBD PPRC_INITIATOR OTHER 2 offline HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4252 HOSTA1 WINNT THIS 1 260013 OL this HOST ID 1000 0000 C920 A7B9 ADAPTER ID 1000 0000 C920 A7B9 HOSTA2 WINNT OTHER l 200013 OL other HOST ID 1000 0000 C921 3F4E ADAPTER ID 1000 0000 C921 3F4E HOSTB1 WINNT THIS I 220013 OL this HOST ID 1000 0000 C921 3E98 ADAPTER ID 1000 0000 C921 3E98 HOSTB2 WINNT OTHER l 250013 OL other HOST ID 1000 0000 C921 3E
64. CLONE_HOME bin generate_cfg pl com cs RemoteCopyName idnum where RemoteCopyName identifies the DRM controller remote copy name of the initiator subsystem for example tulsa This name is case sensitive in Tru64 UNIX and must match the case of the name used later when configuring the application action list idnum specifies the device identification number for the operating system that Command Scripter uses to communicate with controller tulsa As shown in Figure 4 1 this is the identification number of the CCL for the local controller Controller A communicating with the initiator host Host A In Tru64 UNIX the device identification number can be obtained by entering the command hwmgr view device The resulting output shows the CCLs see HP Tru64 UNIX CCL Setup on page 3 2 The device identification number or idnum in that example is scp1 To find the controller remote copy name for a particular CCL enter the following command from the bin subdirectory of CLONE HOME cmdscript f DeviceName show this controller Save the edited shell file into the SCLONE HOME sh subdirectory with a meaningful name like tulsa_gen sh 4 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization 3 Modify the gen_ex sh file again to create a target configuration generation file for the initiator site The device identification number used in this file specifies the de
65. CONTROLLER PROMPT InitiatorControllerNameBottom gt Example set other_controller prompt buildngA Bottom gt 12 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure gt f Set SET THIS_CONTROLLER MIRRORED_CACHE mirrored cache using the following CLI command NOTE This CLI command may fail because internal cache diagnostics are running These diagnostics can take up to 5 minutes to complete so you may need to retry this command gt g Enter the following command SET THIS_CONTROLLER REMOTE_COPY InitiatorRemoteCopyName Example set this_controller remote_copy buildngA gt h Run the Configuration utility to assign a disk name to physical disks with the following CLI command RUN CONFIG You will see a display similar to that in Example Display 3 Example Display 3 BuildngATop gt run config Config Config on the add dis add dis add dis dis dis dis dis dis dis dis dis dis dis dis dis dis dis dis w Q aq aq w Q aq VMMA aaaaqaqa ouo aa AQ DO QOO Os OQ OG OO oo aa Q Config Local Program Invoked is building its tables and determining what devices exist subsystem Please be patient SK10000 SK10100 SK10200 SK10300 SK20000 SK20100 SK20200 SK20300 SK30000 SK30100 SK30200 SK30300 SK40000 SK40300 SK5
66. CS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state INOPERATIVE Unit failsafe locked Target state BUILDNGB D1 is COPYING 0 complete a If all remote copy sets are in failsafe error mode and 1 Every initiator state is failsafe locked proceed to step 5 2 Every initiator state is not failsafe locked proceed to step 5 3 Initiator states are a mixture of failsafe locked or not being in failsafe locked proceed to step 4 b Ifall remote copy sets are in normal mode proceed to step 5 c If the remote copy sets are mixed between failsafe and normal error modes and 1 Every remote copy set in failsafe error mode is failsafe locked and every remote copy set in normal error mode has no targets proceed to step 5 2 Every remote copy set in failsafe error mode is not in a failsafe locked state proceed to step 5 3 Any remote copy sets in failsafe error mode are failsafe locked and remote copy sets in normal error mode have targets proceed to step 4 gt 4 Forremote copy sets in failsafe error mode with mixed initiator states or a mixture of error modes with failsafe error modes that are failsafe locked and normal error mode with targets change all the remote copy sets in failsafe mode to normal mode Use the CLI command SET RemoteCopySetName ERROR_M
67. D_FAILOVER_STEP2 Background fargo drmdispatch PLANNED prefast_failover_step2 ALL NOTFORCED Background denver drmdispatch PLANNED prefast_failover_step2 ALL NOTFORCED END_SHORT_PLANNED_FAILOVER_STEP2 SHORT_PLANNED_FAILOVER_STEP3 Background fargo drmdispatch PLANNED prefast_failover_step3 ALL NOTFORCED Background denver drmdispatch PLANNED prefast_failover_step3 ALL NOTFORCED END_SHORT_PLANNED_FAILOVER_STEP3 EXTENDED PLANNED FAILOVER STEP1 Background fargo drmdispatch PLANNED failover stepl ALL NOTFORCED Background denver drmdispatch PLANNED failover stepl ALL NOTFORCED END EXTENDED PLANNED FAILOVER STEP1 EXTENDED PLANNED FAILOVER STEP2 Background fargo drmdispatch PLANNED prefull failover step2 ALL NOTFORCED Background denver drmdispatch PLANNED prefull failover step2 ALL NOTFORCED END EXTENDED PLANNED FAILOVER STEP2 UNPLANNED_FAILOVER_STEP1 Background fargo drmdispatch DISASTER failover_stepl ALL NOTFORCED Background denver drmdispatch DISASTER failover_stepl ALL NOTFORCED END_UNPLANNED_FAILOVER_STEP1 UNPLANNED_FAILOVER_STEP2 Background fargo drmdispatch DISASTER failover_step2 ALL NOTFORCED Background denver drmdispatch DISASTER failover_step2 ALL NOTFORCED END_UNPLANNED_FAILOVER_STEP2 c 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Structure of the Application Action List Default Application Action List Continued FAST_FAILBACK_STEP1 Background fargo drmdispatch PLANNED fast_failback_stepl ALL NOTFORCED Backg
68. Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide C 3 Structure of the Application Action List Default Application Action List Continued ROLE_REVERSE_FAILOVER_STEP2 Background fargo drmdispatch PLANNED failover_step2 ALL NOTFORCED Background denver drmdispatch PLANNED failover_step2 ALL NOTFORCED END_ROLE_REVERSE_FAILOVER_STEP2 ROLE_REVERSE_FAILOVER_STEP3 Background fargo drmdispatch PLANNED failback_stepla ALL NOTFORCED Background denver drmdispatch PLANNED failback_stepla ALL NOTFORCE END_ROLE_REVERSE_FAILOVER_STEP3 ROLE_REVERSE_FAILOVER_STEP4 Background fargo drmdispatch PLANNED failback_steplb ALL NOTFORCED Background denver drmdispatch PLANNED failback_steplb ALL NOTFORCE END_ROLE_REVERSE_FATLOVER_STEP4 E ROLE REVERSE FAILBACK STEP1 Background fargo drmdispatch PLANNED failback step2 ALL NOTFORCED Background denver drmdispatch PLANNED failback step2 ALL NOTFORCE END ROLE REVERSE FAILBACK STI Pl 2o 3 ET_ERROR_NORMAL_OPERATION ackground tulsa drmdispatch PLANNED failsafe_op ALL NOTFORCED ackground greenbay drmdispatch PLANNED failsafe_op ALL NOTFORCED END SET ERROR NORMAL OPERATION UJ UJ un
69. Display 2 HSG gt show this_controller Controller HSG80 2G84906303 Software V87P Hardware E03 NODE_ID 5000 1FE1 0000 01F0 ALLOCATION_CLASS 0 SCSI_VERSION SCSI 2 Not configured for dual redundancy Device Port SCSI address 7 Time 11 MAY 2001 11 41 36 Command Console LUN is lun 0 IDENTIFIER 98 Host Connection Table is NOT locked Smart Error Eject Disabled Host PORT_1 Reported PORT_ID 5000 1FE1 0000 01F1 PORT_1_TOPOLOGY OFFLINE offline Host PORT_2 Reported PORT_ID 5000 1FE1 0000 01F2 PORT 2 TOPOLOGY OFFLINE offline NOREMOTE_COPY Cache 512 megabyte write cache version 0012 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER DEFAULT 10 seconds Mirrored Cache Not enabled Battery NOUPS FULLY CHARGED Expires 11 MAY 2003 gt c Configure the controllers for multiple bus failover by issuing the following command SET MULTIBUS_FAILOVER COPY THIS_CONTROLLER This command will automatically restart the OTHER controller gt d Optional Set the controller to the preferred SCSI mode with the following CLI command SET THIS_CONTROLLER SCSI_VERSION SCSI x NOTE x 2 or 3 SCSI 2 is the default setting Do not restart the controller gt e Designate a controller prompt with the following CLI commands SI THIS_CONTROLLER PROMPT InitiatorControllerNameTop gt E Fl xample set this_controller prompt buildngA Top gt SET OTHER_
70. FC ADAPTER ID 1000 0000 C921 3EFCHost 2 3 No rejected hosts Set all connections that were renamed back to their appropriate operating system with the following CLI command SET InitiatorHostConnectionName OP Tru64_UNIX VMS or WINNT Example set hostAl operating_system vms ERATING_SYSTEM IBM SUN After you have completed this action verify the connections with the following CLI command SHOW CONN ECTIONS You will see a display similar to that in Example Display 8 12 16 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure Example Display 8 BuildngATop gt show connections Connection Name Operating system Controller Port Address Status Offset BUILDNGBA PPRC_TARGET THIS 2 260413 OL this HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4254 BUILDNGBB PPRC_TARGET OTHER 2 200413 OL other HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4252 BUILDNGBC PPRC_INITIATOR THIS 2 offline HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4254 BUILDNGBD PPRC_INITIATOR OTHER 2 offline HOST_ID 5000 1FE1 0000 4250 ADAPTER_ID 5000 1FE1 0000 4252 HOSTA1 VMS THIS 260013 OL this HOST ID 1000 0000 C920 A7B9 ADAPTER ID 1000 0000 C920 A7B9 HOSTA2 VMS OTHER 200013 OL other HOST ID 1000 0000 C921 3F4E ADAPTER ID 1000 0000 C921 3F4E HOSTB1
71. G80 ACS Version 8 7P Scripting User Guide E DRM Power Up and Power Down This appendix describes the procedure for powering up and powering down your DRM systems Power Up Data Replication Manager Systems The procedures below describe how to power on and power off the storage subsystem after it has been configured target site before applying power to the initiator site Powering up in the wrong i CAUTION HP recommends that you power up the controllers and switches at the sequence may cause incorrect configurations Power on the DRM systems in the sequence described in the following procedures NOTE In this chapter procedures performed at the initiator or target sites show symbols at each step for reference The P symbol is used to identify a step performed at the initiator site and a symbol is used to identify a target site step Target Site Power Up Procedure 0000 1 Ensure that all enclosures switches and rack power distribution units PDUs have their power switches in the OFF position 2 Apply power to all PDUs 3 Turn on the power switches for the racks from the target site 4 Ensure that all controllers are on and functional 5 Apply power to all Fibre Channel switches When completed go to the Initiator Site Power Up Procedures Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide E 1 DRM Power Up and Power Down Initiator Site Power Up Procedure gt 1 En
72. ITE CACHED TRANSFER SIZE 32 Access BUILDNGBA BUILDNGBB BUILDNGBC BUILDNGBD HOSTA1 HOSTA2 State ONLINE to this controller Not reserved PREFERRED PATH THIS CONTROLLER Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 D2 DISK20000 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0135 NOIDENTIFIER Switches RUN NOWRITE PROTECT READ CACHE READAHEAD CACHE WRITEBACK CACHE MAX READ CACHED TRANSFER SIZE 32 MAX WRITE CACHED TRANSFER SIZE 32 Access BUILDNGBA BUILDNGBB BUILDNGBC BUILDNGBD HOSTA1 HOSTA2 State ONLINE to this controller Not reserved PREFERRED PATH OTHER CONTROLLER Host Based Logging NOT Specified Size 17769177 blocks Geometry C H S 5258 20 169 gt 6 The following steps requires actions relative to each operating system in your configuration a HP OpenVMS Allow the hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives 2 If the initiator site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 12 18 Data Replication Manager HSG80 ACS Version 8 7P Sc
73. L Example SET RCS1 ERROR_MODE NORMAL When the remote copy sets are in normal mode proceed to the next step gt 5 Continue the planned failover process with the Running the Role Reversal Failover Program File Procedure Running the Role Reversal Failover Program File Procedure 1 Open a command prompt window on the target host 2 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 3 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 13 3 Planned Site Role Reversal Procedure s 4 Run the hsg_fo xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failover to run Enter r for arole reversal failover You will see a confirmation message that asks you to confirm that a role reversal failover is desired Enter y for yes If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Press v for verbose or c for condensed To change the type of display you must delete the app fo rc file from the config
74. NE_HOME directory Run the hsg_fb xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter f for a full failback You will see a confirmation message that asks you to confirm that a full failback is desired Enter y for yes If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app fb1 rc file from the config subdirectory of the CLONE HOME directory and run the program file The display indicates when mirroring is complete At this time you are disaster tolerant and can operate in this mode until you choose to complete the failback process IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide 10 11 12 13 When you are ready to complete the failback to the original initiator site run the hsg fb2 xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating syste
75. ODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state INOPERATIVE Unit failsafe locked Target state BUILDNGB D1 is COPYING complete a Ifall remote copy sets are in failsafe error mode and 1 Every initiator state is failsafe locked proceed to step 5 2 Every initiator state is not failsafe locked proceed to step 5 13 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Planned Site Role Reversal Procedure 3 Initiator states are a mixture of failsafe locked or not being in failsafe locked proceed to step 4 b Ifall remote copy sets are in normal mode proceed to step 5 c If the remote copy sets are mixed between failsafe and normal error modes and 1 Every remote copy set in failsafe error mode is failsafe locked and every remote copy set in normal error mode has no targets proceed to step 5 2 Every remote copy set in failsafe error mode is not in a failsafe locked state proceed to step 5 3 Any remote copy sets in failsafe error mode are failsafe locked and remote copy sets in normal error mode have targets proceed to step 4 gt 4 Forremote copy sets in failsafe error mode with mixed initiator states or a mixture of error modes with failsafe error modes that are failsafe locked and normal error mode with targets change all the remote copy sets in failsafe mode to normal mode Use the CLI command SET RemoteCopySetName ERROR_MODE NORMA
76. ODE NORMAL Example SET RCS1 ERROR_MODE NORMAL When the remote copy sets are in normal mode proceed to the next step gt s Change to the bat sh or vms subdirectory in the CLONE_HOME directory 7 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode gt 6 Runthe hsg_op xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system gt 7 Youwillseea message that you are performing a DRM operation to modify error mode and asks whether normal or configured error mode is desired Enter n for normal mode gt 8 You will see a confirmation message that asks you to confirm that a normal error mode is desired Enter y for yes gt 9 If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app op rc file from the config subdirectory of the CLONE HOME directory and run the program file gt 10 When an operation completion status result is displayed similar to Figure 6 1 verify that the script removed the targets by entering the following command SHOW REMOTE COPY SETS FULL You will see a display similar to that in Exampl
77. RROR_MODE NORMAL Example SET RCS1 ERROR_MODE NORMAL When the remote copy sets are in normal mode proceed to the next step v Change to the bat sh or vms subdirectory in the CLONE HOME directory v Run the hsg_op xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system 11 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode gt 8 Youwillseea message that you are performing a DRM operation to modify error mode and asks whether normal or configured error mode is desired Enter n for normal mode gt 9 You will see a confirmation message that asks you to confirm that a normal error mode is desired Enter y for yes gt 10 If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app op rc file from the config subdirectory of the CLONE HOME directory and run the program file gt 11 When an operation completion status result is displayed similar to Figure 6 1 verify that the script removed the targets by entering the following command SHOW REMOTE COPY SETS FULL You will see a display similar to that in
78. Replication Manager HSG80 ACS Version 8 7P Scripting User Guide A 3 DRM Scripting Kit Files Table A 1 Installed DRM Scripting Kit Files Continued Directory Filename Description Tru64 UNIX IBM AIX and Sun Solaris files Default Hsgtru64 pm Perl module containing library files used by Perl scripts This file should not be modified Default Hsgibm pm Perl module containing library files used by Perl scripts This file should not be modified Default Hsgsun pm Perl module containing library files used by Perl scripts This file should not be modified SH gen_ex sh Example file to create a configuration generation file for each controller pair hsg fb1 sh Initiates a Perl script that performs step 1 of a planned site failback on controllers identified in the application action list This file should not be modified hsg fb2 sh Initiates a Perl script that performs step 2 of a planned site failback on controllers identified in the application action list This file should not be modified hsg fo sh Initiates a Perl script that performs a two step planned site failover on controllers identified in the application action list This file should not be modified hsg op sh Runs a Perl script to change the remote copy set error mode on the initiator controller This file should not be modified Windows only files Default Hsgwindows pm Perl module containing library files used by Perl scripts This file should not be modified
79. SG80 ACS Version 8 7P Scripting User Guide 13 5 Planned Site Role Reversal Procedure e Microsoft Windows 2000 Allow the hosts to recognize new units 1 If you have not changed the UNIT_OFFSET of any host connections since the hosts have been booted you do not need to reboot the target site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice 2 If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the host has rebooted log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice f Sun Solaris Allow hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and
80. Site Failover with Full Failback Procedure This chapter provides the procedure for performing an unplanned loss of the initiator site with a full failback to the existing hardware This procedure is used after experiencing an unplanned loss of the initiator site The loss could have been caused by a power failure or other event that did not damage initiator site hardware The duration of the outage at the initiator site is unknown After a failover to the target site a full failback is performed to the initiator site The sequence of steps are e Running the Unplanned Failover Program File Procedure e Target Host Setup Procedure e Running the Full Failback Program Files Procedure Initiator Site Cleanup Procedure NOTE In this chapter the P symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Running the Unplanned Failover Program File Procedure 1 Open a command prompt window on the target host 2 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 3 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 6 1 Unplanned Site Failover with Full Failback Procedure 4 Run the hsg_fo
81. Tru64 IBM AIX and Sun Solaris it is a shell sh file and for Microsoft Windows platforms it is a batch bat file Although scripting may make procedures easier to perform the operator must still be able to perform failover failback or failsafe mode transitions with CLI commands if the scripts encounter an abnormal condition that prevents their satisfactory completion Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 1 DRM Scripting Overview This guide explains e Failover and failback planning considerations e How to obtain and install the necessary files needed for using scripts e How to set up the Command Console LUN CCL e How to customize the following files to your DRM configuration Configuration generation batch shell and command files Target controller configuration files The application action list e How to run the failover failback and resumption of operation program files Site Failover Basic Description If the initiator site is no longer available or if there is anticipated downtime that will prevent operation at the initiator site you must decide whether to perform a site failover to the target site Performing a failover enables the target site to assume the role of the initiator and access write read data until the problem is resolved and a failback can be issued Transferring control of system operation to the target site ensures that there will be minimal interruptio
82. VMS THIS 220013 OL this HOST ID 1000 0000 C921 3E98 ADAPTER ID 1000 0000 C921 3E98 HOSTB2 VMS OTHER 250013 OL other HOST ID 1000 0000 C921 3EFC ADAPTER ID 1000 0000 C921 3EFC No rejected hosts Unit size to the value used on the original initiator Obtain your record of SHOW command output that details the original initiator configuration Using the output as a reference set the maximum cached transfer size to the original initiator value You can enhance host I O performance by resetting the maximum cached transfer 3 using the following CLI command InitiatorValue 32 SET UnitName MAXIMUM CACHED TRANSFER SIZE Example set dl maximum cache transfer size NOTE The default maximum cache transfer size is 32 Repeat this step for all remote copy set units After you have completed this action verify host access with the following CLI command SHOW UNITS FULL You will see a display similar to that in Example Display 9 Repeat this step for all units Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 17 Unplanned Site Failover with Failback to New Hardware Procedure Example Display 9 BuildngATop gt show units full LUN Uses Used by D1 DISK10000 BUILDNGA RCS1 LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0134 IDENTIFIER 1 Switches RUN NOWRITE_PROTECT READ_CACHE READAHEAD CACHE WRITEBACK CACHE MAX READ CACHED TRANSFER SIZE 32 MAX WR
83. _READ_CACHED_TRANSFER_SIZE D1 32 D1 1 D2 128 D2 1 MAX WRITE CACHED TRANSFER SIZE MAX WRITE CACHED TRANSFER SIZE D1 32 D1 1 D2 128 D2 1 With a text editor modify the values in the target configuration file to match the values of the initiator as shown below MAX READ CACHED TRANSFER SIZE D1 32 D2 128 MAX_WRITE_CACHED_TRANSFER_SIZE D1 32 D2 128 4 22 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization Application Action List Customization During installation the default application action list app_ex act was extracted from the DRM Scripting Kit and placed in the config subdirectory of CLONE_HOME It provides a basic structure that you must customize using the procedure below The customized file is saved as app act and becomes the application action list used by all hosts running the scripts For a discussion of the structure of the application action list and details of the Perl scripts that use this file refer to Appendix C Customizing the Application Action List The following steps are provided as guidelines to prepare your application action list 1 2 3 Rename the app_ex act file to app act in the CONFIG directory With a text editor open the app act file Identify the number of DRM initiator target controller subsystems in your DRM configuration and populate the number of actions to correspond with the number of DRM su
84. able C 1 Structure of an Action Command Foreground Indicates whether the action line must be executed in Background the foreground or background If in the foreground the hsgcontrol pl script executes that action line and waits until the Perl script is finished before continuing with the next line If in the background the hsgcontrol pl script starts a background process for the action line and continues processing the next action line ControllerName The name of the storage subsystem on which the action is performed This parameter is passed without any checking or case conversion to the Perl script PerlScriptName The name of the Perl script Parameters 1 2 3 These four parameters are used by the Perl script and 4 without any parsing or case conversion In an action they appear in the following order parameter 1 failure type parameter 2 control table parameter 3 remote copy sets processed parameter 4 condition clearing Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide C 5 Structure of the Application Action List How the Perl Scripts Use the Application Action List Two important Perl scripts are responsible for invoking failover failback and resumption of operations The hsgcontrol pl script reads the application action list app act and passes parameters to the drmdispatch pl script The drmdispatch pl script reads the control tables and executes the steps for failover fa
85. ailback and resumption of operation scripts e Contact technical support for additional assistance Intended Audience This book is intended for use by DRM customers who are experienced with the following e Configuring a DRM environment to include zoning write history logs association sets and remote copy sets e Running failover failback and resumption of operations with command line interface CLI commands Related Documentation In addition to this guide HP provides corresponding information Table 1 Related Documentation Document Title Part Number Compaq SANworks Command Scripter Version 1 0A AA RN6HB TE Release Notes Compaq SANworks Command Scripter Version 1 0A AE RN6FB TE Installation Card Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide ix About this Guide Table 1 Related Documentation Continued Document Title Part Number Compaq SANworks Command Scripter Version 1 0A AA RN6EB TE User Guide HP StorageWorks Data Replication Manager HSG80 ACS AA RPHZE TE Version 8 7P Configuration Guide HP StorageWorks Data Replication Manager HSG80 ACS AA RPJOD TE Version 8 7P Failover Failback Procedures Guide HP StorageWorks HSG80 Array Controller ACS EK G80CL RA B01 Version 8 7 CLI Reference Guide HP StorageWorks HSG80 Array Controller V8 7 EK G80TR SA B01 Troubleshooting Reference Guide Prerequisites Before you run the DRM scripts make sure you consider the items
86. ailover during a short planned failover action then you would list the four target controllers in the application action list All of the controllers will fail over together when that failover action is run The configuration files and the application action list are system specific You must tailor them to reflect your unique configuration and your failover failback and resumption of operation preferences Chapter 4 File Customization provides instructions for modifying configuration files and application action lists These files can then be used by the scripts to perform the necessary steps Running a Script A script is invoked by running a command file OpenVMS a shell file AIX Solaris and Tru64 or a batch file Windows NT 2000 from a command prompt on the system console Figure 1 1 shows the scripting information flow after you run one of these program files 1 The Perl interpreter processes the script based on the information in the configuration file and the application action list 2 The script reads the control table which defines the order of CLI commands to be issued and sends the appropriate sequence of CLI commands for the controller configuration specified in the configuration file to the Command Scripter 3 The Command Scripter communicates the commands to the HSG80 controller over the Fibre Channel bus and relays SHOW command verification back for the scripts The area in Figure 1 1 within the dashed
87. alled correctly you will see each unit twice 2 If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice f Sun Solaris Allow the hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance gt 2 Continue the resumption of operations process by Running the Resumption of Operations Program File Procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 7 7 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode
88. and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 19 Unplanned Site Failover with Failback to New Hardware Procedure 2 If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice f Sun Solaris Allow the hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance This completes the procedure
89. arget site Save these batch files in the BAT subdirectory on the target host using the same naming convention used previously Sun Solaris Solaris can run using the CCL in SCSI 2 or SCSI 3 mode with Secure Path Version 2 1D Solaris cannot use the CCL with Secure Path Version 3 0 so it must use anon RCS LUN The following procedures show how to use either method Using the CCL 1 Modify the gen_ex sh file to create an initiator controller configuration generation file for each host running the scripts at the initiator site Use a text editor and make the necessary modifications using the following syntax perl I CLONE_HOME CLONE_HOME bin generate_cfg pl com cs RemoteCopyName idnum where RemoteCopyName identifies the DRM controller remote copy name of the subsystem for example tulsa This name is case sensitive in Sun Solaris and must match the case of the name used later when configuring the application action list idnum specifies the device number for the operating system that Command Scripter uses to communicate with controller tulsa As shown in Figure 4 1 this is the device number of the CCL for the local controller Controller A communicating with the initiator host Host A Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 13 File Customization To find the idnum use the format command to find the device assigned to the CCL For example the format entry display
90. aris Perl Interpreter Follow the installation instructions located on the sunfreeware website listed above SANworks Command Scripter The Command Scripter component provides the interface for the Perl scripts to communicate with the HSG80 controller via the Fibre Channel bus Obtaining SANworks Command Scripter To obtain the Command Scripter contact a reseller or HP account representative Refer to HP Authorized Reseller in the About This Guide section for source information For DRM users with a previous version of the Command Scripter check for updates at the following website http www comag com products sanworks softwaredrivers commandscripter index htm Installing SANworks Command Scripter This section describes the procedures for installing the Command Scripter on OpenVMS Tru64 AIX Windows NT 2000 and Solaris The Command Scripter must be installed on the initiator and target site servers where the scripts reside Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 2 5 Installation HP OpenVMS The following procedure installs the Command Scripter on a OpenVMS server 1 Mount the CD ROM Use the command MOUNT OVERRIDE ID DKA400 Create a directory for the Command Scripter files The following command creates a directory named CMDSCRPT CREATE DIRECTORY SYSSSYSDEVICE CMDSCRPT Set the directory created in the previous step as t
91. ator and target controller configuration files Execute the configuration generation file for each controller subsystem See Running Configuration Generation Files on page 4 17 Use pertinent information from the initiator configuration files to copy into or otherwise modify the respective target configuration files See Target Controller Configuration File Customization on page 4 19 Customizations to the target configuration files are Copy association set information from initiator configuration files to the target configuration files Copy remote copy set information from the initiator configuration files to the target configuration files Identify and add target site hosts that are granted access to remote copy set units following a failover Modify the read and write cache values in the target configuration files to maintain consistent values with the initiator configuration files Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 1 File Customization e Modify the application action list app_ex act file to specify all DRM initiator target controller pairs that you want to execute concurrently in the SAN See Application Action List Customization on page 4 23 IMPORTANT The names of remote copy sets stripesets mirrorsets RAlDsets association sets and connection names may not contain a hyphen This is a Perl restriction Underscores _ are allowed
92. ator site is remedied continue the full failback procedure to new hardware at the initiator site with the Initiator Site Preparation Procedure Initiator Site Preparation Procedure gt 1 Follow the step listed below for the operating system in your configuration Refer to Appendix E for DRM system power up or power down procedures a HP OpenVMS If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O to the LUNs with remote copy sets that will be failed over then dismount the volumes associated with these LUNs HP Tru64 UNIX If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O and unmount all file system LUNs that have remote copy sets that will be failed over IBM AIX If the operating system is running remove all I O to the remote copy set LUNs that will be failed over then unmount the file systems associated with these LUNs 12 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure d Microsoft Windows NT X86 If the operating system is running shut it down and power off the hosts e Microsoft Windows 2000 If the operating syst
93. ay cause a failure indication even though the scripts performed successfully This completes the procedure for resuming initiator site processing after an extended planned loss of the target Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 11 5 Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode 11 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 Unplanned Site Failover with Failback to New Hardware Procedure This chapter provides the procedure for performing an unplanned site failover of the initiator site with a full failback to new hardware This procedure is used when some type of disaster lightning flood fire or the like forces a failover to the target site When the damaged components at the initiator site hosts controllers switches for example have been repaired and the site is operational and back online then a failback is performed The sequence of steps are Running the Unplanned Failover Program File Procedure e Target Host Setup Procedure e Initiator Site Preparation Procedure e Running the New Hardware Failback Program Files Procedure Initiator Site Cleanup Procedure NOTE In this chapter the P symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Running the Unplanned Failover Program File Procedure
94. below e This manual assumes a DRM configuration is in place before scripts are used Refer to the platform hardware and software requirements discussed in Chapter 1 that support scripting e HP strongly recommends that detailed functional testing be performed on all scripts before they are used operationally x Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide About this Guide Document Conventions The conventions shown in Table 2 apply in most cases Table 2 Document Conventions Element Convention Key names menu items buttons Bold and dialog box titles File names and application names Italics User input command names system Monospace font responses output and messages COMMAND NAMES are uppercase unless they are case sensitive Variables Monospace italic font Website addresses Sans serif font http www hp com Symbols in Text These symbols may be found in the text of this guide They have the following meanings WARNING Text set off in this manner indicates that failure to follow directions in the warning could result in bodily harm or loss of life CAUTION Text set off in this manner indicates that failure to follow directions could result in damage to equipment or data IMPORTANT Text set off in this manner presents clarifying information or specific instructions NOTE Text set off in this manner presents commentary sidelights or interesti
95. bols indicates that the component exceeds the recommended weight for one individual to handle safely WARNING To reduce the risk of personal injury or damage to the equipment observe local occupational health and safety requirements and guidelines for manually handling material Xii Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide About this Guide Rack Stability sure that p WARNING To reduce the risk of personal injury or damage to the equipment be The leveling jacks are extended to the floor The full weight of the rack rests on the leveling jacks in single rack installations the stabilizing feet are attached to the rack Inmultiple rack installations the racks are coupled Only one rack component is extended at any time A rack may become unstable if more than one rack component is extended for any reason Getting Help If you still have a question after reading this guide contact service representatives or visit our website HP Technical Support In North America call HP technical support at 1 800 652 6672 available 24 hours a day 7 days a week NOTE For continuous quality improvement calls may be recorded or monitored Outside North America call HP technical support at the nearest location Telephone numbers for worldwide technical support are listed on the HP website http thenew hp com Be sure to have the following information available before cal
96. bsystems The number of initiator target subsystems will match the number of entries for each action For example two DRM subsystems will comprise two entries under the action SHORT_PLANNED_FAILOVER_STEPI For each entry within a failover or failback action section modify the controller name to that of the target controller name for each DRM initiator target pair For each entry containing failsafe lock actions SET_ERROR_NORMAL_OPERATION and SET_ERROR_CONFIGURED_OPERATION modify the controller name to that of the initiator controller name for each DRM initiator target pair Save the app act file and copy it to the other site Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 23 File Customization Example Customization of an Application Action List Assume that you have three DRM initiator target controller pairs in your system called sun moon mars venus and jupiter saturn 1 Rename the app_ex act file to app act at the initiator site 2 Atthe initiator site open the app act file with a text editor and start with the action called SHORT PLANNED FAILOVER STEPI What you see is SHORT PLANNED FAILOVER STEPI Background fargo drmdispatch PLANNED failover stepl ALL NOTFORCED Background denver drmdispatch PLANNED failover stepl ALL NOTFORCED END SHORT PLANNED FAILOVER STEPI NOTE This is what is seen in the actual app ex actfile The names fargo and denve
97. can Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice 2 If you have changed the UNIT OFFSET of any host connections you must reboot that host After the host has rebooted log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice f SunSolaris Allow hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges 10 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Extended Planned Site Failover with Full Failback Procedure 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance 3 After the completion of maintenan
98. cation Manager HSG80 ACS Version 8 7P Scripting User Guide 8 Resumption of Operations After Unplanned Loss of Target Site Procedure Normal Mode This chapter provides the procedure for performing a resumption of operations after an unplanned loss of the target site while running in normal error mode This procedure is used when the initiator remote copy sets are in normal error mode and an unplanned loss of the target occurs Remote copy sets to the target site are removed All of the steps in this procedure are performed from the initiator site The sequence of steps are e Verification of Lost Connections Procedure e Running the Resumption of Operations Program File Procedure NOTE In this chapter the P symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Verification of Lost Connections Procedure gt 1 Verify that the connection to the target site is lost If you are connected to the initiator site controllers when connection to the target site is lost you will see a confirmation message on your terminal as shown in Example Display 1 Refer to Appendix D for the meaning of the instance code Example Display 1 BuildngATop gt EVL BuildngATop gt 06 JUN 2001 12 57 13 Instance Code OEOF8BO1 Template 144 90 Occurred on 06 JUN 2001 at 12 57 13 Power On Time 2 Years 72 Days 5 Hours 31 Minutes 56 Seconds C
99. ce eee eee 6 7 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode Verification of Lost Connections Procedure 2 0 0 cee eee eee eee eee 7 1 Running the Resumption of Operations Program File Procedure 7 3 Initiator Site Cleanup Procedure 20 0 eee cee eee eee 7 6 Running the Resumption of Operations Program File Procedure 7 8 Resumption of Operations After Unplanned Loss of Target Site Procedure Normal Mode Verification of Lost Connections Procedure 2 0 0 cece eee eee eee eee 8 1 Running the Resumption of Operations Program Files Procedure 8 2 Short Planned Site Failover with Fast Failback Procedure Initiator Site Preparation Procedure 0 2 0 cece eee eee 9 1 Running the Short Planned Failover Program File Procedure 9 2 Target Host Setup Procedure 0 eee eee e 9 3 Running the Fast Failback Program Files Procedure 0 002 e ee eee 9 6 Initiator Site Cleanup Procedure 0 0 cece eee eee 9 8 10 Extended Planned Site Failover with Full Failback Procedure 11 Initiator Site Preparation Procedure 0 0 0 0 eee eee eee eee 10 1 Running the Extended Planned Failover Program File Procedure 10 4 Target Host Setup Procedure 0 eee eee ees 10 5 Running the Full Failback Program Files Procedure 222222220200 10
100. ce work at the initiator site continue with the Running the Full Failback Program Files Procedure Running the Full Failback Program Files Procedure 1 Before performing the failback procedure locate your record of SHOW command output that details the initiator configuration Verify that your initiator controller configuration is the same as your target controller configuration Compare the status of the controllers association sets remote copy sets units and connections Refer to the HP StorageWorks Data Replication Manager HSG80 ACS 8 7P Failover Failback Procedures Guide for the full status comparison procedure Open a command prompt window on the target host Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Run the hsg_fb xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter f for a full failback You will see a confirmation message that asks you to confirm that a full failback is desired Enter y for yes Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 7 Extended Planned Site Failover with Full Failbac
101. ck setup exe 2 From the Welcome screen click Next 3 The license agreement displays Click Yes to accept the license agreement 4 Accept the default or choose a destination for the program installation Click Next 5 Click Finish A Command Scripter program icon is added to the Programs menu 6 Copy cmdscript exe to JCLONE_HOME BIN the subdirectory under the default directory where the DRM scripting files reside for example C scripts bin 7 The Windows version of Command Scripter requires that a controller be configured on each storage subsystem to direct the inband Fibre Channel data To test the connection use the following procedure for each controller a Go to the Windows command prompt b Switch to the CLONE_HOME bin directory where CLONE_HOME is the name given to the default scripts directory Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Installation 8 c Enter the following CLI command cmdscript f DeviceName show this_controller DeviceName is the LUN communicating with the controller It can be a drive letter for example Q or a string in the format Scsi3 1 124 0 Refer to Communicating Via Command Scripter on page 5 2 You will see the expected SHOW command response from the controller Take note of what drive letters correspond to each controller name You will need these for later configuration tasks Installation is complete Sun S
102. command from the initiator and target hosts CMDSCRIPT f 1 GGAValue SHOW THIS CONTROLLER with Value being the ID number of the initiator or target CCLs HP OpenVMS Job Queue Setup If you have an existing batch queue you would like to use for the scripts then define the logical name CLONE QUEUE to point to this queue For example use the command DEFINE SYSTEM CLONE QUEUE HostName BATCH If you do not have a batch queue set up on your system refer to the VMS System Manager s Manual and define one After defining a batch queue point to it using the step above HP Tru64 UNIX CCL Setup Tru64 is supported in SCSI 2 and SCSI 3 mode The following steps enable the CCL and configure it on the Tru64 UNIX platform 1 To enable the controller CCL in SCSI 2 mode enter the following commands from each controller subsystem SE HIS_CONTROLLER SCSI_VERSION SCSI 2 SE HIS_CONTROLLER COMMAND_CONSOLE_LUN To enable the CCL in SCSI 3 mode enter the following command from each controller subsystem SET THIS_CONTROLLER SCSI_VERSION SCSI 3 3 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide CCL Setup 2 Verify the CCLs can be seen by the host by entering this command from the host hwmgr view device The following is a sample output for this command HWID Device Name Mfg Model Location 4 dev dmapi dmapi 5
103. correctly set up For example check that you have an association set with a log disk for the initiator site that all remote copy sets belong to an association set a log disk is available for the target site or any other desired setup properties are enabled Also check that each unit has a preferred path at both the initiator and target sites IMPORTANT Any time your controller configuration changes the controller configuration generation program files you create will have to be run again to create new controller configuration files HP OpenVMS The following procedure describes how to create initiator and target configuration generation files in SCSI 3 mode when using a CCL 1 Modify the gen_ex com file to create an initiator configuration generation file for the initiator site Use a text editor and make the necessary modifications using the following syntax perl I CLONE HOME BlN generate cfg pl com cs RemoteCopyName idnum where Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 3 File Customization RemoteCopyName identifies the initiator DRM controller remote copy name of the subsystem for example tulsa The remote copy name can be obtained from running a SHOW THIS command from the controller idnum specifies the device identification number for the operating system that Command Scripter uses to communicate with initiator controller tulsa As shown in Figure 4 1 this is the identification
104. ct 2211 drm_grant_remote_hsg Configuration file problem Update access cannot read configuration file configuration entry for remote copy set 2212 drm_grant_remote_hsg Configuration file problem Node access remote copy set does specified is not an initiator or target not contain this controller Update configuration file 2220 parse_connection_info Communication error Identify unknown header information communication problem and correct 2221 parse_connection_info Communication error Identify cannot parse connection communication problem and correct information 2231 change_rcs_characteristic Control table problem Table has been cannot change characteristic possibly corrupted Restore control due to incorrect value table 2232 change_rcs_characteristic Problem is with remote copy set on cannot change characteristic controller Correct error on controller for remote copy set on HSG as needed controller D 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Troubleshooting Confirmation Message Instance Codes In the examples provided in the scripting failover failback and resumption of operation procedures some controller displays show confirmation messages with the event log symbol EVL and an instance code Information about instance codes and their meanings can be found in the HP StorageWorks HSG80 Array Controller Version 8 7 Troubleshooting Reference Guide
105. d Check unit and underlying storage for error possible problems Correct any error conditions detected on the controller D 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Troubleshooting Table D 1 Scripting Error Codes Continued Error Code 1240 get_subclone_list source A mismatch exists between unit does not exist configuration file and the actual controller configuration Check unit section or clone section for discrepancies and repair 1241 get_subclone_list cannot Verify connectivity with the controller read source unit and repair error conditions as necessary Check unit specified in the error message and verify that it is a valid unit on the controller unit 1320 reconstructing cannot read Verify specified mirror set on controller mirror properties and correct as necessary 1242 9eget subclone list cannot Configuration problem on unit match storage type on source specified Update configuration file 1422 cli cannot send command Communications error Identify communications problem and correct 1500 get cannot match show Communications error Identify output to storage type communications problem and correct 1520 9eparse mirror info unknown Communications error Identify header information communications problem and correct 1521 9eparse mirror info cannot Communications error Identify parse mirror name communications problem and correct 1524 9eparse mirro
106. d from a single disk drive or tape drive A unit may also be created from a more complex container such as a RAIDset The controller supports a maximum of eight units on each target unplanned failover As applied to the Data Replication Manager an unplanned outage of the controllers This may occur when the site communication is lost or due to some other failure whereby remote copy sets cannot be implemented The controllers do not perform an orderly shutdown See also planned failover verbose display A screen output displayed while the failover and failback scripts run The display shows the status of the controller and remote copy sets A user preference for this type of display is set in a resource RC file created by the failover and failback batch files See also condensed display write history logging As applied to the Data Replication Manager the use of a log unit to record a history of write commands and data from the host Write history logging is used for mini merge and fast failback See also mini merge fast failback Glossary 12 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide A action command structure C 5 ActivePerl 2 4 app act file 4 23 4 24 C 6 app_ex actfile 4 2 4 23 4 24 A 1 C 1 application action list 1 10 customization 4 23 defined 1 10 structure C 5 applications names convention defined xi Array Controller Software ACS 1 16 Association Set Section customization 4 19
107. d of a drive letter this is the device identifier of the LUN for the target controller Controller B communicating with the initiator host Host A 4 Save this edited configuration generation file in the sh subdirectory with a meaningful name like fargo_gen sh 5 Repeat the above steps to create initiator and target controller configuration generation program files for each host running the scripts at the target site Use the correct device identification for the non RCS LUNs communicating with the controllers Save these edited files in the sh subdirectory on the target host using the same naming conventions used previously Running Configuration Generation Files After creating initiator and target configuration generation files for each controller subsystem at each site execute each file on the initiator and target hosts running the scripts For example the initiator configuration generation file for controller tulsa is run from the initiator site hosts and a configuration generation file for controller tulsa is run from the target hosts Remember that any time your controller configuration changes these generation files must be run again to create new controller configuration files The configuration generation files run a Perl script called generate_cfg pl When this script runs e Many SHOW commands are sent to the applicable HSG80 controller The script creates a controller configuration file based on the received responses
108. d pane If Secure Path is not installed correctly you will see each unit twice 2 If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You should be able to see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice f Sun Solaris Allow the hosts to recognize new units 1 Reboot the servers using the command reboot r at the initiator site and log in using an account that has administrative privileges 10 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Extended Planned Site Failover with Full Failback Procedure 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance This completes the procedure for running an extended planned site failover with a full failbac
109. dentify communication problem and correct Mismatch between configuration file and controller Connection no longer exists on controller Update configuration file Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Troubleshooting Table D 1 Scripting Error Codes Continued Error coas weaning aeron 2090 drm_deny_server_access Mismatch between configuration file connection configured for unit and controller Connection no longer does not exist in controller exists on controller Update configuration file 2100 parse_associationset_info Communication error Identify unknown header information communication problem and correct 2101 parse_associationset_info Communication error Identify cannot parse association set communication problem and correct name 2110 parse_remotecopyset_info Communication error Identify unknown header information communication problem and correct 2111 parse_remotecopyset_info Communication error Identify cannot parse remote copy set communication problem and correct name 2120 drm_read_associationset_ No association set defined in config association set configuration file Update configuration information configuration file missing 2130 drm_read_remotecopyset_ No remote copy set defined in config remote copy set configuration file Update configuration information configuration file missing 2140 drm_change_hostport_ Failover or failback control
110. e new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance This completes the role reversal failback procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 13 9 Planned Site Role Reversal Procedure 13 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide A DRM Scripting Kit Files This appendix provides a list of files that are extracted from the DRM Scripting Kit The table below lists the kit files provided their location in the CLONE_HOME subdirectory and gives a brief description Table A 1 Installed DRM Scripting Kit Files Directory Filename Description Default Hsgcs pm Perl modules containing library files used by Perl Hsgcustom pm scripts These should not be modified Hsgdrm pm Hsggen pm drmdispatch pl Generates command line interface CLI commands to perform failover or failback This file should not be modified generate_cfg pl Creates a controller configuration file and saves it with a ControllerName cfg file name This file should not be modified
111. e 1 16 Maximum Read Write Cached Transfer Size Section customization 4 22 example B 2 mda conf 3 6 menu items convention defined xi sequences convention defined xi Microsoft Windows NT 2000 CCL setup 3 5 Command Scripter installation 2 8 creating configuration files 4 9 DRM Scripting Kit description 1 8 installation 2 3 environmental variable 2 3 non RCS LUN 4 1 Perl interpreter 1 16 platform requirements 1 14 program file 1 1 running configuration generation files 4 18 script termination 5 6 SCSI 2 mode 1 15 4 9 SCSI 3 mode 1 15 3 5 4 11 using the CCL 4 11 N non RCS LUN 1 14 4 1 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Index 3 Index note defined xi P Perl interpreter 1 8 to 1 10 1 12 1 16 2 4 description 1 9 for IBM AIX 1 16 for OpenVMS 1 16 2 4 for Sun Solaris 1 16 2 5 for Tru64 UNIX 1 16 for Windows NT 2000 1 16 2 4 installing ActivePerl 2 4 installing for OpenVMS 2 4 installing for Sun Solaris 2 5 restrictions 4 2 Perl modules 1 13 platforms supported 1 14 power down procedures initiator site E 2 target site E 3 power up procedures initiator site E 2 target site E 1 prerequisites x R RC file deleting 5 4 location 5 4 related documentation ix remote copy name 4 4 4 6 4 8 4 9 4 11 Remote Copy Set Section customization 4 20 example B 2 requirements hardware 1 15 platforms 1 14 software 1 16 resumption of operations defined 1 1 scenarios
112. e Display 4 Example Display 4 BuildngATop show remote copy sets full Name Uses Used by RCS1 remote copy D1 Reported LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0134 Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE NORMAL FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to the other controller No targets gt 11 Continue with the Initiator Site Cleanup Procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 7 5 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode Initiator Site Cleanup Procedure gt 1 The following steps requires actions relative to each operating system in your configuration a HP OpenVMS Allow hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives 2 If the initiator site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 If the initiator hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI b
113. e This resulting configuration file framework is named by the script in the format ControllerName cfg In the examples above the tulsa_gen xxx program file creates a configuration file called tulsa cfg provided that tulsa is the controller name used in the configuration generation bat sh or com file and places it in the config subdirectory of the platform s CLONE_HOME directory The following steps outline how the configuration generation files run on each platform Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 17 File Customization HP OpenVMS 1 Locate the configuration generation files in the CLONE HOME vms directory 2 Run the command file for the first controller These files must be run from a privileged account 3 Continue to run the configuration generation files for each initiator and target controller on the initiator and target hosts HP Tru64 UNIX IBM AIX and Sun Solaris 1 Locate the configuration generation files in the CLONE_HOME sh directory 2 Run the shell file for the first controller NOTE It may be necessary to change the file permissions to make them executable 3 Continue to run the configuration generation files for each initiator and target controller on the initiator and target hosts Microsoft Windows NT 2000 1 In Windows Explorer locate the configuration generation batch files in the CLONE_HOME bat directory 2 Double click or run the batch fil
114. e for the first controller 3 Continue to run the configuration generation batch file for each initiator and target controller on the initiator and target hosts 4 18 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization Controller Configuration File Customization The configuration files that are created after running the generation files and placed in the config subdirectory represent a picture in time of the controller configuration The information in these files enables the Perl scripts to issue the correct commands to the necessary devices The sections in these files have names like ASSOCIATIONSET CONNECTIONS CONTROLLER and so on Appendix B shows an example of a controller configuration file After the configuration files have been generated you will have files for both the initiator and target controllers residing on each host Please note that e The configuration files for the initiator controllers are complete and do not have to be modified e The configuration files for the target controllers must be modified each time they are created This is done to put them into a state that should exist after failover e The configuration file for any controller must be recreated by rerunning the configuration generation executable file every time there are changes made to the configuration of that controller Target Controller Configuration File Customization The target controller conf
115. e operation completion status result is displayed similar to Figure 6 1 verify that the error mode is failsafe that the initiator state is online and the target state is normal by entering the following command SHOW REMOTE COPY SETS FULL You will see a display similar to that in Example Display 5 7 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode Example Display 5 BuildngATop gt show remote_copy_sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to this controller Not reserved Target state BUILDNGB D1 is NORMAL IMPORTANT If the operation completion status indicates a failure check the remote copy sets to see if they are in a failsafe mode and their target state is normal Sometimes the screen refresh may cause a failure indication even though the scripts performed successfully This completes the procedure for resuming initiator site processing after an unplanned communication loss while in failsafe mode Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 7 9 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode 7 10 Data Repli
116. e the type of display you must delete the app fb2 rc file from the config subdirectory of the CLONE HOME directory and run the program file Continue with the full failback procedure at the initiator site with Initiator Site Cleanup Procedure 10 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Extended Planned Site Failover with Full Failback Procedure Initiator Site Cleanup Procedure gt 1 Verify that the initiator host can connect to the remote copy set units with this command SHOW UNITS FULL In the Access field of the display all units that are used by remote copy sets should show that the initiator hosts are enabled gt 2 The following steps require actions relative to each operating system in your configuration a HP OpenVMS Allow hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives 2 If the initiator site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 If the initiator hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for la
117. ea 3 1 HP OpenVMS Job Queue Setup 0 cee nen 3 2 HP Tru64 UNIX CCE Setup io Ree RR REDIERE EE SEI 3 2 IBM ATX CCL S tup oc Reve ae be De ERR UI CR e Es 3 4 Microsoft Windows NT 2000 CCL Setup 0 0 0 cee e 3 5 iv Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Contents Sun Solaris CCL Setup Sog sere Et Soe et Eee tae PES es 3 5 kernel drv sd conf ese cel ee eme ee eese etn 3 6 Entries to mda conf and LdLite conf 0 0 0 0 eee eee 3 6 4 File Customization Introduction sies a mis ae ar ra Es 4 1 File Customization Steps se renatin cece cece ee eee ene nee 4 1 Configuration Generation File Creation 2 0 0 0 eee eee eee eee 4 3 HP OpenVMS i dura ean he We RES ii 4 3 HP Tr 64 UNIX Ll eng ren aa 4 6 IBM AIR cb een NE die egi Pere tegrta 4 7 Microsoft Windows NT 2000 0 cee cece eee nee 4 9 SCSI 2 Mode with No CCL Enabled 0 0 0 0 000 eee ee eee 4 9 SCSI 3 Mode with CELS 22 2 2 22 Liv Ree eee RW 4 11 Sun Solaris eee IR PUE et feat iiie thee 4 13 Usingthe CL 2 4 4 35 ae eee ertet eri 4 13 Using the Non RCS LUN No CCL sseeeeeeeeee eee 4 15 Running Configuration Generation Files lleeeeeee eee 4 17 HP OpenVMS ach ans anna usq RP Rer t osa dfe 4 18 HP Tru64 UNIX IBM AIX and Sun Solaris 0 2 0 0 000000005 4 18 Microsoft Windows NT 2000 2 0 eee III 4 18 Controller Configuration File Customi
118. eck the remote copy sets to see if they are in a normal mode and their target state is normal Sometimes the screen refresh may cause a failure indication even though the scripts performed successfully This completes the procedure for resuming initiator site processing after an unplanned communication loss while in normal mode 8 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 9 Short Planned Site Failover with Fast Failback Procedure This chapter provides the procedure for performing a short planned site failover with a fast failback This procedure ensures the proper functioning of a failover and subsequent failback during short duration maintenance activities When a planned site failover is used with a fast failback each remote copy set must be in an association set with a write history log enabled Because of the short duration of the planned outage the write history log will be able to accommodate the accumulated writes These are the sequence steps Initiator Site Preparation Procedure Running the Short Planned Failover Program File Procedure Target Host Setup Procedure Running the Fast Failback Program Files Procedure Initiator Site Cleanup Procedure NOTE In this chapter the gt symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Initiator Site Preparation Procedure gt a Before performing t
119. ectly you will see each drive twice f SunSolaris Allow the hosts to recognize new units 1 Reboot the servers using the command reboot r at the target site and log in using an account that has administrative privileges Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 9 5 Short Planned Site Failover with Fast Failback Procedure 2 If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find the new WWIDs of the units use the SHOW UnitName command on the controller You may also need to adjust the kernal drv mda conf and kernal drv sd conf files to accommodate for extra LUNs After editing the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance 4 Continue the planned failback procedure at the target site with the Running the Fast Failback Program Files Procedure Running the Fast Failback Program Files Procedure 1 Before performing the failback procedure locate your record of SHOW command output that details the initiator configuration Verify that your initiator controller configuration is the same as your target controller configuration Compare the status of the controllers
120. ed would look like the following c1t10040 drive type unknown So for this example the entry for idnum would be dev rdsk c1t100d0s2 where s2 is the slice number of the drive Save the edited generation file into the CLONE_HOME sh subdirectory with a meaningful name like tulsa_gen sh Modify the gen_ex sh file again to create a target configuration generation file for the initiator site The device identification number used in this file specifies the device identification number for the CCL of the controller on the target site Controller B communicating with the initiator host Host A Use the syntax perl I CLONE_HOME SCLONE_HOME bin generate_cfg pl com cs RemoteCopyName idnum where RemoteCopyName identifies the target DRM controller remote copy name of the subsystem for example fargo This name is case sensitive in Sun Solaris and must match the case of the name used later when configuring the application action list idnum specifies the device number for the operating system that Command Scripter uses to communicate with controller fargo As shown in Figure 4 1 this is the device number of the CCL for the remote controller Controller B communicating with the initiator host Host A Save this program file in the sh subdirectory with a meaningful name like fargo_gen sh Repeat the above steps to create initiator and target configuration generation program files for all hosts running the scrip
121. efers to storagesets single disk units and all other storage devices that are installed in a subsystem and accessed by the host A storage unit can be any entity that is capable of storing data whether it is a physical device or a group of physical devices StorageWorks Command Console SWCC A graphical user interface GUI that provides local and remote management of StorageWorks controllers It is a tool for monitoring configuring and troubleshooting storage subsystems SWCC issues commands and interprets the responses sent by the controller The user interface displays the logical and physical layout and status of a selected subsystem in graphical form surviving controller The controller in a dual redundant configuration pair that serves its companion s devices when the companion controller fails swcc See StorageWorks Command Console switch fabric 1 The internal interconnect architecture used by a switching device that redirects the data coming in on one of its ports out to another of its ports 2 The combination of interconnected switches used throughout a campus or large geographic area which collectively provide a routing infrastructure synchronous mode A mode of operation of the remote copy set whereby the data is written simultaneously to the cache of the initiator subsystem and the cache of the target subsystem The I O completion status is not sent until all members of the remote copy set are updated See also async
122. elow Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 Verify that all configuration files are current for each controller pair Ensure that commands and controller names used in case sensitive platforms like HP Tru64 IBM AIX and Sun Solaris are consistent For example a configuration generation file may list a controller as tulsa so the app act file must list it with the same lower case Otherwise the script will not execute Ensure that all target configuration files with manually updated sections are correctly updated with matching fields from the corresponding initiator configuration file Also ensure that no spaces line feeds returns and so on were manually added to the updated sections These may cause the script to hang Ensure that application action files follow correct formatting and that they use correct DRM controller remote copy names Failover and failback actions use the target controller name in the action statement Resumption of operation actions use the initiator controller name in the action statement Remember that Tru64 UNIX IBM AIX and Sun Solaris are case sensitive and it is generally helpful to use lower case to be consistent Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide D 1 Troubleshooting e Inspect script generated log
123. em is running shut it down and power off the hosts f Sun Solaris If the operating system is running and is being used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is being used for other applications remove all I O and unmount all volumes that have remote copy sets that will be failed over gt 2 Manually reconfigure the controllers but do not re create the original remote copy sets This procedure includes the following steps NOTE Steps c f and g will cause the controller pair to restart gt a Set node ID and checksum for THIS controller this information can be found on top of the controller See the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Configuration Guide for information on the node ID and World Wide Name This node ID for the OTHER controller will be set automatically by the command in step c gt b HP Tru64 UNIX IBM AIX Microsoft Windows and Sun Solaris Go to Step 2c HP OpenVMS only Set the identifier to its previous value with the following command SET THIS IDENTIFIER Value Example set this identifier 98 Verify the identifier setting with the following command SHOW THIS_CONTROLLER You will see a display similar to that in Example Display 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 12 7 Unplanned Site Failover with Failback to New Hardware Procedure Example
124. emove Secure Path units and LUNs you must also add or remove the entries for them in sd conf Entries to mda conf and LdLite conf Entries to the CCL in the mda conf IdLite conf and sd conf files are similar to other LUNs with the following differences mda conf The LUN ID assigned to the CCL can be determined by issuing the following CLI command show this_controller The LUN ID assigned to the CCL is displayed in the controller data and is usually LUN 0 Similar to any storage LUN on the RAID storage system there must be two entries for each CCL in mda conf one for each path An example of an entry in mda conf containing an initiator CCL is name mda parent pci 1f 4000 fibre channel 2 target 33 lun 0 qdepth 32 name mda parent pci 1f 4000 fibre channel 4 target 32 lun 0 qdepth 32 An example of an entry in mda conf containing a target CCL is name mda parent pci 1f 4000 fibre channel 2 target 35 lun 0 qdepth 32 name mda parent pci 1f 4000 fibre channel 4 target 34 lun 0 qdepth 32 3 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide CCL Setup ldLite conf The World Wide LUN ID that is needed in dLite conf is created by appending the Node ID ofthe RAID storage system with the hexadecimal equivalent ofthe SCSI ID of the CCL device HSG80CCL You can determine the Node ID by issuing the CLI command show this_controller The Node ID is a 16 character hexadecimal
125. enarios explains the conditions for their use and gives the name of the scripting procedure to follow Table 1 3 Failover Failback and Resumption of Operations Procedure Choices Event or Condition Unplanned loss of initiator site function Initiator site hardware will not be replaced Unplanned loss of target site function Anticipated Duration of Event Unknown Unknown Initiator Mode of Operation Normal or failsafe Failsafe Scripting Procedure to Follow Chapter 6 Unplanned Site Failover with Full Failback Chapter 7 Resumption of Operations After Unplanned Loss of Target Site Failsafe Mode 1 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Table 1 3 Failover Failback and Resumption of Operations Procedure Choices Continued Anticipated Initiator Duration of Mode of Scripting Procedure to Event or Condition Event Operation Follow Unplanned loss of target Unknown Normal Chapter 8 site function Resumption of Operations After Unplanned Loss of Target Site Normal Mode Planned maintenance Short up to Normal or Chapier 9 outage at initiator site several hours failsafe Short Planned Site Failover with Fast Failback Planned maintenance Extended Normal or Chapter 10 outage at initiator site several hours failsafe Extended Planned Site or longer Failover with Full Failback Planned maintenance at Extended
126. er is not only able to distribute an input output request to both the initiator and target sites but it can also transfer the initiator s role to the target site as needed Thus Data Replication Manager is a distributed computing model that supports full disaster tolerant storage device identifier A unique 16 digit hexadecimal identifier of a Fibre Channel device disaster tolerance DT As applied to DRM disaster tolerance provides the ability for rapid recovery of user data from a remote location when a significant event or a disaster occurs at the primary computing site disk mirroring The recording of redundant data for fault tolerant operation Data is written on two partitions of the same disk or on two separate disks within the same system Disk mirroring uses the same controller RAID 1 provides for mirroring which is usually accomplished with SCSI drives See also RAID disk striping The spreading of data over multiple disk drives to improve performance Data is interleaved by bytes or by sectors across the drives For example with four drives and a controller designed to overlap reads and writes four sectors could be read in the same time it normally takes to read one Disk striping does not inherently provide fault tolerance or error checking It is used in conjunction with various other methods See also RAID DRM controller name The identification of a controller in a DRM environment drmdispatch pl See script DRM
127. erforming a DRM operation to modify error mode and asks whether normal or configured error mode is desired Enter n for normal mode gt 6 You will see a confirmation message that asks you to confirm that a normal error mode is desired Enter y for yes gt 7 Ifthis is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app_op rc file from the config subdirectory of the CLONE_HOME directory and run the program file 8 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Operations After Unplanned Loss of Target Site Procedure Normal Mode gt s Whenan operation completion status result is displayed similar to Figure 6 1 verify that the script removed the targets by entering the following command SHOW REMOTE COPY SETS FULL You will see a display similar to that in Example Display 2 Example Display 2 BuildngATop show remote copy sets full Name Uses Used by RCS1 remote copy D1 Reported LUN ID 6000 1FE1 0000 01F0 0009 8490 6303 0134 Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE NORMAL FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to the other controller No targets gt 9 Once the connection to the target site is re established the remote copy s
128. et hosts can connect to the LUNs use this command SHOW UNITS FULL In the Access field of the display all units that are used by remote copy sets should show that the target hosts connections are enabled This should also show the initiator controller connections The following step requires actions relative to each operating system in your configuration a HP OpenVMS Allow hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives 2 If the target site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 If the target hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters is located with the following command hwmgr scan scsi bus x where x is the SCSI bus number 12 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure c IBM AIX Allow the hosts to recognize new units 1 If the target site h
129. ets need to be added back in Run the hsg_op xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system gt 10 You will see a message that you are performing a DRM operation to modify error mode and asks whether normal or configured error mode is desired Enter c for configured error mode gt 11 You will see a confirmation message that asks you to confirm that a configured error mode is desired Enter y for yes gt 12 When an operation completion status result is displayed similar to Figure 6 1 verify that the error mode is normal that the initiator state is online and the target state is normal by entering the following command SHOW REMOTE_COPY_SETS FULL You will see a display similar to that in Example Display 3 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 8 3 Resumption of Operations After Unplanned Loss of Target Site Procedure Normal Mode Example Display 3 BuildngATop gt show remote_copy_sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE NORMAL FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to this controller Not reserved Target state BUILDNGB D1 is NORMAL IMPORTANT If the operation completion status indicates a failure ch
130. f CLI commands to be issued and sends the appropriate sequence of CLI commands for the configuration specified in the configuration file to the Command Scripter The Command Scripter runs the script file and issues the commands to the HSG80 controller over the Fibre Channel bus copying member In a mirrorset a copying member is a container introduced to the mirrorset after the mirrorset has already been in use None of the blocks can be guaranteed to be the same as other members of the mirrorset Therefore the copying member is made the same by copying all the data from a normal member This is in contrast to normalizing where all blocks written since creation are known to be the same When all blocks on the copying member are the same as those on the normal member the copying member becomes a normal member Until it becomes a normal member the copying member contains undefined data and is not useful See also normalizing member Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary 3 Glossary Data Replication Manager Data Replication Manager provides controller based mirroring across a Fibre Channel link The HSG80 Array Controller storage system is used on a host port to host port basis which allows data to be synchronously migrated from one storage system or physical site to another even if they are located at different physical sites Using the required multiple bus failover configuration the controll
131. f all LUNs is normal Use the command SHOW REMOTE_COPY_SETS FULL 15 Continue with the fast failback procedure at the initiator site with Initiator Site Cleanup Procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 9 7 Short Planned Site Failover with Fast Failback Procedure Initiator Site Cleanup Procedure The following steps requires actions relative to each operating system in your configuration gt 1 HP OpenVMS Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives If the target site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG gt 2 HP Tru64 UNIX Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives If the initiator hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number gt 3 IBM AIX Allow the hosts to recognize new units a If the initiator site hosts are
132. failover procedure locate your record of SHOW command output that details the current initiator configuration Verify that your target controller configuration is the same as your initiator controller configuration Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 1 Extended Planned Site Failover with Full Failback Procedure gt 2 Follow the step listed below for each operating system in your configuration Refer to Appendix E for DRM system power up or power down procedures a HP OpenVMS If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O to the LUNs with remote copy sets that will be failed over then dismount the volumes associated with these LUNs HP Tru64 UNIX If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O and unmount all file system LUNs that have remote copy sets that will be failed over IBM AIX If the operating system is running remove all I O to the remote copy set LUNs that will be failed over then unmount the file systems associated with these LUNs Microsoft Windows NT X86 If the operating system is running shut it down and power off the hosts Microsoft Windows 2000 If the o
133. file topology invalid port number problem File is possibly corrupt Restore control file 2141 drm_change_hostport_ Failover or failback control file topology invalid topology problem File is possibly corrupt Restore control file 2160 drm_change_unit_ Controller problem Check the unit characteristic cannot change status and repair as needed characteristic for unit 2161 drm_change_unit_ Possible controller problem Check characteristic incorrect value unit status and repair as needed for characteristic Action Possible configuration problem continues but using defaults Repair file as needed Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide D 7 Troubleshooting Table D 1 Scripting Error Codes Continued Error Code 2180 drm_upload_saved_config Communications error Identify communication problem and correct 2200 drm_deny_remote_hsg Communications error Identify access cannot remove access communication problem and correct 2201 drm_deny_remote_hsg Configuration file problem Update access cannot read configuration file configuration entry for remote copy set 2202 drm_deny_remote_hsg Configuration file problem Node access remote copy set does specified is not an initiator or target not contain this controller Update configuration file 2210 drm_grant_remote_hsg Communication error Identify access cannot grant access communication problem and corre
134. files located in the LOG subdirectory of the CLONE_HOME directory A LOG file is created for each step in the failover failback process and traces the commands and responses that were sent and received from the controllers To see error messages look at the CHK files that are written to this subdirectory Refer to Table D 1 for a list of all error codes their meanings and what actions should be taken e Be sure to look at the most recent LOG and CHK files These files are marked with a numeric revision number with the controller name passed to drmdispatch pl from the application action list For example running the DISASTER_FAILOVER_STEP1 action for a controller named tulsa may result in the following files in the LOG subdirectory tulsa_disaster_failover_step1 chk 1 tulsa_disaster_failover_step1 log 1 Scripting Error Codes This appendix describes error codes that you may encounter when you use the Perl scripts Table D 1 lists the error codes meanings and what actions you should take Table D 1 Scripting Error Codes 1120 new cannot read controller Check presence and file access to configuration file specified configuration file 1180 delete_unit cannot read unit Check unit and underlying storage for or unit does not exist possible problems Correct any error conditions detected on the controller Verify connectivity with the controller and repair error conditions as necessary 1181 delete_unit cli comman
135. following example TULSA TOP gt show d34 LUN Uses Used by D34 DISK50100 partition LUN ID 6000 1FE1 0009 0A30 0001 0280 4798 00B3 NOIDENTIFIER Switches RUN NOWRITE_PROTECT READ_CACHE READAHEAD_CACHE WRITEBACK_CACHE MAX_READ_CACHED_TRANSFER_SIZE 128 MAX_WRITE_CACHED_TRANSFER_SIZE 128 Access TgtHostl HBA TgtHost2 HBA InitHostl HBA InitHost2_HBA State ONLINE to this controller Not reserved NOPREFERRED PATH Host Based Logging NOT Specified Size 4442290 blocks Geometry C H S 1315 20 169 Next run the following commands to get a listing of all recognized devices cd opt CPQswsp bin spmgr display Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 15 File Customization A listing of all devices similar to the following displays Server SUN002 Report Created Thu Dec 13 16 14 03 2001 Command spmgr display Storage 5000 1FE1 0009 0D90 Load Balance Off Auto restore Off Path Verify On Verify Interval 30 HBAs fca pci 0 fca pci 1 fca pci 4 fca pci 5 Controller 2G05103924 Operational 2G02103935 Operational Devices c6t0dO0 c6t0d1 c6t0d2 c6t0d6 c6t0d3 c t0d4 c6t0d5 TGT LUN Device WWLUN ID Parent 4 Paths 0 0 c7t0d0 6000 1FE1 0009 0A30 0001 0280 4798 00B3 4 Locate your device by matching the WWLUN_ID of this display with the LUN ID obtained from the previous SHOW command The device indentifier is listed for the non RCS LUN for example c7t0dO and
136. for running an unplanned site failover with failback to new hardware 12 20 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 13 Planned Site Role Reversal Procedure This chapter provides the procedure for performing a site role reversal This procedure is used to implement a planned failover of an initiator site with subsequent simple failback to an alternate target site The sequence of steps are e Initiator Site Preparation Procedure e Running the Role Reversal Failover Program File Procedure e Target Host Setup Procedure e Running the Role Reversal Failback Program File Procedure Initiator Site Cleanup Procedure NOTE In this chapter the P symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Initiator Site Preparation Procedure gt 1 Before performing the failover procedure locate your record of SHOW command output that details the current initiator configuration Verify that your target controller configuration is the same as your initiator controller configuration gt 2 Follow the step listed below for each operating system in your configuration Refer to Appendix E for DRM system power up or power down procedures a HP OpenVMS If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for
137. fy error mode and asks whether normal or configured error mode is desired Enter c for configured error mode gt 7 You will see a confirmation message that asks you to confirm that a configured error mode is desired Enter y for yes gt s If this is the first time the program file is run or you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app op rc file from the config subdirectory of the CLONE HOME directory and run the program file gt 9 Whenan operation completion status result is displayed similar to Figure 6 1 verify that the error mode of the remote copy set is failsafe by entering the following command SHOW REMOTE_COPY_SETS FULL You will see a display similar to that in Example Display 4 Example Display 4 BuildngATop gt show remote_copy_sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state ONLINE to this controller Not reserved Target state BUILDNGB D1 is NORMAL IMPORTANT If the operation completion status indicates a failure check the remote copy sets to see if they are in a failsafe mode and their target state is normal Sometimes the screen refresh m
138. g User Guide Sample Controller Configuration File TERMINALSERVER backup password primary UNIT D1 DISK30200 D11 DISK60300 D2 DISK50100 D3 DISK60100 D50 HISTLOG D6 DISK40300 D1 D11 D2 D3 D50 D6 UNIT IDENTIFIERS Example of Controller Configuration File Continued Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide B 3 Sample Controller Configuration File B 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide C Structure of the Application Action List Default Application Action List The default application action list app ex act is provided in the DRM Scripting Kit and installed in the CONFIG subdirectory of the CLONE HOME directory the default directory where the scripts reside It contains groupings of various failover failback and resumption of operation actions and lists the DRM initiator or target controllers to be included in each action The scripts need the target controller name in a failover or failback action and the initiator controller name in a resumption of operation action Only one controller name is needed for an initiator target controller pair because the script can determine and perform its role based on the pairing information contained in the controller configuration file The default application action list is shown below It contains actions for two separate DRM controller pairs The two target control
139. ge 5 2 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Run the hsg_fb xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter r for a role reversal failback You will see a confirmation message that asks you to confirm that a role reversal failback is desired Enter y for yes If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app fb1 rc file from the config subdirectory of the CLONE HOME directory and run the program file IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide 9 Continue the role reversal failback procedure at the initiator site with Initiator Site Cleanup Procedure Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 13 7 Planned Site Role Reversal Procedure Initiator Site Cleanup Procedure The following steps requires actions relative to each operating system in y
140. guration file ASSOCIATIONSET CX07539A NOTE Ensure an exact copy ofthe data is made with no extra lines returns and so on Figure 4 3 Copying association set information Remote Copy Set Section Because the information is not available when the script runs this is another section that does not exist in the target configuration file Using a text editor copy the remote copy set information from the corresponding initiator controller configuration file into the target file as shown in Figure 4 4 4 20 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization Initiator configuration file REMOTECOPYSET RD6 INITIATOR TULSA D6 TARGET FARGO D6 OPERATION SYNCHRONOUS ERROR NORMAL O 20 RSQLDATA1 INITIATOR TULSA D1 TARGET FARGO D1 OPERATION SYNCHRONOUS ERROR NORMAL IO 20 RSQLDATA2 INITIATOR TULSA D2 TARGET FARGO D2 OPERATION SYNCHRONOUS ERROR NORMAL IO 20 RSQLLOG INITIATOR TULSA D3 TARGET FARGO D3 OPERATION SYNCHRONOUS ERRORSFAILSAFE IO 20 Target configuration file REMOTECOPYSET CX07540A NOTE Ensure an exact copy ofthe data is made with no extra lines returns and so on Figure 4 4 Copying remote copy set information Connections Section This section in the configuration file specifies which host has access to the LUNs In a newly created target configuration file the information may look like the following CONNECTIONS D1 TULSAC TULSAD D2 TULSAC TULSAD On
141. han bytes or groups of bytes are interleaved across multiple disks RAID 3 Data is striped across three or more drives Used to achieve the highest data transfer rate because all drives operate in parallel Parity bits are stored on separate dedicated drives RAID 4 Similar to RAID 3 but manages disks independently rather than in unison Not often used RAID 5 Most widely used Data is striped across three or more drives for performance parity bits are used for fault tolerance The parity bits from two drives are stored on a third drive RAID 6 Highest reliability but not widely used Similar to RAID 5 but does two different parity computations or the same computation on overlapping subsets of the data RAID 10 Also designated RAID 1 0 It is a combination of RAID 1 and 0 mirroring and striping RCS See remote copy set redundancy The provision of multiple interchangeable components to perform a single function to cope with failures and errors A RAIDset is considered to be redundant when user data is recorded directly to one member and all of the other members and associated parity also are recorded If a member is missing from the RAIDset its data can be regenerated as needed but the RAIDset is no longer redundant until the missing member is replaced and reconstructed Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary 9 Glossary remote copy set RCS A feature that allows data to be
142. he default directory with the following command SET DEFAULT SYS SYSDEVICE CMDSCRPT Copy the self extracting zip file using the following command COPY DKA400 OVMS_V71 SCRPT10V EXE Unzip the files with the following command RUN SCRPT10V EXE A message displays showing that the files are unzipped Enter the following command DIRECTORY The following message displays Directory SYS SYSDEVICE CMDSCRIPT COMPAQ AXPVMS CPQCMDSCR V0100 45 1 PCSI 1 SCRIPT10V EXE 1 Install Command Scripter using the following command PRODUCT INSTALL CPQCMDSCR SOURCE When asked to continue enter Yes Follow the on screen prompts A verification message is displayed When installation is complete add the following line to the system SYLOGIN COM file CMDSCRIPT CMDSCRIPT 2 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Installation To run Command Scripter you must have certain process privileges Two in particular are the privileges DIAGNOSE and LOG_IO The Perl scripts must be run from a privileged account Installation is complete HP Tru64 UNIX The following procedure installs the Command Scripter on a Tru64 UNIX server l Mount the CD ROM using the following command mount r t cdfs o rrip dev disk cdrom0c mnt Enter the following commands CD mnt CD unix install sh The license agreement displays E
143. he failover procedure locate your record of SHOW command output that details the current initiator configuration Verify that your target controller configuration is the same as your initiator controller configuration Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 9 1 Short Planned Site Failover with Fast Failback Procedure gt 2 Follow the step listed below for each operating system in your heterogeneous configuration Refer to Appendix E for DRM system power up or power down procedures a HP OpenVMS If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O to the LUNs with remote copy sets that will be failed over then dismount the volumes associated with these LUNs HP Tru64 UNIX If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O and unmount all file system LUNs that have remote copy sets that will be failed over IBM AIX If the operating system is running remove all I O to the remote copy set LUNs that will be failed over then unmount the file systems associated with these LUNs Microsoft Windows NT X86 If the operating system is running shut it down and power off the hosts Microsoft Windows
144. hp StorageWorks data replication manager hsg80 acs version 8 7p scripting user guide Part Number EK DRMSC OA E01 Fifth Edition September 2002 Product Version ACS Version 8 7P DRM Scripting Kit Version 3 0 This user guide provides installation configuration and operating procedures for running scripts with HP StorageWorks Data Replication Manager The scripts run on HP OpenVMS HP Tru64 IBM AIX Microsoft Windows NT 2000 and Sun Solaris hosts and handle eight specific failover failback and resumption of operation situations invent Hewlett Packard Company 2002 All rights reserved Hewlett Packard Company makes no warranty of any kind with regard to this material including but not limited to the implied warranties of merchantability and fitness for a particular purpose Hewlett Packard shall not be liable for errors contained herein or for incidental or consequential damages in connection with the furnishing performance or use of this material This document contains proprietary information which is protected by copyright No part of this document may be photocopied reproduced or translated into another language without the prior written consent of Hewlett Packard The information contained in this document is subject to change without notice HP Compaq StorageWorks SANworks OpenVMS and Tru64 are trademarks of Hewlett Packard Company in the U S and or other countries Microsoft Windows and W
145. hronous mode target The site that is set up for data replication Data processing occurs at the initiator site and data is replicated or copied to the target site If a significant failure occurs at the initiator site data processing can be resumed at the target site where the data is intact See also initiator target configuration file A configuration file at the target site See also configuration file target Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary 11 Glossary TCP IP Transmission Control Protocol Internet Protocol A communications protocol that is the protocol of the Internet and has become the global standard for communications TCP provides transport functions which ensure that all bytes sent are received correctly at the other end TCP IP is a routable protocol and the IP part of TCP IP provides the routing capability In a routable protocol all messages contain not only the address of the destination station but the address of a destination network This allows TCP IP messages to be sent to multiple networks within an organization or around the world hence its use in the worldwide Internet Every client and server in a TCP IP network requires an IP address See also IP address this controller The controller that is serving the current CLI session through a local or remote terminal See also other controller unit A container made accessible to a host A unit may be create
146. iguration files are built by running configuration generation program files that execute the generate_cfg pl script However these files require additional information to allow the target site to assume the initiator role after failover You can copy some of this information directly from similar sections of the corresponding initiator controller configuration file You will need to edit other sections manually Association Set Section This section exists in the initiator configuration file but not in the target configuration file because association sets do not exist on the target site when the script is executed To make the necessary changes Using a text editor copy the association set information from the initiator configuration file into the target configuration file as shown in Figure 4 3 e Set up a target log unit for each association set Refer to the procedure described in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Configuration Guide if you are not sure how to do this Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 19 File Customization NOTE HP recommends that the log unit on the initiator match a designated log unit on the target and that it be set up as a write history log unit Initiator configuration file ASSOCIATIONSET ALOG MEMBERS RSQLLOG FAIL ALLINOORDER ALL NOLOG UNIT ASQL MEMBERS RSQLDATA1 RSQLDATA2 NOFAIL ALL ORDER ALL LOG UNIT D90 Target confi
147. ilback and resumption of operations The following sections discuss these two Perl scripts hsgcontrol pl The hsgcontrol pl script runs from a batch file The script processes actions that are read from the application action list The following command line shows the script syntax for the Windows platform where CLONE HOME is the environmental variable that represents the directory where the scripts reside Its structure is shown in Table C 2 CLONE_HOME S BIN hsgcontrol pl FileName ActionLabel RCFile Table C 2 Structure of the hsgcontrol pl Script Command CLONE_HOME BIN The variable pointing to the default directory of the script files and the BIN subdirectory This is the path to the hsgcontrol pl script hsgcontol pl The Perl script name FileName The name of the application action list that contains the actions to be performed ActionLabel The named action to be performed from the application action list For example PLANNED FAILOVER STEP 1 RCFile This file sets the user s preference to receive status reporting in either verbose or condensed mode For more information about these preferences see Chapter 5 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Structure of the Application Action List An example of the hsgcontrol pl Perl script command executed by a Windows batch file is Perl I CLONE HOME CLONE_HOME bin hsgcontrol pl app PLANNED SHORT
148. ile scripts_30 tar into the directory created in step 1 From the command line enter the following command tar xvfp scripts_30 tar The kit files will self extract into the default directory you created Verify that the subdirectories bin config log sh and tmp were created in the default directory See Appendix A for a list and description of the installed files Set the environmental variable In the ksh shell use the following two commands CLONE HOME ScriptDefault Example CLONE HOME scripts export CLONE HOME These lines can also be added to the profile file to make the environmental variable load after rebooting 2 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Installation Microsoft Windows NT 2000 1 Create a directory for example C scripts to be a default directory for the scripts 2 Copy the scripting kit self extracting file script_win_30 exe into the directory created in step 1 3 From Windows Explorer or a command line prompt double click or execute the script_win_30 exe file The kit files self extract into the default directory you created 4 Verify that the subdirectories bat bin config log and tmp are created in the default directory See Appendix A for a list and description of the installed files 5 Add an environmental variable named CLONE_HOME to set the default directory of the scripts a b C
149. ile for the initiator site Use a text editor and make the necessary modifications using the following syntax perl I CLONE HOME SCLONE_HOME bin generate_cfg pl com cs RemoteCopyName idnum where Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 7 File Customization RemoteCopyName identifies the DRM controller remote copy name of the initiator subsystem for example tulsa This name is case sensitive in IBM AIX and must match the case of the name used later when configuring the application action list idnum specifies the device number for the operating system that Command Scripter uses to communicate with controller tulsa As shown in Figure 4 1 this is the device number of the CCL for the local controller Controller A communicating with the initiator host Host A To find the idnum use the 1sdev Cc disk command to find the device assigned to the CCL A listing of all available hdisks appears with those that are CCLs identified as such So for this example the idnum for an hdisk CCL might be dev hdisk1 To verify that this is the correct controller subsystem to which this device communicates enter the following command cmdscript f dev hdiskl show this The resulting display shows the characteristics of the controller beginning with the controller remote copy name If atypical show this controller response is not received then an incorrect idnum was used for that
150. ilover Failback Procedures Guide for the full status comparison procedure Make sure that any status change is reflected on the target To make a status comparison bring up a terminal emulator session and enter a SHOW THIS command Continue the new hardware failback phase at the target site with Running the New Hardware Failback Program Files Procedure Running the New Hardware Failback Program Files Procedure 1 2 Open a command prompt window on the target host Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 Change to the bat sh or vms subdirectory in the CLONE_HOME directory Run the hsg_fb xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter n for a new hardware failback You will see a confirmation message that asks you to confirm that a new hardware failback is desired Enter y for yes 12 14 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display E
151. indows NT are trademarks of Microsoft Corporation in the U S and or other countries UNIX is a trademark of The Open Group in the U S and or other countries All other product names mentioned herein may be trademarks of their respective companies Hewlett Packard Company shall not be liable for technical or editorial errors or omissions contained herein The information is provided as is without warranty of any kind and is subject to change without notice The warranties for Hewlett Packard Company products are set forth in the express limited warranty statements accompanying such products Nothing herein should be construed as constituting an additional warranty Printed in the U S A Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Fifth Edition September 2002 Part Number EK DRMSC OA E01 Contents About this Guide Intended Audienee 23 528 ae qe Em lex e Rs ERA EN ix Related Documentation 415 5 22 325 sei erat cete Ae A do een e i e A ix Prerequisites 2 3 2 2 2 2 Settee a wets We ae EE Les Sa eee x Document Conventions 0 esses RR RR RR RR es xi Symbols in Lext28 4th 8 sie ss ses ode o Verre e opes ges aea ee M Ade xi Symbols on Equipment denta netanna eaa e iapa e rh xii Rack Stability toe ala Eoo Ee LNG an ae Pa ru eae tds xiii Get ng Helps co Sra see Xano air Bu xiii HP Techlinical S ppott eei ree eger uer que e Ng e xiii HP Website a eet ores preter red te ge xiv HP Autho
152. ing Error Codes Continued drm_delete_remotecopyset remote copy set still exists after removal drm_add_target_unit the remote copy set does not exist drm_add_target_unit the remote copy set already has a target Delete that first drm_add_target_unit cannot add target to existing remote copy set Check connection to remote site drm_remove_target_unit the remote copy set does not exist drm_remove_target_unit mismatch between configuration file and current controller setup drm_remove_target_unit target unit still there after removal drm_grant_server_access connection configured for unit does not exist in controller Cannot delete remote copy set Check for communication error and repair as needed Also possible issue with error mode on controller If in failsafe set the error mode to normal Remote copy set does not exist on the controller Update the configuration file Remote copy set is present but already has a target configured There is a mismatch between the configuration file and controller configuration Update the configuration file Check target unit status for access problems and verify the path to remote site is functional The remote copy set does not exist on the controller Update the configuration file The controller and its configuration file specify a target unit but conflict with each other Update the configuration file Communication error I
153. ion Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization SET ERROR NORMAL OPERATION Background sun drmdispatch PLANNED failsafe op ALL NOTFORCED Background mars drmdispatch PLANNED failsafe op ALL NOTFORCED Background jupiter drmdispatch PLANNED failsafe op ALL NOTFORCED END SET ERROR NORMAL OPERATION NOTE The initiator controller names in the actual app ex act file that would be replaced are tulsa and greenbay 6 Savethe file The same app act file you just created is also used at the target site Copy this file from your initiator host to your target host The file customization is now complete for both sites Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 25 File Customization 4 26 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 5 Scripting File Descriptions and Behaviors This chapter describes the function of the program files used for failover failback and resumption of operations The chapter also explains how to verify that the scripts can communicate with the controllers via Command Scripter and how to work with some of the scripting features Program File Descriptions The program files that start the failover failback and resumption of operation scripts are provided in the DRM Scripting Kit and are placed in the bat sh or vms subdirectory of the CLONE_HOME directory
154. is used to create the necessary DeviceID syntax for the configuration generation file as follows dev rdsk c7t0d0s2 where s2 is the slice number of the drive To verify that this is the correct controller subsystem to which this device communicates enter the following command cmdscript f dev rdsk c7t0d0s2 show this The resulting display shows the characteristics of the controller beginning with the controller remote copy name If atypical show this controller response is not received then an incorrect device ID was used for the controller Save the edited generation file into the SCLONE HOME sh subdirectory with a meaningful name like tulsa_gen sh Modify the gen ex sh file again to create a target configuration generation file for the initiator site Use the syntax perl I CLONE HOME S CLONE HOME bin generate cfg pl com cs RemoteCopyName DeviceID where RemoteCopyName identifies the target DRM controller remote copy name of the subsystem for example fargo This name is case sensitive in Solaris and must match the case of the name used later when configuring the application action list 4 16 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization DevicelD specifies the identifier for the non RCS LUN you created to allow Command Scripter to communicate with the initiator controller tulsa The setup is similar to that shown in Figure 4 2 where instea
155. ite history log to the initiator site If space on the write history log is exceeded a full copy of the disk is performed The display indicates when copying is complete At this time you are disaster tolerant and can operate in this mode until you choose to complete the failback process IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide 10 When you are ready to complete the failback to the original initiator site run the hsg_fb2 xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system 11 You will seea message asking what type of failback to run Enter s for a fast failback 12 You will see a confirmation message that asks you to confirm that a fast failback is desired Enter y for yes 13 If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app_fb2 rc file from the config subdirectory of the CLONE_HOME directory and run the program file 14 After normalization occurs check that the target state o
156. k Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 11 Extended Planned Site Failover with Full Failback Procedure 10 12 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 11 Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode This chapter provides the procedure for performing a resumption of replication after an extended planned loss of the target site while running in failsafe error mode This procedure is used when the error mode of the remote copy set is set for failsafe and when the target site will be shut down for an extended length of time Setting the error mode of the remote copy set to normal will allow host I O to continue while the target site is offline Because there is no log unit configured a full copy will be performed once the target is back online The sequence of steps are Running the Resumption of Replication Program File Procedure e Continuing the Resumption of Replication Program File Procedure NOTE In this chapter the gt symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Running the Resumption of Replication Program File Procedure gt 1 Use the following CLI command to verify that the error mode of the remote copy sets is set to failsafe and to make note of the initiator states SHOW REMOTE_COPY_SETS FULL
157. k Procedure 8 If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app_fb1 rc file from the config subdirectory of the CLONE HOME directory and run the program file When an operation completion status result is displayed similar to Figure 6 1 shut down all target site hosts IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide 10 When you are ready to complete the failback to the original initiator site run the 11 12 13 14 hsg fb2 xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter f for a full failback You will see a confirmation message that asks you to confirm that a full failback is desired Enter y for yes If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To chang
158. laris 6 and 7 and Windows NT 2000 Compaq SANworks Command Scripter Version 1 0A These components are limited to the requirements listed on page 1 14 A brief description of each scripting component follows HP DRM Scripting Kit Specific DRM Scripting Kits are available for the following operating systems e HP OpenVMS HP Tru64 UNIX IBM AIX and Sun Solaris e Microsoft Windows NT and Microsoft Windows 2000 These virtual kits consist only of files and are downloaded from the HP website They contain Perl scripts Perl support files example files and program files necessary for the scripts to perform failover failback and resumption of operation procedures Program files are command files OpenVMS shell files AIX Tru64 and Solaris and batch files Windows NT 2000 specific to an operating system Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Perl Interpreter Perl is the interpreted programming language in which the scripts are written The Perl interpreter translates and processes the scripts Every Perl script must pass through the interpreter in order to execute SANworks Command Scripter Compaq SANworks Command Scripter is application software that provides an interface to communicate the CLI commands generated by the Perl scripts to the HSG80 controllers via the Fibre Channel bus How the Failover Failback and Resumption of Operation Scripts Work
159. ler Configuration File ASSOCIATIONSET ASS_1 MEMBERS RCS_M1 RCS_M2 NOFAIL_ALL NOORDER_ALL LOG_UNIT D50 Q LONES Q LONESTORAGESETCONFIG CONNECTIONS D1 FARGOC FARGOD HORN_B HORN_T D11 SUN018_B SUN018_T D2 FARGOC FARGOD HORN_B HORN_T D3 FARGOC FARGOD HORN_B HORN_T D50 None D6 TRU050_B TRU050_T CONTROLLER CCLLUN 0 CCLid 2 Failover MULTIBUS_FAILOVER Firmware V86P SANName TULSA SCSI SCSI 3 SerialNumbers 2G91606296 2G91205687 WWid 5000 1FE1 0000 1520 commmode cs device zg91606296 name tulsa Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide B 1 Sample Controller Configuration File Example of Controller Configuration File Continued MAX_READ_CACHED_TRANSFER_SIZE D1 32 D11 32 D2 32 D3 32 D50 32 D6 256 MAX_WRITE_CACHED_TRANSFER_SIZE D1 32 D11 32 D2 32 D3 32 D50 32 D6 256 MIRRORSET STLOG DISK10000 DISK20000 Ml m PREFERRED PATH D1 THIS CONTROLLER Dl11 THIS CONTROLLER D2 THIS CONTROLLER D3 THIS CONTROLLER D50 D6 OTHER CONTROLLER RAID5SET REMOTECOPYSET RCS_M1 INITIATOR TULSA D1 TARGET FARGO D1 OPERATION SYNCHRONOUS ERROR NORMAL 10 200 RCS_M2 INITIATOR TULSA D2 TARGET FARGO D2 OPERATION SYNCHRONOUS ERROR NORMAL IO 200 RCS_M3 INITIATOR TULSA D3 TARGET FARGO D3 OPERAT ION S YNCHRONOUS ERROR NORMAL IO 200 SNAPSHOTS STRIPESET B 2 Data Replication Manager HSG80 ACS Version 8 7P Scriptin
160. lers are named fargo and denver The two initiator controllers are tulsa and greenbay Default Application Action List This file specifies the actions that need to be performed on the HSG controllers This configuration file is used by the CLONE HOME bin hsgcontrol pl The structure of the file is as follows comment sign Only comment signs on the lst position of the line are allowed ACTION Start of the Action specification for this application END ACTION End of Action definition for this application Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide C 1 Structure of the Application Action List Default Application Action List Continued An action line is constructed as follows Foreground Background lt ControllerName gt lt PerlScriptName gt Parameterl lt Parameter2 gt lt Parameter3 gt lt Parameter4 gt SHORT_PLANNED_FAILOVER_STEP1 Background fargo drmdispatch PLANNED failover_stepl ALL NOTFORCED Background denver drmdispatch PLANNED failover_stepl ALL NOTFORCED END_SHORT_PLANNED_FAILOVER_STEP1 SHORT_PLANNE
161. ling Technical support registration number if applicable Product serial numbers Product model names and numbers Applicable error messages Operating system type and revision level Detailed specific questions IMPORTANT To be eligible for technical support you must be using the DRM scripts provided by HP as described in the User Guide in a supported configuration Scripts that have been modified in any way are not supported Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide xiii About this Guide HP Website The HP website has the latest information on this product as well as the latest drivers Access the HP website at http thenew hp com country us eng storage html From this website select the appropriate product or solution HP Authorized Reseller For the name of your nearest HP Authorized Reseller e Inthe United States call 1 800 345 1518 e In Canada call 1 800 263 5868 e Elsewhere see the HP website for locations and telephone numbers xiv Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 DRM Scripting Overview Introduction HP StorageWorks Data Replication Manager provides a means to prevent data loss through the use of hardware redundancy and software data replication A Data Replication Manger DRM configuration consists of paired storage sites The initiator site carries out primary data processing A target site is set up for data replicati
162. lock for block contains exactly the same data as that on the other members within the mirrorset Read requests from the host are always satisfied by normal members other controller The controller in a dual redundant pair that is not connected to the controller serving the current CLI session with a local terminal See also this controller local terminal peripheral device Any unit distinct from the CPU and physical memory that can provide the system with input or can accept output from it Terminals printers tape drives and disks are peripheral devices Perl Practical Extraction Report Language A programming language that combines syntax from several UNIX utilities and languages Perl is widely used to write Web server programs for such tasks as automatically updating user accounts and news group postings processing removal requests synchronizing databases and generating reports Perl interpreter A program through which Perl programs are passed at run time for execution The Perl interpreter translates the program internally and then executes it immediately Perl module Code written in the Perl language to perform a specific task planned failover As applied to the Data Replication Manager an orderly shutdown of the controllers for installation of new hardware updating the software and so on The host applications are quiesced and all write operations are permitted to complete before the shutdown The controllers must be in
163. lover with Full Failback Procedure This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters is located with the following command hwmgr scan scsi bus x where x is the SCSI bus number c IBM AIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives 2 If the target hosts are not shut down use the following commands to recognize the drives and verify that they are present cfgmgr v lsdev Cc disk Use the following commands to access file systems on the failed over LUNs importvg y volumegroupname hdiskx mount all NOTE volumegroupname is the name of the volume group originally created at the initiator site and x is the number of the hdisk assigned to the failed over LUN If the y volumegroupname parameter is omitted AIX will create a default volume group name for example vg00 d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the target site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 Allow hosts to recognize new units 1 If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to
164. ly the DRM initiator controller connections will have been inserted in this section by the remote copy set LUNs You must modify this section to enable the target site hosts access to these LUNs following a site failover To do this modify the target configuration file with the names of the desired host connections Spaces between the connection names are not allowed For example assume that SERVA_T and SERVA_B are target site host connections to be given access to the LUNs After you modify the file with a text editor the resulting section would look like the following CONNECTIONS D1 TULSAC TULSAD SERVA_T SERVA_B D2 TULSAC TULSAD SERVA_T SERVA_B Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 21 File Customization Maximum Read Write Cached Transfer Size Section When the target configuration file is created maximum read and write cached transfer size default values are inserted that do not correspond to the values in the initiator configuration file Change these values to match those of the initiator configuration file Ensure that the units being modified are mapped to the correct remote copy sets For example the newly created initiator and target configuration files may show the following Initiator Configuration File Target Configuration File MAX_READ_CACHED_TRANSFER_SIZE MAX
165. m You will see a message asking what type of failback to run Enter f for a full failback You will see a confirmation message that asks you to confirm that a full failback is desired Enter y for yes If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app fb2 rc file from the config subdirectory of the CLONE HOME directory and run the program file 6 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Unplanned Site Failover with Full Failback Procedure 14 Boot all of the target hosts now to be ready for a future failover 15 Continue with the full failback procedure at the initiator site with Initiator Site Cleanup Procedure Initiator Site Cleanup Procedure The following steps require actions relative to each operating system in your configuration gt 1 HP OpenVMS Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives b Ifthe initiator site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG gt 2 HP Tru64 UNIX Allow the hosts to recognize new units a If the initiator
166. mmand Scripter installation 2 6 creating configuration generation files 4 3 device identification number 4 4 Index 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Index DRM Scripting Kit installation 2 2 job queue setup 3 2 obtaining CCL listing 4 4 Perl interpreter 1 16 2 4 privileges 2 7 4 18 6 3 program file 1 1 running configuration generation files 4 18 script termination 5 5 SCSI 3 mode 1 15 3 1 4 3 HP Tru64 UNIX CCL identification 4 6 CCL setup 3 2 Command Scripter installation 2 7 creating configuration files 4 6 device identification number 4 6 4 7 DRM Scripting Kit installation 2 2 Perl interpreter 1 16 program file 1 1 running configuration generation files 4 18 script termination 5 6 SCSI 2 mode 1 15 3 2 SCSI 3 mode 1 15 3 2 hsgcontrol pl 1 10 1 12 command structure C 6 description A 1 syntax C 6 4 6 4 6 IBM AIX CCL setup 3 4 Command Scripter installation 2 7 creating configuration files 4 7 device number 4 8 DRM Scripting Kit description 1 8 installation 2 2 Perl interpreter 1 16 platform requirements 1 14 program file 1 1 running configuration generation files 4 18 script termination 5 6 SCSI 2 mode 1 15 3 4 4 7 SCSI 3 mode 1 15 3 4 4 7 using the CCL 4 7 IdLite conf 3 6 3 7 important defined xi initiator site 1 1 initiator site symbol and definition xi instance codes D 9 K keyboard keys convention defined xi M Management Applianc
167. n in data access after a failure IMPORTANT Verify that all components at the target site are operational before you begin the site failover NOTE If you decide to perform a failover operation keep in mind that all components must be failed over Therefore if only one component fails fixing that single component may be preferable to performing a complete failover Table 1 1 outlines example criteria that calls for a failover as well as criteria that does not call for a failover 1 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview Table 1 1 Possible Failover Situations Type of Failure Total initiator site loss Recommended Action Remote Copy Set Error_Mode Normal Manual intervention to fail over data and processing to target site Remote Copy Set Error_Mode Failsafe Manual intervention to fail over data and processing to target site Loss of initiator site fabric Manual intervention to fail over data and processing to target site Manual intervention to fail over data and processing to target site Loss of initiator controller pair Manual intervention to fail over data to target site and restart of processing at both sites Manual intervention to fail over data to target site and restart of processing at both sites Loss of all intersite links Total target site loss Failover not necessary Failover not necessary Decide on
168. name of the target subsystem for example fargo Device specifies the non RCS LUN that Command Scripter uses to communicate with the target controller from the initiator host for example R in Figure 4 2 Save this batch file in the BAT subdirectory with a meaningful name like fargo_gen bat Repeat the above steps to create initiator and target configuration generation batch files for all hosts running scripts at the target site Use the correct non RCS LUN device letters that provide communications to the controllers Save these batch files in the BAT subdirectory on the target host using the same naming convention used previously SCSI 3 Mode with CCLs In SCSI 3 mode the Windows scripting hosts communicate with controllers through CCLs Chapter 3 describes how to set up CCLs and identify their serial numbers Follow these steps to create initiator and target controller generation batch files 1 Modify the gen_ex bat file to create an initiator controller configuration generation file for each host running scripts at the initiator site Use a text editor and make the necessary modifications using the following syntax Perl I CLONE_HOME CLONE_HOME bin generate_cfg pl com cs RemoteCopyName SerNum where RemoteCopyName identifies the DRM controller remote copy name of the initiator subsystem for example tulsa The remote copy name can be obtained by running a SHOW THIS command from the controller
169. ng points of information P Identifies a procedural step to be performed at the initiator site Identifies a procedural step to be performed at the target site Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide xi About this Guide Symbols on Equipment AA Any enclosed surface or area of the equipment marked with these symbols indicates the presence of electrical shock hazards Enclosed area contains no operator serviceable parts WARNING To reduce the risk of injury from electrical shock hazards do not open this enclosure ANS AA Any RJ 45 receptacle marked with these symbols indicates a network interface connection WARNING To reduce the risk of electrical shock fire or damage to the equipment do not plug telephone or telecommunications connectors into this receptacle Any surface or area of the equipment marked with these symbols indicates the presence of a hot surface or hot component Contact with this surface could result in injury WARNING To reduce the risk of injury from a hot component allow the surface to cool before touching AN Power supplies or systems marked with these symbols indicate the presence of multiple sources of power E Ea WARNING To reduce the risk of injury from electrical shock nn remove all power cords to completely disconnect power from the power supplies and systems ANA Any product or assembly marked with these sym
170. nnel bus With Command Scripter users can edit and run script files that contain CLI commands Glossary 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary condensed display A screen output displayed while the failover and failback scripts run The display shows only the status of the controller A user preference for this type of display is set in a resource RC file created by the failover and failback batch files See also verbose display configuration file A file that tells the failover failback scripts what devices are attached to a controller and how the controller is configured with respect to devices and storagesets There is a configuration file for each controller subsystem on the initiator and target hosts See also initiator configuration file and target configuration file container 1 Any entity that is capable of storing data whether it is a physical device or a group of physical devices 2 A virtual internal controller structure representing either a single disk or a group of disk drives linked as a storageset Stripesets and mirrorsets are examples of storageset containers that the controller uses to create units controller A hardware device that uses software to facilitate communications between a host and one or more storage devices organized in an array The HS series StorageWorks family of controllers are all array controllers control table A file that controls the order o
171. nter Yes to accept the license agreement terms The following message displays Starting the Command Scripter Installation Press Enter to continue with the installation The files are copied Copy cmdscript to CLONE_HOME bin the subdirectory under the default directory where the DRM scripting files reside for example scripts bin Installation is complete IBM AIX The following procedure installs the Command Scripter on an IBM AIX server 1 Mount the CD ROM using the following command mount cdrom Enter the following commands CD cdrom install sh The license agreement displays Enter Yes to accept the license agreement terms Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 2 7 Installation 4 The following message displays Starting the Command Scripter Installation Press Enter to continue with the installation The files are copied 5 Copy cmdscript to CLONE HOMB bin the subdirectory under the default directory where the DRM scripting files reside for example scripts bin Installation is complete Microsoft Windows NT 2000 The following procedure installs the Command Scripter on Windows NT or Windows 2000 servers 1 Insert the Command Scripter CD ROM The InstallShield Wizard runs automatically NOTE If the CD ROM does not automatically run open Windows Explorer and click the CD ROM drive Double click the Windows folder then double cli
172. nter v for verbose or c for condensed To change the type of display you must delete the app fb1 rc file from the config subdirectory of the CLONE HOME directory and run the program file Monitor the display for the indication that mirroring is complete At this time you are disaster tolerant and can operate in this mode until you choose to complete the failback process IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide 9 10 11 12 13 When you are ready to complete the failback to the original initiator site run the hsg fb2 xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system You will see a message asking what type of failback to run Enter n for a new hardware failback You will see a confirmation message that asks you to confirm that a new hardware failback is desired Enter y for yes If this is the first time the program file is run or if you have deleted the RC file you are prompted to select a verbose or condensed reporting display Enter v for verbose or c for condensed To change the type of display you must delete the app fb2 rc file from the config subdirectory of the
173. o step 5 c If the remote copy sets are mixed between failsafe and normal error modes and 1 Every remote copy set in failsafe error mode is failsafe locked and every remote copy set in normal error mode has no targets proceed to step 5 2 Every remote copy set in failsafe error mode is not in a failsafe locked state proceed to step 5 3 Any remote copy sets in failsafe error mode are failsafe locked and remote copy sets in normal error mode have targets proceed to step 4 gt 4 Forremote copy sets in failsafe error mode with mixed initiator states or a mixture of error modes with failsafe error modes that are failsafe locked and normal error mode with targets change all the remote copy sets in failsafe mode to normal mode Use the CLI command SET RemoteCopySetName ERROR_MODE NORMAL Example SET RCS1 ERROR_MODE NORMAL When the remote copy sets are in normal mode proceed to the next step Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 3 Extended Planned Site Failover with Full Failback Procedure gt 5 Continue the planned failover process with the Running the Extended Planned Failover Program File Procedure Running the Extended Planned Failover Program File Procedure o Q9 1 2 Open a command prompt window on the target host Verify that the script server can communicate with the target controller via Command
174. ocedure e Running the Resumption of Operations Program File Procedure e Initiator Site Cleanup Procedures e Running the Resumption of Operations Program File Procedure NOTE In this chapter the gt symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Verification of Lost Connections Procedure gt 1 Verify that the connection to the target site is lost and host I O is paused If you are connected to the initiator site controllers when connection to the target site is lost you will see a confirmation message on your terminal similar to that shown in Example Display 1 Refer to Appendix D for the meaning of the instance codes Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 7 1 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode Example Display 1 BuildngATop gt EVL BuildngATop 06 JUN 2001 12 57 13 Instance Code 0E098901 Template 144 90 Occurred on 06 JUN 2001 at 12 57 13 Power On Time 2 Years 72 Days 5 Hours 31 Minutes 56 Seconds Controller Model HSG80 Serial Number 2G84906303 Hardware Version E03 2B Software Version V86 1P Target Controller Board Serial Number 2G94115654 Initiator WWLID 6000 1FE1 0000 4250 0009 9411 5654 003E nitiator Node Name BUILDNGA nitiator Unit Number 1 00000001 Target WWLID 6000 1FE1 0000 4250 0009 9411 5654 003E
175. olaris The following procedure installs Command Scripter on a Sun Solaris server 8 Bl Oye Be DDr Insert the CD ROM Select the UNIX folder Select install sh Enter OK There are no arguments This returns the license agreement terms Enter YES Files are copied into directories When the system installs CPQelm select Enter Copy cmdscript to CLONE_HOME BIN the subdirectory under the default directory where the DRM scripting files reside for example usr scripts bin Close the install window and eject the CD ROM Installation is complete Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 2 9 Installation 2 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 3 CCL Setup Introduction This guide assumes that a Data Replication Manager DRM configuration is already in place to allow the scripts to run However there are some setup procedures that are important enough that the concepts are reiterated here to ensure a smooth transition to the scripting environment In particular if the Command Console LUN CCL will be used then it must be enabled on the controller and configured so the host can make use of it The CCL is a special pseudo disk device on a RAID storage system that allows the servers to communicate with the RAID array This chapter discusses how to set up a CCL for each supported operating system If the CCL is not being used for example a non RCS
176. on Data processing occurs at the initiator site and is replicated or copied to the target site If a significant failure occurs at the initiator site data processing can be resumed at the target site where the data is intact In a DRM environment a site failover makes the data available at the target site most likely after some type of failure or maintenance action Failback moves data operations back to the initiator after the initiator site has been brought back online Failsafe locked is an error mode you can set to cease initiator site I O whenever the target becomes inaccessible or the initiator unit fails You may want to transition between a failsafe locked mode and normal mode to continue or resume operations at the initiator site This transition does not constitute a failover or failback event but as a resumption of operations Usually to perform failover failback or a resumption of operations an operator must manually issue a complex series of command line interface CLI commands to a controller The use of scripts greatly reduces the need to issue many of these commands manually This can be especially beneficial with configurations containing many remote copy sets or numerous subsystems that all need to perform actions quickly during a crisis You only need to run a program file to begin issuing the appropriate CLI commands for an event action for example an unplanned failover For HP OpenVMS it is a command com file for HP
177. onfiguration BuildngA BuildngB Host A BuildngA CXO7172A Figure 1 3 Data Replication Manager basic configuration This figure uses fictional Buildng A as the initiator site and Buildng B as the target site The scripts use the following sequence for failover and failback e Failover from Buildng A to Buildng B planned unplanned or role reversal e Failback step 1 from Buildng B back to Buildng A fast full new hardware or role reversal e Failback step 2 from Buildng B back to Buildng A fast full new hardware or role reversal Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 13 DRM Scripting Overview NOTE The previous figure refers to Buildng A as the initiator site and Buildng B as the target site This does not change even after failover has occurred to Buildng B and before failback has occurred to Buildng A While in failover mode the controllers in Buildng B are acting as the initiator for all remote copy sets and are referred to as the target in this document Notice that failback with the scripts is performed in two steps e The first step adds the initiator back into the remote copy set It also performs a normalization if any new data was written to the target controllers while the initiator was inoperable It is often desirable for the system to operate in this semi failed back state while remotely mirr
178. ons via the storage system Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide C 7 Structure of the Application Action List Table C 3 Structure of the drmdispatch pl Script Command Continued ErrorLog An optional parameter that specifies the error log file when the script is run An example of the drmdispatch pl Perl script command called by the hsgcontrol pl script with parameters passed by the app act files is Perl CLONE_HOME S bin drmdispatch pl fargo PLANNED failover stepl ALL NOTFORCED Er This script starts a planned failover using the steps defined in the failover_step_1 control table All remote copy sets are processed Error conditions for a controller are not cleared by the script Error messages are placed in a log file called Er in the CLONE_HOME LOG directory where CLONE HOME is the default directory of the scripts C 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide D Troubleshooting This appendix provides several troubleshooting aids Recommendations are listed for troubleshooting common problems when using scripts A listing of error codes provides help in interpreting error messages A listing of confirmation message instance codes are also provided for examples used in this book that display these codes Troubleshooting Recommendations If errors are encountered while running the scripts consider the recommendations given b
179. ontroller Model HSG80 Serial Number Z2G84906303 Hardware Version E03 2B Software Version V86 1P Target Controller Board Serial Number 2G94115654 Initiator WWLID 6000 1FE1 0000 4250 0009 9411 5654 003E Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 8 1 Resumption of Operations After Unplanned Loss of Target Site Procedure Normal Mode Initiator Node Name BUILDNGA Initiator Unit Number 1 00000001 Target WWLID 6000 1FE1 0000 4250 0009 9411 5654 003E Target Node Name BUILDNGB Target Unit Number 1 00000001 Number of Targets 1 00000001 Remote Copy Set Name RCS1 Association Set Name AS_D1 Log Unit Number Not Available Instance Code 0E098901 gt 2 Continue the restore process with Running the Resumption of Operations Program Files Procedure Running the Resumption of Operations Program Files Procedure gt 1 Open a command prompt window on the initiator host gt 2 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 v Change to the bat sh or vms subdirectory in the CLONE HOME directory V p Run the hsg_op xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system gt 5 Youwillseea message that you are p
180. oring data before reverting to the original initiator and target roles e The second step reverses the initiator and target roles Requirements This section specifies the hardware and software required for DRM scripting Platforms Supported platforms are e HP OpenVMS Versions 7 2 1H1 7 2 2 and 7 3 e HP Tru64 UNIX Versions 5 1 and 5 1A e IBM AIX Versions 4 3 3 and 5 1 e Microsoft Windows NT Server Version 4 with Service Pack 6a e Microsoft Windows 2000 Server with Service Pack 2 Advanced Server with Service Pack 2 and Datacenter Server with Service Pack 2 e Sun Solaris Versions 2 6 7 and 8 The scripts work in a heterogeneous DRM environment with the supported platforms mentioned above to perform site failovers failbacks and other related procedures However the servers for a DRM initiator target pair must be running the same operating system The scripts can be run from any one of the scripting supported platforms and can manipulate storage that presents LUNs to other DRM supported platforms for example HP UX and Novell Netware A Command Console LUN CCL or a non RCS LUN communicates scripting operations to the controllers So the host running the scripts may or may not have access to similar types of LUNs on the controller when performing scripting operations 1 14 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview To manipulate storage on a platform not s
181. osts are shut down boot them now Booting the hosts enables AIX to recognize the drives 2 If the target hosts are not shut down use the following commands to recognize the drives and verify that they are present cfgmgr v lsdev Cc disk Use the following commands to access file systems on the failed over LUNs importvg y volumegroupname hdiskx mount all NOTE volumegroupname is the name of the volume group originally created at the initiator site and x is the number of the hdisk assigned to the failed over LUN If the y volumegroupname parameter is omitted AIX will create a default volume group name for example vg00 d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the target site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 Allow hosts to recognize new units 1 If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the target site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is
182. other applications remove all I O to the LUNs with remote copy sets that will be failed over then dismount the volumes associated with these LUNs Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 13 1 Planned Site Role Reversal Procedure b HP Tru64 UNIX If the operating system is running and is used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is used for other applications remove all I O and unmount all file system LUNs that have remote copy sets that will be failed over c IBM AIX If the operating system is running remove all I O to the remote copy set LUNs that will be failed over then unmount the file systems associated with these LUNs d Microsoft Windows NT X86 If the operating system is running shut it down and power off the hosts e Microsoft Windows 2000 If the operating system is running shut it down and power off the hosts f Sun Solaris If the operating system is running shut it down and power off the hosts gt 3 Check the error mode initiator state and target status of all remote copy sets with the CLI command SHOW REMOTE_COPY_SETS FULL Example Display 1 provides a sample output Example Display 1 BuildngATop show remote copy sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_M
183. our configuration gt 1 HP OpenVMS Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables OpenVMS to recognize the drives If the initiator site hosts are not shut down use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG gt 2 HP Tru64 UNIX Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables Tru64 UNIX to recognize the drives If the initiator hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number gt 3 IBM AIX Allow the hosts to recognize new units a If the initiator site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives If the initiator site hosts are not shut down use the following commands to recognize the drives and mount the file systems cfgmgr v mount all 13 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Planned Site Role Reversal Procedure gt 4 Microsoft Windows NT Allow the host to recognize new units Reboot the servers a
184. own boot them now Booting the hosts enables AIX to recognize the drives Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 5 Extended Planned Site Failover with Full Failback Procedure 2 If the target hosts are not shut down use the following commands to recognize the drives and verify that they are present cfgmgr v lsdev Cc disk Use the following commands to access file systems on the failed over LUNs importvg y volumegroupname hdiskx mount all NOTE volumegroupname is the name of the volume group originally created at the initiator site and x is the number of the hdisk assigned to the failed over LUN If the y volumegroupname parameter is omitted AIX will create a default volume group name for example vg00 d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the target site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 1 If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the target site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Res
185. perating system is running shut it down and power off the hosts Sun Solaris If the operating system is running and is being used exclusively for DRM operations shut down the operating system and power off the hosts If the operating system is being used for other applications remove all I O and unmount all volumes that have remote copy sets that will be failed over gt 3 Check the error mode initiator state and target status of all remote copy sets with the CLI command SHOW REMOTE_COPY_SETS FULL Example Display 1 provides a sample output 10 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Extended Planned Site Failover with Full Failback Procedure Example Display 1 BuildngATop gt show remote_copy_sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state INOPERATIVE Unit failsafe locked Target state BUILDNGB D1 is COPYING 0 complete a If all remote copy sets are in failsafe error mode and 1 Every initiator state is failsafe locked proceed to step 5 2 Every initiator state is not failsafe locked proceed to step 5 3 Initiator states are a mixture of failsafe locked or not being in failsafe locked proceed to step 4 b Ifall remote copy sets are in normal mode proceed t
186. play 2 7 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode Example Display 2 BuildngATop gt show remote_copy_sets full Name Uses Used by RCS1 remote copy D1 AS_D1 Reported LUN ID 6000 1FE1 0000 4250 0009 9411 5654 003E Switches OPERATION_MODE SYNCHRONOUS ERROR_MODE FAILSAFE FAILOVER_MODE MANUAL OUTSTANDING_IOS 20 Initiator BUILDNGA D1 state INOPERATIVE Unit failsafe locked Target state BUILDNGB D1 is COPYING 0 complete gt 3 Continue the restore process with Running the Resumption of Operations Program File Procedure Running the Resumption of Operations Program File Procedure gt 1 Open a command prompt window on the initiator host gt 2 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 gt 3 Check the error mode initiator state and target status of all remote copy sets with the CLI command SHOW REMOTE_COPY_SETS FULL Example Display 3 provides a sample output Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 7 3 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode Example Display 3 BuildngATop gt show remote_copy_sets full Name Uses Used by R
187. porting display Enter v for verbose or c for condensed To change the type of display you must delete the app fo rc file from the config subdirectory of the CLONE HOME directory and run the program file When an operation completion status result is displayed similar to Figure 6 1 continue the failover procedure at the target site with the Target Host Setup Procedure IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide Target Host Setup Procedure 1 2 To verify that failover completed successfully issue this CLI command SHOW REMOTE COPY SETS FULL The output shows the status of remote copy sets Be sure that the units you see listed under nitiator State are at the target site To verify that the target hosts can connect to the LUNs use this command SHOW UNITS FULL Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 9 3 Short Planned Site Failover with Fast Failback Procedure In the Access field of the display all units that are used by remote copy sets should show that the target host connections are enabled This should also show the initiator controller connections The following step requires actions relative to each
188. process flow for failover failback and resumption of operation scripts 1 The user runs a program file batch command or shell file to invoke the hsgcontrol pl Perl script The Perl interpreter processes the scripting instructions Parameters specified in the failover failback program file tells the script to read the application action list and what actions to perform The hsgcontrol pl script then calls the drmdispatch pl script 1 12 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Overview 3 The drmdispatch pl script performs the work of failover failback and resumption of operation The script is given parameters from the application action list app act file that specify how the actions are processed The controller configuration files and control table are read by the drmdispatch pl script and followed until all actions are performed Perl modules containing library routines can be accessed by the scripts when needed 4 The drmdispatch pl script then sends commands to the Command Scripter for inband transmission to the controller SHOW command responses are returned from the controllers and used by the scripts to verify that the commands issued to the controllers were successfully executed The disaster tolerant configuration that supports DRM involves two HSG80 Array Controller subsystems one at an initiator site and one at a target site Figure 1 3 depicts a basic DRM c
189. pter Version 1 0A e HP StorageWorks Array Controller Software ACS Version 8 7P 1 16 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 2 Installation Introduction This chapter discusses acquiring and installing software components necessary for the scripting processes to run The following software components are required to prepare a server attached to a Storage Area Network SAN for script operation e The HP DRM Scripting Kit e A Perl interpreter The Compaq SANworks Command Scripter These software components must be installed on each host at the initiator and target sites that will use scripting HP DRM Scripting Kit This software kit is obtained from the HP website http www compag com products sanworks softwaredrivers drm index html The kit contains program files Perl scripts Perl modules control tables and example files Several kits are available for various operating systems and they provide the scripts to perform failover failback and resumption of operation procedures Installing the HP DRM Scripting Kit Files The DRM Scripting Kit should be installed on an initiator and target host at a minimum to provide redundancy It can also be installed on any host to manipulate storage through a compatible controller but this method may not provide disaster tolerance when a site goes offline Use the following procedures to install the DRM Scripting Kit on your operating systems NOTE
190. pts must reside on hosts in a zone having access to the HSG80 controllers at the initiator and target sites The HSG80 controller does not distinguish between commands issued from in band command tools HP StorageWorks Command Console and HP OpenView Storage Management Appliance and commands issued out of band through the serial port Serial port commands should only be performed when the customer has restricted commanding from other sources Special care must be taken with the Management Appliance as it periodically issues polling commands that can interpret serial port communications All management appliances should be removed from switch zones in which controllers are commanded through a serial port Software The following software is required to use scripts to perform DRM failover failback and resumption of operations e HP DRM Scripting Kit Version 3 0 e Perl interpreter Open VMS Version 5 6 system privileges needed Tru64 UNIX Version 5 005 part of operating system IBM AIX Version 5 5 3 0 part of operating system Windows NT 2000 ActivePerl Version 5 6 1 628 binary kit for Win32 Solaris 6 Version 5 005 for Intel Solaris or Version 5 6 1 SPARC Solaris Solaris 7 Version 5 005 for Intel Solaris or Version 5 6 1 SPARC Solaris Solaris 8 Version 5 005 for Intel Solaris part of operating system or Version 5 6 1 for SPARC Solaris part of operating system e Compaq SANworks Command Scri
191. r are names of the target controllers that must be substituted with your target controller names 3 Three initiator target controller pairs are identified in our example but the app act file shows only two To add a third action copy and paste an existing action line with a text editor After copying another action line you would have SHORT PLANNED FAILOVER STEP1 Background fargo drmdispatch PLANNED failover stepl ALL NOTFORCED Background denver drmdispatch PLANNED failover stepl ALL NOTFORCED Background denver drmdispatch PLANNED failover stepl ALL NOTFORCED END SHORT PLANNED FAILOVER STEP I 4 For failover or failback actions change the sample DRM controller names to the actual target controller remote copy names moon venus and saturn An example would be SHORT PLANNED FAILOVER STEP1 Background moon drmdispatch PLANNED failover stepl ALL NOTFORCED Background venus drmdispatch PLANNED failover stepl ALL NOTFORCED Background saturn drmdispatch PLANNED failover stepl ALL NOTFORCED END SORT PLANNED FAILOVER STEPI1 5 For the failsafe actions SET ERROR NORMAL OPERATION SET ERROR CONFIGURED OPERATION STEPI and SET ERROR CONFIGURED OPERATION STEP2 change the sample DRM controller names to the actual initiator controller names sun mars and jupiter An example would be 4 24 Data Replicat
192. r connecting multiple hosts to dedicated storage systems With Fibre Channel the hosts can talk not only to the storage system via SCSI but also to each other via IP over the same network Fibre Channel supports existing peripheral interfaces and communications protocols including SCSI and IP Its name is somewhat misleading as Fibre Channel not only supports single mode and multi mode fiber connections but coaxial cable and twisted pair as well heterogeneous SAN A storage area network configured to support more than one host server operating system homogeneous SAN A storage area network in which all host servers run the same operating system HSG80 Array Controller An intelligent mass storage controller that interfaces between host computer systems using a Fibre Channel bus and Ultra Wide attached mass storage devices using Ultra Wide Single Ended SCSI buses hsgcontrol pl See script initiator The site that carries out primary data processing If a significant failure occurs at the initiator site data processing can be resumed at the target site where the data is intact See also target initiator configuration file A configuration file at the initiator site See also configuration file initiator IP address An acronym for Internet Protocol address The IP address is a number that is used as the address specifying a particular computer or other device connected to the Internet local terminal A terminal plugged into the
193. r info cannot Communications error Identify parse mirror size communications problem and correct 1540 9eparse unit info unknown Communications error Identify header information communications problem and correct 1541 9eparse unit info cannot Communications error Identify parse unit name communications problem and correct 1542 9eparse unit info cannot Communications error Identify parse unit size communications problem and correct Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide D 3 1522 9eparse mirror info cannot Communications error Identify parse mirror membership info communications problem and correct Troubleshooting Table D 1 Scripting Error Codes Continued Error Code 1560 parse_storage_type Communications error Identify unknown header information communications problem and correct 1561 parse_storage_type Communications error Identify unknown storage information communications problem and correct 1580 get_size cannot read size of Communications error Identify the disk communications problem and correct 1644 connect no controller Communications error Identify prompt communications problem and correct 2000 drm_site_failover cannot Configuration file problem Verify read configuration entry for correctness of the remote copy set remote copy set section to the remote copy set specified 2003 drm_site_failover cannot A site failover was attempted five
194. r of tulsa is ZG91412410 by cross referencing the device name controller name and serial number NOTE Scripts may be run from either the top or bottom controller of a controller subsystem pair So the scripts for controller tulsa can be run from the top or bottom controller of tulsa 2 Save the edited initiator configuration generation batch file into the CLONE_HOME bat subdirectory with a meaningful name like tulsa_gen bat Modify the gen ex bat file again to create a target configuration generation file for the initiator site The syntax is Perl I CLONE HOMES CLONE_HOME bin generate_cfg pl com cs RemoteCopyName SerNum 4 12 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization where RemoteCopyName identifies the DRM controller remote copy name of the target subsystem for example fargo The remote copy name can be obtained by running a SHOW THIS command from the controller SerNum specifies the device serial number that Command Scripter uses to communicate with target controller fargo from the initiator host As shown in Figure 4 1 this is the serial number of the CCL for Controller B fargo that is communicating with Host A Save this batch file in the BAT subdirectory with a meaningful name like fargo_gen bat Repeat the above steps to create initiator and target configuration generation batch files for each host running the scripts at the t
195. re Path Version 3 0 so non RCS LUNs must be used to communicate with the controller To enable the controller CCL in SCSI 2 mode with Secure Path Version 2 1D enter the following command from each controller subsystem SE SE P HIS CONTROLLER SCSI VERSION SCSI 2 P HIS CONTROLLER COMMAND CONSOLE LUN To enable the CCL in SCSI 3 mode with Secure Path Version 2 1D enter the following command from each controller subsystem S ET THIS_CONTROLLER SCSI_VERSION SCSI 3 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 3 5 CCL Setup The following sections describe file changes needed when running Secure Path Version 2 1D with Sun Solaris kernel drv sd conf All Secure Path devices must have a corresponding sd target entry Secure Path creates its own sd conf entries one per targN of the dLite conf file entries These entries are placed at the head of the sd conf file and allow Secure Path devices to configure prior to other sd targets Placing the dLite entries at the head of the sd file prevents conflicts between Secure Path entries and other SCSI bindings Secure Path entries have the format name sd parent ldLite target N lun 0 where N represents the value of targN dev Name in dLite conf For example Secure Path device target 20 would have the following entry name sd parent ldLite target 20 lun 0 When you add or r
196. rge configurations If this is the case scan only the SCSI bus where one of the DRM host adapters is located with the following command hwmgr scan scsi bus x where x is the SCSI bus number c IBM AIX Allow the hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 9 Extended Planned Site Failover with Full Failback Procedure 2 If the initiator site hosts are not shut down use the following commands to recognize the drives and mount the file systems cfgmgr v mount all d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the initiator site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 Allow hosts to recognize new units 1 If you have not changed the UNIT_OFFSET of any host connections since the hosts have been booted you do not need to reboot the initiator site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right han
197. ripting User Guide Unplanned Site Failover with Failback to New Hardware Procedure 2 If the target hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number c IBM AIX Allow the hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives 2 If the initiator site hosts are not shut down use the following commands to recognize the drives and mount the file systems cfgmgr v mount all d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the initiator site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 Allow hosts to recognize new units 1 If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the initiator site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu
198. rized Reseller mereretur TRAE T eut NN xiv 1 DRM Scripting Overview Introduction uns ie hue b ere Hau s ue oce e eee ec Maas 1 1 Site Failover Basic Description lleleeeee ee ences 1 2 Failback Procedure Choices 0 0 0 eee ete eee eee nes 1 5 Summary of Failover Failback Procedure Choices leere 1 6 Benefits Of Scripts oes eei ed m ease Tus WE bd esce B rd e c eS 1 8 Components for Scripting 2 0 eee een 1 8 HP DRM Scripting Kit ana ea Glad aa 1 8 Perl Interpreters al Lata S oOo Mode ie eee ee eas 1 9 SANworks Command Scripter 0 0 eee neces 1 9 How the Failover Failback and Resumption of Operation Scripts Work 1 9 Perl Scripts Joe peVE erden rar dena pn 1 9 User Customized Script Support Files llle 1 9 Running a Script os ccs ol tebe bee e ES Erie 1 10 The Scripting Process Flow socer re ep Me eR 1 12 Requirements 4442 n 2 as ee ar nes dana We eet a ga eve wg ges 1 14 Platforms pits b een RU EE e mb PE EE 1 14 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide iii Contents Hardwate S ee hb ee cbt ona eG lo marae a ea eats Be 1 15 Switch Zoni g eee a Ree en e C eaa 1 16 SOL Wale soper sao oua thane Bl wie Ple ei ea icu tuse a 1 16 2 Installation Introduction 2 8 ined dc eT t e eas Tug t c in CE 2 1 HP DRM Scripting Kit iiie bier tan c PS ede bids 2 1 Installing the HP DRM Scripting Kit Files 2 2 0 0 0 eee eee
199. round denver drmdispatch PLANNED fast failback stepl ALL NOTFORCED END FAST FAILBACK STEP1 FAST FAILBACK STEP2 Background fargo drmdispatch PLANNED failback step2 ALL NOTFORCED Background denver drmdispatch PLANNED failback step2 ALL NOTFORCED END FAST FAILBACK STEP2 ULL FAILBACK STEP1A ackground fargo drmdispatch DISASTER failback stepla ALL NOTFORCED ackground denver drmdispatch DISASTER failback stepla ALL NOTFORCED ND FULL FAILBACK STEP1A w w r 7 FULL_FAILBACK_STEP1B Background fargo drmdispatch DISASTER failback_steplb ALL NOTFORCED Background denver drmdispatch DISASTER failback_steplb ALL NOTFORCED z 0 FULL_FAILBACK_STEP1B Ei FULL_FAILBACK_STEP1C Background fargo drmdispatch DISASTER failback_steplc ALL NOTFORCED Background denver drmdispatch DISASTER failback_steplc ALL NOTFORCED END_FULL_FAILBACK_STEP1C FULL_FAILBACK_STEP2 Background fargo drmdispatch DISASTER failback_step2 ALL NOTFORCED B ackground denver drmdispatch DISASTER failback_step2 ALL NOTFORC END_FULL_FAILBACK_STEP2 Eri iw ROLE_REVERSE_FAILOVER_STEP1 Background fargo drmdispatch PLANNED failover_stepl ALL NOTFORCED Background denver drmdispatch PLANNED failover_stepl ALL NOTFORCED END_ROLE_REVERSE_FATLOVER_STEP1 I
200. s 6 Hours 50 Minutes 57 Seconds Peer to Peer Remote Copy target controller HSG80 Serial Number Z2G94115654 Hardware Version E10 28 V86 1P Other Controller Serial Number Z2G94319198 Target Number 0 00 LUNs Taken By This Controller Open a command prompt window on the initiator host Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 1 codes Example Display 3 BuildngATop gt SEVL BuildngATop Template 5 05 Power On Time 2 Event reported by Controller Model Software Version Instance Code 07050064 Last Failure Code SEVL BuildngATop Template 4 04 Power On Time 2 Event reported by Controller Model Software Version Failed Controller 00000003 00000000 00000000 00000000 00000000 00000000 00000000 00000000 Instance Code 43010064 2 3 4 Change to the bat sh or vms subdirectory in the CLONE HOME directory 11 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Replication After Extended Planned Loss of Target Procedure Failsafe Mode gt 5 Runthe hsg_op xxx program file where xxx denotes com OpenVMS sh Tru64 AIX and Solaris or bat Windows NT 2000 depending on your operating system gt 6 Youwillseea message that you are performing a DRM operation to modi
201. s configuration generation files It includes the complete directory path to the device name Refer to Chapter 4 on how to obtain the complete device name DeviceName in Windows can either be the drive letter of the non RCS LUN when not using the CCL for example Q or the controller CCL name in the format Scsi3 1 124 0 Windows systems do not assign device names to the CCL A pseudo name for the CCL is created in the format Adapter Bus Target LUN Chapter 4 describes how this format can be converted to a serial number when creating a configuration generation file The general syntax for checking whether Command Scripter can communicate through a device when knowing its serial number as in Windows systems that use the CCL is cmdscript n SerialNumber show this 5 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Scripting File Descriptions and Behaviors The general syntax for checking whether Command Scripter can communicate through a device when knowing its worldwide unique identifier is cmdscript w WorldwideID show this Whenever you want to obtain a list of available controllers seen by Command Scripter use the command cmdscript f subsysdata Verbose and Condensed Displays The first time a program file is run to launch a script you are prompted to select either a verbose or condensed reporting display A verbose display lists the status of the controller and all remote copy
202. s of each other Mirrorsets provide the highest level of data availability at the highest cost Another name for RAID I Also called mirrored units or mirrored virtual disks 2 Two or more physical disks configured to present one highly reliable virtual unit to the host 3 A virtual disk drive consisting of multiple physical disk drives each of which contains a complete and independent copy of the entire virtual disk s data multiple intersite links Each intersite link ISL is a fiber link between two switches As applied to Data Replication Manager increasing bandwidth between switches is handled by adding additional connections between the switches with a maximum of two connections normalization A state in which block for block data written by the host to a mirrorset member is consistent with the data on other normal and normalizing members The normalizing state exists only after a mirrorset is initialized normalizing member A mirrorset member whose contents are the same as all other normal and normalizing members for data that has been written since the mirrorset was created or since lost cache data was cleared A normalizing member is created by a normal member when either all of the normal members fail or all of the normal members are removed from the mirrorset See also copying member Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary 7 Glossary normal member A mirrorset member that b
203. s of implementing SANs centralized and decentralized A centralized SAN ties multiple hosts into a single storage system which is a RAID device with large amounts of cache and redundant power supplies The cabling distances allow for local as well as campus wide and metropolitan wide hookups over peripheral channels rather than over an overburdened network SCSI distances have also been extended This centralized storage topology is commonly employed to tie a server cluster together for failover Fibre Channel is a driving force in the SAN arena because it supports existing peripheral interfaces as well as network interfaces Fibre Channel can be configured point to point in an arbitrated loop FC AL or via a switch With Fibre Channel the hosts can talk not only to the storage system via SCSI but they can also communicate with each other via IP over the same topology If a centralized storage system is not feasible a SAN can connect multiple hosts with multiple storage systems Glossary 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary storage array An integrated set of storage devices Storage arrays can be manipulated as one unit with a single command storageset 1 A group of devices configured with RAID techniques to operate as a single container 2 Any collection of containers such as stripesets mirrorsets striped mirrorsets JBODs and RAIDsets storage unit The general term that r
204. sets while the script runs see Figure 5 1 A condensed display lists only the controller name and its status see Figure 5 2 Terminal window Edit Options UNPLANNED_FAILOVER_STEP2 Status start time 07 Sep 2001 14 20 17 current time 07 Sep 2001 14 20 48 Processing Working 0 Sep 2001 14 2 Queued 07 Sep 2001 14 2 0 0 Queued 07 Sep 2001 14 20 Queued 07 Sep 2001 14 20 Figure 5 1 Verbose status display Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 5 3 Scripting File Descriptions and Behaviors Window Edit Options Help FULL_FAILBACK_STEP1A Status start time 07 Sep 2001 15 44 37 current time 0 Sep 2001 15 44 43 Figure 5 2 Condensed status display The verbose or condensed option is set in a RC file That option remains in effect until you change it To change the display option you must delete the associated RC file Table 5 1 lists the specific RC files created from a particular failover failback or resumption of operation program file Delete the RC file in the config subdirectory of the CLONE_HOME directory for the type of failover or failback you plan to run You are then prompted again for the verbose or condensed output format the next time the program file is run Your choice of display output causes that program file to create another RC file to store that preference NOTE The verbose status display is not currently supported with HP OpenVMS Table
205. sing an account that has administrative privileges 2 Open Computer Management and click Disk Management 3 After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice b If you have changed the UNIT OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice This completes the procedure for performing an unplanned site failover with a full failback 6 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 7 Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode This chapter provides the procedure for performing a resumption of operations after an unplanned loss of the target site while running in failsafe error mode When the error mode of the remote copy set is in failsafe and the connection to the target site is lost host I O is paused This occurs because the initiator state of the remote copy set becomes inoperative while being failsafe locked These procedures are used to resume host I O until the connection to the target site is re established The sequence of steps are e Verification of Lost Connections Pr
206. splay scien eie ee ieee 5 3 5 2 Condensed status display 0 0 cee cc eee eens 5 4 6 1 Operation completion status result display 0 0 0 0 cece eee eee 6 3 Tables 1 Related Documentation 0 0 0 0 cece cece eect ene nee ix 2 Document Conventions 0 cece eee eee een eee xi 1 1 Possible Failover Situations lees 1 3 1 2 Types of Failover and Failback eee eee eee 1 5 1 3 Failover Failback and Resumption of Operations Procedure Choices 1 6 1 4 DRM Heterogeneous Operating Systems 0 000000 eee eee 1 15 4 1 Created or Customized Host Files 0 0 0 0 cece eee eee eee 4 2 5 1 RC File Location vene a bbe et beds ar SEP I ERA ees 5 4 A 1 Installed DRM Scripting Kit Files 0 eee ee eee A 1 C 1 Structure of an Action Command 0 0 cee eee ee eee C 5 C 2 Structure of the hsgcontrol pl Script Command C 6 C 3 Structure of the drmdispatch pl Script Command C 7 D 1 Scripting Error Codeso iee crore cen n da i e ee D 2 D 2 Instance Code Legend e erconissro creeat a stee kunne ee D 9 viii Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide About this Guide This user guide provides information to help you e Acquire and install software required for using Data Replication Manager DRM scripts e Configure and customize files required for the scripts e Run failover f
207. st after the data is safe on the initiating controller and prior to the completion of the target command Asynchronous mode can provide greater performance and faster response time but the data on all members at any one point in time cannot be assumed to be identical See also synchronous mode batch file A text file containing Windows operating system commands that are used to invoke Perl scripts The batch file hsg_fo bat for example invokes a Perl script that performs a site failover on controllers identified in the application action list CCL See command console LUN CLI See command line interface CLI command CLI commands allow users to manage their subsystems by viewing and modifying the controller and its attached devices A primary function of CLI commands is to control the failover mode of a controller pair CLI SHOW command See SHOW commands clone A utility that physically duplicates data on any unpartitioned single disk unit stripeset mirrorset or striped mirrorset command console LUN CCL A special pseudo disk device on the RAID storage system that allows the server to communicate with the RAID array command line interface CLI The configuration interface that operates the controller software Also known as command line interpreter Command Scripter Application software that provides an interface to communicate the CLI commands generated by the Perl scripts to the HSG80 controllers via the Fibre Cha
208. subdirectory of the CLONE HOME directory and run the program file IMPORTANT If a script encounters an abnormal condition and exits with a warning or error abort the scripting activities and perform the remaining steps manually These procedures are located in the HP StorageWorks Data Replication Manager HSG80 ACS Version 8 7P Failover Failback Procedures Guide Continue with the planned failover at the target site with Target Host Setup Procedure Target Host Setup Procedure 1 2 To verify that the target site host can connect to the LUNs use this command SHOW UNITS FULL In the Access field of the display all units that are used by remote copy sets should show that the target host connections are enabled The display should also show the initiator controller connections The following step requires actions relative to each operating system in your configuration a HP OpenVMS Allow the hosts to recognize new units 1 If you have shut down the host boot it now Booting the host enables OpenVMS to recognize the drive 13 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Planned Site Role Reversal Procedure 2 If you did not shut down the host use the following command from a privileged account to enable OpenVMS to recognize the drives MCR SYSMAN IO AUTOCONFIGURE LOG b HP Tru64 UNIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boo
209. sure that all enclosures switches and rack PDUs have their power switches in the OFF position 2 Apply power to all PDUs 3 Turn on the power switches for the racks from the initiator site Make sure that all controllers are on and functional VV VV PN 5 Apply power to all Fibre Channel switches Power Down Data Replication Manager Systems Power down the DRM systems in the sequence described in the following procedures If the initiator site will be powered down for a long period of time you may need to disable cache batteries Initiator Site Power Down Procedure gt 1 Issue the following CLI commands in the order shown SHUTDOWN OTHER_CONTROLLER SHUTDOWN THIS_CONTROLLER gt 2 Turn off the Fibre Channel switches gt 3 Turn off the power to the enclosures gt 4 Turn off the PDUs When completed go to the Target Site Power Down Procedures E 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Power Up and Power Down Target Site Power Down Procedure 1 Issue the following CLI commands in this order SHUTDOWN OTHER_CONTROLLER SHUTDOWN THIS_CONTROLLER 2 Turn off the Fibre Channel switches 3 Tum off the power to the enclosures 4 Turn off the PDUs Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide E 3 DRM Power Up and Power Down E 4 Data Replication Manager HSG
210. synchronous operation mode before starting a planned failover See also disaster tolerance DT synchronous mode port In general terms a port is e A logical channel in a communications system e The hardware and software that connect a host controller to a communications bus such as a SCSI bus or serial bus With respect to a controller a port is the logical route for data in and out of a controller that can contain one or more channels all of which contain the same type of data With respect to a SCSI system a port is the hardware and software that connect a controller to a SCSI device Glossary 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary RAID Redundant Array of Independent Disks A disk subsystem that increases performance and provides fault tolerance RAID is a set of two or more hard disks and a specialized disk controller that contains the RAID functionality RAID improves performance by disk striping which interleaves bytes or groups of bytes across multiple drives so more than one disk is reading and writing simultaneously Fault tolerance is achieved by mirroring or parity RAID 0 Provides disk striping only which interleaves data across multiple disks for better performance It does not provide safeguards against failure RAID 1 Uses disk mirroring which provides 100 duplication of data Offers highest reliability but doubles storage cost RAID 2 Bits rather t
211. t perform other processing tasks while the scripts are running This includes running more than one instance of the scripts issuing commands through Command Scripter and issuing commands directly on the controller through a terminal server or serial port This may cause the controller to hang Communicating Via Command Scripter Command Scripter accesses storage subsystems through device drivers on the local host system when given a supplied device name controller serial number or worldwide unique identifier Command Scripter will use the supplied device name to attempt to locate a Command Console LUN CCL on the same controller If a CCL is found access to the controller will be through the CCL Otherwise access will be through the specified device The general syntax for checking whether Command Scripter can communicate through the specified device is cmdscript f DeviceName show this_controller DeviceName in OpenVMS is the device identification number described in Chapter 4 when setting up the OpenVMS configuration generation files and obtained by entering the SHOW DEVICE GG command DeviceName in Tru64 UNIX Version 5 x systems is the name of the device special file name This is the CCLName described in Chapter 4 when setting up the Tru64 configuration generation files and obtained by using the hwmgr view devic command DeviceName in AIX and Solaris is the idnum described in Chapter 4 when setting up the AIX or Solari
212. t the initiator site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice gt 5 Microsoft Windows 2000 Allow hosts to recognize new units a If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the initiator site hosts 1 On each host log in using an account that has administrative privileges 2 Open Computer Management and click Disk Management 3 After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice b If you have changed the UNIT_OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You see all of the units in Disk Management If Secure Path is not installed correctly you will see each drive twice gt 6 Sun Solaris Allow the hosts to recognize new units a Power on or reboot the servers using the command reboot r atthe target site and log in using an account that has administrative privileges b If Secure Path was not configured for these units you will not see the drives You need to edit the WWIDs in the kernal drv IdLite conf file To find th
213. t them now Booting the hosts enables Tru64 UNIX to recognize the drives 2 Ifthe target hosts are not shut down use the following command to recognize the drives hwmgr scan scsi This may take awhile for large configurations If this is the case scan only the SCSI bus where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number c IBM AIX Allow the hosts to recognize new units 1 If the target site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives 2 Ifthe target hosts are not shut down use the following commands to recognize the drives and verify that they are present cfgmgr v lsdev Cc disk Use the following commands to access file systems on the failed over LUNs importvg y volumegroupname hdiskx mount all NOTE volumegroupname is the name of the volume group originally created at the initiator site and x is the number of the hdisk assigned to the failed over LUN If the y volumegroupname parameter is omitted AIX will create a default volume group name for example vg00 d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the target site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice Data Replication Manager H
214. tain potentially catastrophic error conditions The failsafe locked condition prevents further write operations from the host to the remote copy set members The failsafe error mode is enabled by the user to fail any I O whenever the target is inaccessible or the initiator unit fails When either of these conditions occurs the remote copy set goes into the inoperative offline state fast failback The synchronization of the initiator site with the target site during a planned failover of the initiator subsystem The write operations are logged to the target site write history log and during the fast failback the initiator site is updated from the write history log See also mini merge disaster tolerance DT planned failover write history logging fiber An optical strand used in fiber optic cable Spelled Fibre when used in Fibre Channel protocol See also fiber optic cable Fibre Channel fiber optic cable A transmission medium that transmits digital signals in the form of pulses of light Fiber optic cable is noted for its properties of electrical isolation and resistance to electrostatic contamination Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Glossary 5 Glossary Fibre Channel A high speed transmission technology that can be used as a front end communications network a back end storage network or both at the same time Fibre Channel is a driving force in the storage area network SAN arena fo
215. te controller Controller B communicating with the initiator host Host A 4 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization Setup for One Controller Subsystem Initiator Site Target Site Host A Host B CCL Controller A Controller B Setup for Maximum Controllers Per Host Initiator Site Target Site Host A Host B Controller B CCL Controller C Controller D ji N CCL CCL Controller E Controller F CXO7733A Figure 4 1 Generation file setup using a CCL Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 5 File Customization 4 Save this program file in the vms subdirectory with a meaningful name like fargo_gen com Repeat the above steps to create initiator and target configuration generation files for the target site Use the correct device identification numbers of the CCLs communicating with the controllers Save these files in the vms subdirectory on the target host using the same naming convention as used previously HP Tru64 UNIX The following procedure describes how to create initiator and target configuration generation files when using a CCL in SCSI 2 or SCSI 3 mode 1 Modify the gen_ex sh file to create an initiator configuration generation file for the initiator site Use a text editor and make the necessary modifications using the following syntax perl I CLONE_HOME
216. tested with the scripts was Perl Version 5 6 0 and is labeled as OpenVMS Alpha 7 2 1 Dec C Sockets Installing the OpenVMS Perl Interpreter Follow the OpenVMS installation instructions located on the website listed above Microsoft Windows NT 2000 Obtaining the Windows Perl Interpreter ActivePerl The Perl interpreter for the Windows platforms can be downloaded from http aspn activestate com ASPN Downloads ActivePerl HP tested the scripts with the ActivePerl 5 6 1 628 MSI package for Windows Previous versions of the ActivePerl program are also available at the site Installing Windows ActivePerl Follow the Windows installation instructions located on the Activestate website listed above Windows NT Server users must have installed or must download Microsoft Windows Installer version 1 1 or later and must be operating with Service Pack 5 or later No additional software is needed for Windows 2000 servers 2 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Installation Sun Solaris Versions 6 and 7 Obtaining the Sun Solaris Versions 6 and 7 Perl Interpreter The Perl interpreter for the Sun Solaris platforms can be downloaded from http www sunfreeware com At the time this document was prepared there were two versions of the Solaris Perl interpreter Perl Version 5 6 1 was available for the SPARC Solaris and Perl Version 5 005 was available for Intel Solaris Installing the Sun Sol
217. tiator and target hosts and the application action list app act When you are finished the files on the initiator host and target host are nearly identical 4 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization Configuration Generation File Creation During installation of the DRM Scripting Kit for a specific operating system one of three types of example configuration generation files were extracted and placed in a CLONE_HOME subdirectory where CLONE_HOME is the default directory where the scripts reside e OpenVMS files reside in the vms subdirectory e Tru64 UNIX AIX and Solaris files reside in the sh subdirectory e Windows files reside in the bat subdirectory The example configuration generation file gen_ex xxx where xxx is the bat sh or com file extension is provided as a template file It must be modified to create one customized configuration generation file for each HSG80 controller subsystem at both the initiator and target sites The syntax of the configuration generation file differs depending on the operating system and whether a CCL or non RCS LUN is used The modified template file is saved as an executable program file and then run at the initiator site and again at the target site to create individual controller configuration files on all hosts that will use the scripts Before creating configuration files for your scripts ensure that your DRM configuration is
218. ting the IdLite conf mda conf and sd conf files reboot the server using the reboot r command You should now be able to see the drives using the format command Refer to the Compaq StorageWorks Secure Path for Sun Solaris Installation and Reference Guide for assistance 4 Once the problem that disabled the initiator site is remedied continue the full failback procedure at the initiator site with the Running the Full Failback Program Files Procedure Running the Full Failback Program Files Procedure gt 1 Before performing the failback procedure locate your record of SHOW command output that details the initiator configuration Verify that your initiator controller configuration is the same as your target controller configuration Compare the status of the controllers association sets remote copy sets units and connections Refer to the HP StorageWorks Data Replication Manager HSG80 ACS 8 7P Failover Failback Procedures Guide for the full status comparison procedure gt 2 Open a command prompt window on the initiator host gt 3 Verify that the script server can communicate with the target controller via Command Scripter Use the cmdscript command as described in the section Communicating Via Command Scripter on page 5 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 6 5 Unplanned Site Failover with Full Failback Procedure Change to the bat sh or vms subdirectory in the CLO
219. ting with the controllers Save these program files in the sh subdirectory on the target host using the same naming conventions used previously Microsoft Windows NT 2000 Windows NT 2000 can be run in SCSI 2 mode with no CCLs or in SCSI 3 mode with CCLs The following procedures show how to use either method SCSI 2 Mode with No CCL Enabled In SCSI 2 mode with no CCL enabled the scripting hosts communicate with controllers through LUNs These must be non RCS LUNs with sticky drive letters meaning that they are persistent during restarts of the host Hosts at the initiator and target sites must see these LUNs as the same drive letter For example if Host A initiator sees this LUN as Q then Host B target must see the same LUN as Q also refer to Figure 4 2 There is a maximum limit of three controllers per host Follow these steps to create SCSI 2 initiator and target controller generation batch files 1 Ensure you have one non RCS LUN enabled on each initiator and target controller subsystem that each host running the scripts can access see Figure 4 2 This means that the initiator host must have access to LUNs at the initiator and target sites and the target host has access to LUNs at the initiator and target sites If zoning is enabled make sure the initiator and target hosts reside in the same zone Modify the gen_ex bat file to create an initiator configuration generation file for each initiator site controller
220. ts at the target site Use the correct device identification numbers of the CCLs communicating with the controllers Save these program files in the sh subdirectory on the target host using the same naming conventions used previously 4 14 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide File Customization Using the Non RCS LUN No CCL l Modify the gen_ex sh file to create an initiator configuration generation file for each host running the scripts at the initiator site Use a text editor and make the necessary modifications using the following syntax perl I CLONE_HOME CLONE_HOME bin generate_cfg pl com cs RemoteCopyName DeviceID where RemoteCopyName identifies the DRM controller remote copy name of the subsystem for example tulsa This name is case sensitive in Solaris and must match the case of the name used later when configuring the application action list DeviceID specifies the identifier for the non RCS LUN you created to allow Command Scripter to communicate with the initiator controller tulsa The setup is similar to that shown in Figure 4 2 where instead of a drive letter this is the device identifier of the LUN for the local controller Controller A communicating with the initiator host Host A The device identifier can be found using the following method Run a SHOW CLI command on the non RCS LUN you created and note the LUN ID or WWLUN_ID number See the
221. uld not be modified CONFIG fast_failback_ Contains step 1 fast failback instructions read by the step 1 tbl armdispatch pl Perl script This file should not be modified CONFIG _failsafe_op tbl Contains instructions that modify the error mode of the remote copy sets to normal mode in order to resume operations at the initiator site due to a planned or unplanned loss of the target site while in failsafe error mode This file should not be modified CONFIG failsafe_revop1 tbl Contains step 1 of instructions that reverts the remote copy set to a former error mode This file should not be modified CONFIG failsafe_revop2 tbl Contains step 2 of instructions that reverts the remote copy set to a former error mode This file should not be modified CONFIG hsg80class msg Contains a translation between error codes and text messages that are displayed in the CHK files generated by the scripts CONFIG prefast_failover_ Contains step 2 pre fast failover instructions read by the step2 tbl armdispatch pl Perl script This file should not be modified CONFIG prefast_failover_ Contains step 3 pre fast failover instructions read by the step3 tbl armdispatch pl Perl script This file should not be modified A 2 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide DRM Scripting Kit Files Table A 1 Installed DRM Scripting Kit Files Continued Directory Filename Description CONFIG prefull_failover_ step2 tbl
222. units in Disk Management If Secure Path is not installed correctly you will see each drive twice This completes the procedure for running a short planned site failover with a fast failback Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 9 9 Short Planned Site Failover with Fast Failback Procedure 9 10 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 10 Extended Planned Site Failover with Full Failback Procedure This chapter provides the procedure for performing an extended planned site failover with a full failback to existing hardware This procedure is used to ensure the proper functioning of a planned failover and subsequent full failback for extended initiator site maintenance activities It is expected that the duration of the event exceeds the ability of the write history log to capture all of the host I O Therefore a full normalization of the remote copy sets is needed after failback The sequence of steps are Initiator Site Preparation Procedure Running the Extended Planned Failover Program File Procedure Target Host Setup Procedure Running the Full Failback Program Files Procedure Initiator Site Cleanup Procedure NOTE In this chapter the gt symbol is used to identify a procedural step performed at the initiator site and the symbol is used to identify a target site procedural step Initiator Site Preparation Procedure gt a Before performing the
223. upported by the scripts use a controller shared by a host capable of running the scripts The scripts can manipulate all storage on a controller when communicating through a LUN Hardware DRM supports compatible operating systems sharing the same HSG80 controller To be compatible all of the operating systems must support the same level of SCSI command and control either SCSI 2 with the CCL turned on or off or SCSI 3 with the CCL turned on Table 1 4 lists the current DRM supported operating systems and the SCSI level that each supports Table 1 4 DRM Heterogeneous Operating Systems Operating System SCSI 2 SCSI 3 HP OpenVMS No Yes HP Tru64 UNIX Yes Yes HP UX Yes Yes IBM AIX Yes Yes Microsoft Windows NT 2000 Yes Yes Novell Netware Yes No Sun Solaris Yes Yes IMPORTANT Although Windows can be run in SCSI 2 or SCSI 3 mode in a DRM environment SCSI 2 mode is only supported with no CCL enabled When sharing a controller in SCSI 2 mode between Windows and another operating system using a CCL HP recommends that scripting be performed with the Windows host Sun Solaris running Secure Path Version 2 1D supports SCSI 2 mode with or without a CCL but not SCSI 3 mode Sun Solaris running Secure Path Version 3 0A supports SCSI 2 or SCSI 3 but without the use of CCLs Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 1 15 DRM Scripting Overview Switch Zoning The scri
224. uration generation files 4 3 Connections Section 4 21 controller configuration file 4 19 list of files on host 4 2 Maximum Read Write Cached Transfer Size 4 22 process steps 4 1 Remote Copy Set Section 4 20 D Data Replication Manager DRM configuration basics 1 13 description 1 1 device identification number IBM AIX 4 8 OpenVMS 4 4 Sun Solaris 4 13 4 14 Tru64 UNIX 4 6 4 7 dialog box names convention defined xi document conventions xi prerequisites x documentation related ix DRM Scripting Kit description 1 8 download location 2 1 files included A 1 HP OpenVMS installation 2 2 HP Tru64 UNIX installation 2 2 IBM AIX installation 2 2 Microsoft Windows NT 2000 installation 2 3 required component 1 8 Sun Solaris installation 2 2 types available 1 8 version required 1 16 drmdispatch pl 1 12 command structure C 7 description A 1 C 6 syntax C 7 E environmental variable 2 3 error codes D 2 F failback defined 1 1 scenarios 1 6 types of 1 5 5 1 failover defined 1 1 1 2 planned 1 4 role reversal 1 4 scenarios 1 6 situations 1 3 types of 1 5 5 1 unplanned 1 4 failsafe locked defined 1 1 mode of operation 1 6 Fibre Channel bus 1 10 file customization steps 4 1 file names convention defined xi finding controller serialnumber 3 5 4 12 G gen_ex file 4 3 to 4 16 generate cfg pl 4 17 A 1 getting help xiii H help obtaining xiii heterogeneous DRM environment 1 14 1 15 HP OpenVMS CCL setup 3 1 Co
225. us where one of the DRM host adapters are located with the following command hwmgr scan scsi bus x where x is the SCSI bus number c IBM AIX Allow the hosts to recognize new units 1 If the initiator site hosts are shut down boot them now Booting the hosts enables AIX to recognize the drives 2 If the initiator site hosts are not shut down use the following commands to recognize the drives and mount the file systems cfgmgr v mount all 7 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide Resumption of Operations After Unplanned Loss of Target Site Procedure Failsafe Mode d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the initiator site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 Allow hosts to recognize new units 1 If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the initiator site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not inst
226. vailable 20 58 00 6 0 DEC HSG80 Command Console LUN hdisk2 Available 20 58 00 6 1 DEC HSG80 RAID Array 3 4 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide CCL Setup Microsoft Windows NT 2000 CCL Setup Windows supports the CCL in SCSI 3 mode only The following steps enable the CCL and configure it on a Windows platform 1 To enable the CCL enter the following command from each controller subsystem SET THIS_CONTROLLER SCSI_VERSION SCSI 3 Check which CCLs can be seen by entering the following command from each host used by the scripts CMDSCRIPT F SUBSYSDATA The system displays a list of CCLs similar to the following Available devices Device Name Vendor Product ID Serial Num FW Scsi4 1 0 0 DEC HSG80CCL 2G91412410 V87P Scsi4 1 2 0 DEC HSG80CCL 2G91205687 V87P Scsi5 1 1 0 DEC HSG80CCL 2G91416136 V87P Scsubilt3 s0 DEG HSG80CCL 2G91606296 V87P To ensure that a host can see the CCL enter the following command using a device name like one of those shown in the example display in the previous step CMDSCRIPT F DeviceName SHOW THIS CONTROLLER Example cmdscript f Scsi4 1 0 0 show this controller The resulting display should show all the characteristics of the controller verifying that the host sees the CCL Sun Solaris CCL Setup Sun Solaris supports the CCL when running Secure Path Version 2 1D in SCSI 2 mode and SCSI 3 mode A CCL is not supported with Secu
227. value displayed as the NODE_ID in the controller data The SCSI ID of the CCL must also be converted to a hexadecimal value The hexadecimal value for HSG80CCL is 4853 4738 3043 434C For example to assign the CCL to target 20 of a StorageWorks HSG80 Storage System with a Node ID of 5000 1FE1 0001 ED 10 dLite conf must have the following entry targ20 devName 5000 1FE1 0001 ED10 4853 4738 3043 434C Reboot the servers with the command reboot r You should now be able to see the CCLs using the format command Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 3 7 CCL Setup 3 8 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide 4 File Customization Introduction This chapter takes you through the customizations required for the configuration generation files target controller configuration files and the application action list These customizations must be made to files on both the initiator and target hosts so the scripts may be run from either site for redundancy File Customization Steps The following list summarizes the file configuration process explained in this chapter Create executable files to simplify the configuration generation task see Configuration Generation File Creation on page 4 3 One executable configuration generation file is created for each controller subsystem on each host The executable files are run on each host to create initi
228. vice identification number for the CCL of the controller at the target site Controller B with access to the initiator host Host A Use the syntax perl I CLONE_HOME SCLONE_HOME bin generate_cfg pl com cs RemoteCopyName idnum where RemoteCopyName identifies the DRM controller remote copy name of the target subsystem for example fargo This name is case sensitive in Tru64 UNIX and must match the case of the name used later when configuring the application action list idnum specifies the device identification number for the operating system that Command Scripter uses to communicate with controller fargo As shown in Figure 4 1 this is the identification number of the CCL for the remote controller Controller B communicating with the initiator host Host A 4 Save this program file in the sh subdirectory with a meaningful name like fargo_gen sh 5 Repeat the above steps to create initiator and target configuration generation program files for the target site Use the correct device identification numbers of the CCLs communicating with the controllers Save these generation files in the sh subdirectory on the target host using the same naming conventions used previously IBM AIX The following procedure describes how to create initiator and target configuration generation files when using a CCL in either SCSI 2 or SCSI 3 mode 1 Modify the gen_ex sh file to create an initiator configuration generation f
229. volumegroupname hdiskx mount all NOTE volumegroupname is the name of the volume group originally created at the initiator site and x is the number of the hdisk assigned to the failed over LUN If the y volumegroupname parameter is omitted AIX will create a default volume group name for example vg00 d Microsoft Windows NT Allow the host to recognize new units Reboot the servers at the target site and log in using an account that has administrative privileges You should be able to see all of the units in My Computer If Secure Path is not installed correctly you will see each drive twice e Microsoft Windows 2000 Allow hosts to recognize new units 1 If you have not changed the UNIT OFFSET of any host connections since the hosts have been booted you do not need to reboot the target site hosts a On each host log in using an account that has administrative privileges b Open Computer Management and click Disk Management c After Disk Management has initialized go to the Action Menu and click Rescan Disks All of the failed over units should appear in the right hand pane If Secure Path is not installed correctly you will see each unit twice 2 If you have changed the UNIT OFFSET of any host connections you must reboot that host After the server has rebooted log in using an account that has administrative privileges You should be able to see all of the units in Disk Management If Secure Path is not installed corr
230. zation 0 0 0 cece eee eee ee ee 4 19 Target Controller Configuration File Customization 2222222000 4 19 Association Set Section cesis ee eee d En EET eee nee 4 19 Remote Copy Set Section sose sa masie a eee Roper Te en 4 20 Connections Section llle eens 421 Maximum Read Write Cached Transfer Size Section 4 22 Application Action List Customization 4 23 Customizing the Application Action List 2 0 0 0 cee eee eee eee 4 23 Example Customization of an Application Action List 2222220 4 24 5 Scripting File Descriptions and Behaviors Program File Descriptions 0 0 cece e E a e eens 5 1 Communicating Via Command Scripter 0 00 c ee eee eee ee 5 2 Verbose and Condensed Displays 0 0 cece eee eee eens 5 3 Terminating a Script a2 ee Seda ohh ide Disco ee a 5 5 HP Open MS e352 eed ee re 5 5 HP Tru64 UNIX IBM AIX and Sun Solaris 00005 5 6 Data Replication Manager HSG80 ACS Version 8 7P Scripting User Guide v Contents Microsoft Windows NT 2000 2 2 cc mouse eee eee 5 6 Unplanned Site Failover with Full Failback Procedure Running the Unplanned Failover Program File Procedure 2 6 1 Target Host Setup Procedure 0 0 cece cette eee eee 6 2 Running the Full Failback Program Files Procedure 0 000000 0 00 6 5 Initiator Site Cleanup Procedure 0 0 ce

Data Replication Manager HSG80 Version 8.7P Scripting User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents