Home

Proxy Sniffer User`s Guide

1. The following Alert Conditions are supported 2 PRX Alert Notifications Mozilla Firefox HAr EB http 127 0 0 1 7990 dfischer webadmininterFace PopupPersonalSettingsAlertConfiqur ationWeblet x e fa Job cannot be started es a aie Alert Notifications Local Alert Configuration Pat x e At the Start of a Job information Default Setting Alerts are Enabled by default Alerts are Disabled by default 1 1 Generic Message Prefix Message Prefix for Internal Errors fatal 9 If an Internal Error occurs during the Execution of a Message Prefix for Measured Errors error i Message Prefix for Exceeded Limits exceeded Job Message Prefix for Information Messages info Message Prefix for Cancelled Alerts cancelled e During the Execution of a Job In Periodically 7 Send Email Alerts to SMTP Server Test Connection from All Exec Agents V Y Send SMS Alerts to Clickatell HTTP Gateway Test Connection from AIl Exec Agents v i y a SMTP Server IP Address or DNS Name 1192 16 4 31 Clickatell Username miller Intervals configurable interval time In minutes SMTP Server Auth Username Clickatell Password f SMTP Server Auth Password l E Clickatell API ID 3240755 O At Eve ry nterval info rm atio n From Email Address prxsniffalert d fischer com To Mobile Numbers 41774582420
2. x 1 1 O00 sec 10463 bytes 422 ms d ccr http www praxy sniffer dei 200 0K TEXT HTML x 2 2 0 53sec 5593 bytes 172ms A GET http pros sniffer deiformatcss 200 0K TEXTICSS x 3 E 0 63 sec b 4 bytes 313 ms B GET http www google analytics somi rehin is gt 00 OK TEXTUAVASCRIPT x 4 4 0 72 ser 4d bytes 172ms MIGET hip pros dif 200 OK IMAGEIGIF x 5 5 Of2sec 58 bytes 203 ms iii GET http ed 12xBH qgif 200 OK IMAGESGIF x B B O73sec 22428 bytes B72ms ll GET T BcreenShotClusterPreview jpq 200 OK IMAGENPEG x 1 O73 sec 5 925 bytes 297 ms Ml GET http en remsa uif 200 OK IMAGEIGIF x E 8 0 77 sec 19774 bytes 515 ms MBIGET httpimrds sni ages en BereenShotRaealtimePreview jpg 200 OK IMAGE JPEG x 3 B 0 73sec 7069 bytes 500ms Ml GET htpdiprdaey s ii iBcreenShotwebAdminiPreview qif 200 OK IMAGE GIF 10 10 0 77 sec 9977 bytes 1 531 ms MB GET htipaprole s nit BereenShotFinalResultPreview qif 200 OK IMAGE GIF x 11 11 0 89 sec 35 bytes 218 ms GET http iwww proxy snier der J n b dir 200 OK IMAGEIGIF x 12 127 0 89 sec 26027 bytes 437 ms GET h http Jhwwwvw praxy sniffer delimages enfReportfebras git 200 OK IMAGE GIF x 313 0 84 sec B bytes Yate lt stil TET y x15 14 1 11 sec 35 bytes Total 114072 bytes O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 19 171 Apica ProxySniffer V5 0 User s Guide English Edition 4 2
3. PRX Main Menu Mozilla Firefox Sele File Edit wiew History Bookmarks Tools Help D PRX Main Menu T 8 127 0 0 1 7990 AA oou l PO AY Apica i SS Y a Y o h e P Main Menu uf P Web P P Proj Load T G Anal Refresh ct st at ProxySniffer Web Admin 5 0 K Help Claud Teal Bonnes Settings Nav aot I purge Load E De m text html Recorded Items 120 C H LIS 00 x er i Search Sessi Start St Reset O i pt i earc eszinn a o ese 3 v sec 135 NH 1 Recording State STOPPED Overall Cutter Recording seems Recording amp El pu Save Export War view d can d na X M I HO Page 4 Proxy Sniffer License and Price Information User s think time 0 seconds 35 Max acceptable response time ms a Item E Position Content Size Time HTTP Request HTTP Response x E 10735 bytes 14ms GET http i 16 4 Sipurchase en html 200 Ok TEXT HTML x29 N 15 583 bytes 1ms ii GET http 19 16 4 S Standardlicense gif 200 OK IMAGEIGIF x 36 No 20 583 bytes O ms ii GET http 19 16 4 5macloudLicenseEC2 gif 200 OK IMAGEJGIF x 37 IB 19 450 bytes 1ms IGET http 92 16 4 SAwsDevPayLicenseEC gif 200 OK IMAGE GIF xS I 16 751 bytes 1 ms IGET http 19 16 4 SfecelLoadTestHint gif 200 OK IMAGE GIF Total 0 02 sec 83132 bytes 5 Requests 4890 12 kbytes sec x 38 3H HE Page 5 Proxy Sniffer EC S
4. 1 _894bytes 31 ms MBIGET hit 192164 Sifaviconico 21 times before logout Total 3 62 sec 178 046 bytes 19 Requests 49 12 kbytes sec x 20 20 HO Page 2 Download user s think time 3 seconds 35 Item Test Offset Position Content Size Time HTTP Request HTTP Response During the load test inner loops execute within the outer normal loops repetitions of aa 00 ESE e LN the web surfing session per user for example if you run a load test with 10 users and 3 x2202 048sc SB bytes 31 ms o Mz 0A ual e 200 TEM x 23 23 048sec MW 959Bbytes 47 ms MGET http 192 16 4 S pdf bookmarks loops with an unlimited test duration each user will execute the recorded web surfing A IA A a v session 3 times Within each repetition outer loop the inner loop s will be executed x 25 25 050sec Ml 73 bytes 62ms MGETnttpir192 16 45 emallintoTransi x 26 26 Q 50sec MEA 27111 bytes 78ms MGET http 19216 4 5 tc recordingexte Total 0 58 sec 60 679 bytes 6 Requests 104 98 kbytes sec Inner loops must be composed of entire web pages and not only a subset of URL calls to A x 27 27 HO Page 3 Support user s think time 3 seconds 35 a single web page however you can define additional page breaks between URL calls iari Ter Onia Poco EEES Tinta SON HTTP FERM CITT RR after the recording has been completed x28 28 0 00sec MM 11 660 bytes 391 ms I GET http 1192 16 4 Sisupport en htra
5. welcome safearea com au x 7 7 236sec 6 502 bytes 1219 ms A GET http iwww safearea com au htmliportalicss isp 7m x 8 8 3 22 sec 8 293 bytes 1 235 ms 9 GET httpwww safearea com au htm themesi classi Welcome x 9 9 322sec 2666 bytes 891 ms 9 GET httpiiwww safearea com au htmljs liferay servic x10 10 542sec 244 bytes 750 ms MGET http iiwww safearea com au htmllthemes classi Safe Area Pty Ltd x 11 11 5 42sec 6 554 bytes 1 110 ms MIGET httpwww safearea com au htmlithemesi classi Fine software x12 12 542sec 177 bytes 782ms MBIGET httpwww safearea com au htmlithemesiclassi E x 13 13 5 45 sec 3 982 bytes 781 ms A GET http www safearea com au image company_lo ee ee OREA oe A E es i eer ee ER en TAS 2 Welcome Proxy Sniffer Google Sitemap Generator Services Blogs About uz x 14 14 5 45 sec 330 bytes 765 ms AN GET httpifwww safearea com au htmlithemesiclassi lt Ir Welcome Safe Area Pty Ltd We re the australian distributor of Proxy Sniffer one ofthe leading web load testing tools We also provide support and Services surrounding load testing af websapplication servers Announcements Entries Prepare a load test script for breakfast xt Internet 4g mios 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 10 171 Apica ProxySniffer V5 0 User s Guide English Edition 2 2 Recording Subsequent Web Pages You must insert a page break before the next web page
6. B http 127 0 0 1 7990 dfischerJwebadmininterFace AnalyseLoadtestResultDiagr amwWeblet La A hin ir arta Analyse Load Tests Load Curves CS O E X Help POF Report Export Close Page 1 Start Page URL 1 to 10 Response Time per URL Average Page 1 Start Page milliseconds 100 120 140 160 150 200 concurrent user Page 1 Start Page URL 11 to 17 Response Time per URL Average Page 1 Start Page milliseconds 100 120 140 160 160 200 concurrent user Page 2 Special Features URL 19 to 28 Response Time per URL Average Page 2 Special Features milliseconds 2012 2013 Ingenieurb ro David Fischer AG Switzerland GET http 192 16 4 5 80 0XXXX g GET hitp 192 16 4 5 80 flagEnglan 12 ms GET http 1 92 16 4 5 80 wolke2 jpe GET http 1192 16 4 5 B0 arrow_rec GET http 192 16 4 5 80 flagGerma 9 ms GET http 192 16 4 5 80 images_er GET http 192 16 4 5 80 images_er GET http 192 16 4 5 80 images_er Request GET Http 1 82 16 4 5 80 mages er GET http 1 92 16 45 807000000 gif GET Http 1 82 16 4 5 80 mages er 7 ms GET http 192 16 4 5 80 noPrograrr 6 ms GET http 192 16 4 5 80 bullet ok r GET http 192 16 4 5 80 C9C9C8 1 GET http 192 16 4 5 80 favicon icc All Rights Reserved Page 137 171 Apica ProxySniffer V5 0 User s Guide 10 3 4 Session Failures This menu option displays a summary about all errors which did occur in the test runs By clic
7. This restriction can be avoided by installing the local Job Controller as a Windows Service or as a Unix Daemon see Application Reference Manual After the cluster job has been scheduled you can leave this menu by closing the window and you can use later the Jobs menu to cancel or modify the schedule of this job 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 94 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 2 1 Real Time Cluster Job Statistics The real time statistics of a cluster job show the most important measured values similar to the values which are shown in the Real Time Statistic of Exec Agent Jobs see chapter 9 1 1 for a detailed description The cluster job itself contains Exec Agent jobs which have been created by the local cluster job controller By clicking on the magnifier icon of a cluster member the real time statistics of the corresponding Exec Agent job can be displayed in its own window PRX Cluster Job 208 Execute Load Test Mozilla Firefox Seles http 127 0 0 1 7990 dfischer webadmininterf ace PopupClusterStartFinalWeblet rox niffer NS SU Mer dise Cluster Job 208 Execute Load Test gt ES EL cl c1 Cluster Job 208 Real Ting Error Overview Diagrams Statistical Overview Diagrams Job Parameter Test01 u 200 4604 combined at Cluster Level 0 percurl 104 combined at Cluster Level Cluster Job Status running Test started at 01
8. Please note that the above list of menus is not exhaustive There are many other menus available for example menus to export data generate PDF reports control search delete or filter functionality and perform configuration of the Proxy Sniffer product itself In addition there are menus to enable and control the execution of load test programs on remote computer systems including the combination of load test execution systems configured ina cluster These menus are all described in this User Guide 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 7 171 Apica ProxySniffer V5 0 User s Guide All menus provide context specific help text available using the Help Icon Example http 127 0 0 1 7990 startDataRecordId 1 1 78658830671 PRX Delete URLs Windows Internet Explorer Delete URLs Delete URLs from Item 13 y toiterm gt C Delete all URLs with ContentType image jpeg AND HTTP Status Code 200 ok y C Host vevew proxy sniffer ch y AND Port 80 y C URL Pattern httpz www proxy sniffer ch images en SessionRecorderP gif Preview list of items to delete Item 14 Page 2 Proxy Sniffer Web Load and Stress Testing Tool Item 16 Page 3 Proxy Sniffer Produkt Features ere d Help Delete URLs gt Help Main Index a 0 00 see B 5 5 5 35 3 6368byes x 2 0s0sc N J 5609byes Input Fields 1 Delete URLs radio button from Item
9. Test Scenario Objectives Test Start Date 28 Sep 2009 14 24 34 Load Test Program Test01 class Load Source Hosts c1 Cluster vO3iid 10 224 200 9 534 users v03j74 10 224 200 22 533 users v03j7g 10 224 200 34 533 users Load Source OS E Target Host ef testix post ch 443 Applied HTTP Version 1 1 Test Input Parameter Concurrent Users 1 500 Planned Test Duration 15 00 min Planned Loops per User unlimited Startup Delay per User 300 millisec Request Timeout per URL 60 sec Statistic Sampling Interval 15 sec Additional Sampling Rate per Web Page Call 100 wer Additional Sampling Rate per URL Call 1 PRX Real Time Comments Mozilla Firefox a CERE 9 F http 127 0 0 1 7990 dfischer webadmininterface PopupAnalyseLoadtestDetailsWebletRtComments key c47ce4cff1 98bb8eade2d309aaSbee01 p lu mM Load Test Result Detail Modify Real Time Comments x a 6 6 User Input Fields o s 6 User s Think Time 15 oor 6 Note modifications are only temporarily applied the corresponding prxres file will not be updated gt y Annotation Real Time Comments A e Elapsed Time 0 1190 Seconds New Real Time Comment a 6 none ERI You can also modify delete or add real time comments as well Page 1 startseite as add or modify the annotat
10. button PRX Execute Load Test Mozilla Firefox BER B http 127 0 0 1 7990 dFischerwebadmininterf ace PopupDireckoryNavigatorStartLoadTestWeblet filePathB64 QzpcRG9rdWw1lIbnRIIHvuz CBFawS5zdGVvsbHvuz2VuxG V eee See Proxy Sniffer H lt S M Project Navigator Execute Load Test gt ae x Background information load test programs can also be transferred and executed on remote systems in the same manner as on the local system Start Load Test Test01 a therefore all data which are needed during program execution must be This load test program must first be zipped with other files before it can be executed Click on the gt gt ZIP and execute button to combine the compiled load test program together with the input file s and the plug in s to a ZIP archive packed to one Zi P arch ive New Zip Archive Test01 zip Load Test Program Test01 class Input Files Accounts bd Plug ins PrxAbortFailedTestVv10 class P and execute Note executable ZIP archives are can also be created manually using the ZIP icon amp inside Project Navigator You only have to create the ZIP archive once afterwards you can just start the zipped load test program Ifthe date of one of the files which has been added to the ZIP archive is newer than the date of If the load test program contains other dependent files which are not Input las BE LE u Files and not Plug Ins for example files which should be uploaded to t
11. e Real Time Comment If real time comments are entered during test execution these comments are later displayed inside all time based diagrams of the load test result detail menu e Job Parameter The name of the load test program and the program arguments test input parameter 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 81 171 Apica ProxySniffer V5 0 User s Guide English Edition The Web Transaction Rate Diagram shows the actual number of Successful completed URL calls per second counted overall simulated users By clicking on this diagram the Response Time Overview Diagrams are shown see chapter 9 1 1 1 e otal Passed URL Calls The total number of passed URL calls since the load test job was started e Total Failed URL Calls The total number of failed URL calls since the load test job was started e Keep Alive Efficiency The efficiency in percent about how often a network connection to the web server was a HM successfully re used instead of creating a new network connection This floating average value is calculated since otal Faile alls i HTTP Keep Alive Efficiency 95 5 the load test job was started AM Web Trans Rate URL callsfsect 194 7 e AV Web Trans Rate URL calls sec The floating average number of successful completed URL calls per second calculated since the load test job was started The Session Failures Ignored Errors Diagram shows
12. uuu mud PRX Page Scanner Result URL Details Mozilla Firefox uuuEEEEN NE http 11127 0 0 1 7990jdfischer webacmininterface PopupPageScannerResultUriDetailweblet displayPagehir 18displayUrlir D amp absoluteUrlB64 aHROcDovL d dySwent Ra Page Scanner Result URL Details x Page 1 http www proxy sniffer ch Proxy Sniffer Web Lasttest und Stresstes 141 ms 42 392 bytes i i ee e i http www proxy sniffer ch 15ms 4 626 bytes format css 16 ms 43 bytes XK gif 15 ms 234 bytes flagGerman gif 3 x 3 3 R 3 x 3 x x x x 12 1 1 1 P 1 1 1 2 O ms 1 212 bytes flagEngland gif 15ms 88 bytes 15 ms 9 791 bytes farrow red 12x9 gif fimages en SessionRecorderP dif 16ms 8706 bytes 16 ms 7 141 bytes images en varHandlerP qif images en ExecAgentClusterP dif 15ms 10 904 bytes 0 ms 35 bytes fimages en imeasurementResultP gif 000000 cif 47ms 31 474 bytes oms ms 67 bytes images enfReportZebra gif en ReportZebra qif bullet ok redai o ok red qif Toa without external links 311 ms 118213 bytes 13 URLs 3 16 2000k X 317 20 X 318 00k Page 2 http iwww proxy sniffer ch index_en html 20 2000k 94 ms 41 597 bytes http veww proxy sniffer ch index en htrml m without external links 94ms 41 597 bytes 1 URL Page 23 http www proxy sniffer ch featu
13. Vat Header Rev Content Rev Total mz O mz 44 mz O mz 5 mz 49 ms Muser data B chunk protocol time milliseconds Sample No 2 Met Con Reg Trim ait Header Rev Content Rew Total SH user data chunk protocol time milliseconds Sample Ho 3 All Rights Reserved English Edition Throughput measurement of a HTML response received from a web portal server It is conspicuous that the most response time is spent in the chunked protocol overhead but that the user data payload is received ina relatively short time One explanation could be that the Web page is calculated piece by piece by the portal server page navigation page main content page footer and that some server internal delay times occurred during the calculations Page 116 171 Apica ProxySniffer V5 0 User s Guide 10 1 4 Results per URL Call Details Displays measurement details about URL calls If this menu is invoked from the URL overview menu by clicking on a test number the selected URL call is marked with a blue background English Edition 0 Page StartPage user s think time 0 0 seconds C x Test gt Av Net t Con gt Av Req Trm gt Av Wait 23 Av Header Rev Ay Content Rev Min Time Av Time Max Time A Av Throughput v olumns 1 46 ms 0 ms 361
14. coocccccccccccccncccnoconoconocononononononononanonanonanonanenanenanes 15 3 1 2 4 HTTPS Client Certificate Authentication PKCS 11 Device Proxy Recorder esses nennen nennen nnn nnne nnns 15 3 1 2 5 NEN Althenticaton Proxy REI EE ee 16 3 1 2 6 KEerbErOS AUTHENICAUON SEPFOXY Hecordel ae 16 91 257 FOR ERNO ee Qmm 17 A Next Steps aiter Recording a Web Sung Session arsiasnee e o 18 4 1 Saving the Recorded Web Surfing Session ccccccccceeccesecescseecceeeccceecscseeesceecccuaeesseeescuncecenesscunescuncesceeescesessausescsaeeceneescuaesceaeeccsessousessenssosess 18 4 2 Reviewing the Recorded Web Surfing Session ccccccccsecescseccceneecceeesceececoececceeensecesouaecceneescenesouaeccoeeescenessaueecceeeeceneseceeeccseeesoeessceeeesseeeseeaes 19 4 2 1 Reviewing the Stressed Web Servers 4 abet ek de di a ea Bela 19 4 2 2 Reviewing the Automatically Applied Content TeSt cccccssscccsseccceeeecencecseeecenseseaeeesecessaeeeseusessaeeeseusessueeesuseseaseesueeessusessueeeseusessueeeseas 20 4 2 3 Configuring Parallel or Serial URL Execution with Web Pages sesssssssssessseseeneenen nennen nnne menn nnne nnn n sna nnn snas n snas nnns sns n a anrea ns 23 43 Executing a Fist Load TeSt uiti ae e eee 26 MES AAA ET 30 5 1 Importing Web Surfing Sessions from External Definition Files occconcnconncccnnconnnnonnnnncnnnnnononnnoncnonnnonnrnconrnnnnnnnnrnnnnnnnnnnnnn
15. id 449647 48 gt 51 Form Param VIEWSTATE 1 _ VIEWSTATE dDwtMTgOMjMwNDc403Q802w8aTwxPjs O2w8dDw7bDxpPDE O2k8NT47PjtsPHQ8O A 51 Form Param Status1 ins_step1 type rbLeben a 51 Form Param Q Status1 ins_step1 subtype rbRate amp 52 CGlParam type tt22 amp 52 CGlParam W changed False 52 gt 54 Form Param VIEWSTATE 2 _ vIEWSTATE dDwtMTgOMjMwNDc4O03Q802w8aTwxPjs O2w8dDw7bDxpPDE O2k8NT47PjtsPHQ8O R 54 Form Param Status1 ins_step22 b Company Scandia R 54 Form Param Status1 ins_step22 t4tPolicyNumber 223er4 R A A a 55 A All Rights Reserved Page 56 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 5 2 Manual Extraction of Dynamically Exchanged Session Parameters The documentation in this subchapter 7 5 2 is still applicable but not more up to date Starting from Proxy Sniffer Version 4 4 G a new function named Var Extractor Wizard had been added to the product Further information is provided in the new manual about Handling of Dynamically Exchanged Session Parameters PDF document HandlingDynamicSessionParameterEN pdf If the automated handling did not succeed you should click on the magnifier icon at the left of the desired parameter All of the URLs in the web session are searched looking for the recorded value A 3 Cal Param E id a 451047 A 1 CG Param 9 id 449647 Blue arrows pointing to the left indicate that the value ht
16. to the user s think time of a web page chapter 7 3 1 to the response verification algorithm of a URL call searched text fragment or size of received content chapter 4 2 2 to the number of iterations and or the pacing delay of an inner loop chapter O to some HTTP request header fields most request header fields are automatically handled by Proxy Sniffer to an input parameter of a Load Test Plug In chapter 7 4 Each variable has also a scope Possible scopes are global all users will see the same value of the variable during the load test user although the variable has been defined only once each user will see its own value during the load test There are as many virtual instances of the variable as there are concurrent users used during the load test loop the variable is bound to the current loop surf session repetition of a user and its value can change during each loop inner loop the variable is bound to an inner loop of a user and can change its value during each iteration of the inner loop Although seemingly complicated the Var Handler is a powerful tool which is easy to use It is possible to satisfy complex requirements in a short period of time with a few mouse clicks as described in the next sections Programming knowledge is not required 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 40 171 Apica ProxySniffer V5 0 User s Guide 7 1 Variable Handler Var H
17. 15 Modifying Woda Test Programs MANU ERE OS ooo 165 16 IRGC esoo aio as UE 0 HD circ REN A E dc Settee Otte dc Settee O haan etd ca tet heen ante ca ctu beast Sante en cae haunt Sete en 167 16 1 example 1 Exracing Rertormance Dale a a 167 162 Example 2 Extracting Error Smapshols uo uo o aee e Idee Loc a 169 17 Manalac ENE cedem D no Un EE 171 UNIX is a trademark of The Open Group Solaris and Java are trademarks of Sun Microsystems Inc Windows is a trademark of Microsoft Corporation 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 5 171 Apica ProxySniffer V5 0 User s Guide 1 Introduction English Edition Thank you for choosing Proxy Sniffer You now have a powerful product to perform professional web load tests The product is easy to use and intuitive However for a better understanding of the concepts behind Proxy Sniffer we suggest that you read this manual 1 1 Menu and Navigation Overview Session Cutter Analyse Load Test Menu A rte AG AER M ui Load Curve Diagrams Execute Load Test lt x Comparison Diagrams S ETE Recorded Web Surfing Session Generate Load Test Program Detail Statistics amp Diagrams Response Verification The Proxy Sniffer menu structure is somewhat different from other applications Menu options are always context sensitive
18. Abort Loop gt gt Next Loop v _ Applyforall URLs with HTTP status code 200 OK v AND contentiype textihtmi J Reset Settings for this URL O forall URLs with HTTP status code any v AND contentiype any Done The search text must occur exactly lt int gt times inside the received content lt int gt lt search text gt The search text must occur a minimum of lt int gt times inside the received content lt int1 gt lt int2 gt lt search text gt Disable Content Verification the received content will not be checked Normally this option should not be used Verify Content by Size only the size of the content is checked This is a good fast approach for completely static content such as images You may also set an acceptable size derivation of 0 if the content never changes Verify Content by Test String a test string is searched for inside the received content This is the best method for testing dynamically generated HTML pages If the content contains HTML or XML text the Proxy Sniffer program analyses the recorded content and gives rated suggestions 0 1 for advisable string fragments Alternatively you can enter your own desired test string In addition to searching for the occurrence of a simple text string inside the received response content of an URL call the following special search patterns are supported I lt search text gt The search tex
19. E http 127 0 0 1 7990 dfischer webadmininterface PopupPersonalSettingsWeblet m authentication are directly requested by the web browser during recoding of a web surfing session This means that no special configuration is required for these two authentication methods inside this eee See Proxy Sniffer T lt 7 26 web Admin Personal Settings Proxy Recorder amp GUI Help Web Refresh Close Next Proxy HTTP Host m Web Browser Protocol NTLM v2 v Domain menu Next Proxy HTTP Port Q ga N Username Password j Next Proxy HTTP Cache disabled gt ann uu l l l Next Proxy HTTPS Host C Note 2 the authentication credentials entered in this ME res eS DE Es E BEN menu can also be transferred into the generated load ext Pro sername Web Server Host Names Full DNS Names Additional IP Addresses and Aliases a O Bu D um ML EANA Ly sm testpdv testpdv lan 192 16 4 23 test prog rams The al location of individual m credentials per simulated user can be selected No Next Proxy for Host Domain Add Web Server Host Name Full DNS Name D _ BEEN P Used only for Basic Authentication For Kerberos Authentication add Proxy to Kerberos Host Names Kerberos Configuration krb5 conf Additional Settings krb5LoginModule conf when generati ng the HTTP S Load Test _ Programs see chapter 8 GUI Settings g p SSL Version v3 Time Zone
20. To Email Addresses direct d fischer com Outbound HTTP Proxy Host Port 192 16 4 31 8080 O If the Session Failure Rate IS greater than a Cc Email Addresses Outbound HTTP Proxy Auth Username miller threshold in percent 1 Bee Email Addresses Outbound HTTP Proxy Auth Password Alert Conditions Send Notifications from Prefix Additional Message Pre O If the Average Hesponse Time per Page IS Ifa Job connate Fr prx fatal STARTJOB failed greater than a threshold In seconds 1 V Atthe Start of a Job prx finfo STARTJOB ok If an Internal Error occurs during the Execution of a Job prx fatal JOB internal error O If the Average Response Time of the Slowest During the Execution of a Job in Periodically Intervals of Sminutes C At Every Interval prx into JOB runtime info Page IS greater than a threshold In seconds Ifthe Session Failure Rate is greater than 2 prx error v SFR at runtime exceeded prx cancelled SFR at runtime ok 1 Ifthe Average Response Time per Page is greater than Sseconds v i prx exceeded Y RTP at runtime exceeded prx cancelled RTP at runtime ok Ifthe Average Response Time ofthe Slowest Page is greater than 10 seconds v prx exceeded RTS at runtime exceeded prx cancelled RTS at runtime ok e At the End of a Job information C Atthe End of a Job prx finfo ENDJOB info Atthe End of a Job ifthe Session Failure Rate is greater than 296 Y prx error v END
21. configuration can be modified by clicking on the IT TEST Em corresponding magnifier icon In the lower part of the A 3 B TestPC Il 1 00 19216 4 35 7993 plain direct network connection window additional Exec Agents can be defined A 36 B Sun Fire V240 1 00 192 16 4 78 7993 plain direct network connection TS ge and or existing Exec Agents can be modified You must click on the Refresh icon in the right upper corner of the windows to add several Exec Agents Input Fields Description arbitrary description of the Exec A Agent Add New Exec Agent ic mica EO Host TCP IP address or host name of the Description E c d L L Exec Agent Host Port 7993 Protocol plain y BER LL LES Ne Username Password Load Test zw 13 sss SZ Port TCP IP port number usually 7993 18 Indirect Network Connection through HTTP S Proxy gt Y _ i i ewe ar Po e n D ON AA Protocol communication protocol Proxy Username Proxy Password Y Y i r niente NE Username Password allows you to restrict Add New Exec Agent ee Web Application d access to the Exec Agent by using a username and a password This option can EE 7 only be used with the HTTP or HTTPS communication protocols optional only for http and https protocol supported All further input fields are only used if the communication should be made over an outgoing proxy server You can test the configuration and the acc
22. lt HEAD gt BODY BGCOLOR 000000 TEXT 000000 LINK 0000FF VLINK 0000F F l outer table start gt TABLE BORDER 0 CELLSPACING 0 CELLPADDING 0 gt l text title LL 15 STR x 319 2000k 31ms 25688 bytes e Broken Links displays a list of all broken hyperlinks Page 4 http www avantec ch hotlinksindex html AYVANTEC Hot Links 4 0 2000k 297 ms 16474 bytes httpztiveww avantec chrhotlinks index html 4 19 404 notfound 141 ms 284 bytes http mew remote exploit arg index php aSniffing remate traffic via GRE tunnels Page 15 httpzivevew avantec chinewsflash html AVAMTEC Mewsflashes 15 0 2000k 469 ms 37225 bytes httpwww avantec chinewsflash html 15 34 MUR 344 ms 298 bytes httpiveww orbitiex seminare ch pages sem mitwoch strmiti 1O 15 80 MAOA oms bytes http www telenetcom ch 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 159 171 Apica ProxySniffer V5 0 User s Guide Duplicated Content displays a list of URLs with duplicate content same content but different URL path Duplicated Content Original Duplicate Original Duplicate Original Duplicate Original Duplicate Largest Web Pages displays a list of the largest web pages Hint you can click on the bars to display the corresponding page details kbytes http www avantec ch checkup http haa avan
23. prosy eniffer proxy sniffer proxy sniffer proxy snif fer Prosy sniffer proxy sniffer proxy snif fer Prosy sniffer Pproxy sniffer 4prauy sniffer prosy aeniffer 4prauy sniffer 4prouy sniffer prosy eniffer 2 5 3 4 Hb 14 pase nr http um Pro y sniffer ch Bi May 2667 21 58 83 ch featurez de html ch featurez html che ch support de Hhtml ch en Enf ree ScreenshotlebAdmin html cheen javadocse jndex al l Atml ch customersAndPartners_de html chens javadocse dFf ischer ut ils HttpTe chens javadocsbe df ischer ut ils Perf or che index_en html chrseruj ces_de html ch aboutus_de html chedown load_de html ch buu de html ch support htmnl All Rights Reserved English Edition Hint you can click on the bars to display the corresponding page details Page 161 171 Apica ProxySniffer V5 0 User s Guide English Edition 13 1 3 Converting a Scan Result into a Web Surfing Session A Page Scanner result can be converted into a normal web surfing session which can be used to create a load test program PRX Page Scanner Result Mozilla Firefox PUO httpi 127 0 0 1 7990 dfischer webadmininterf ace PopupPageScannerResultWeblet Proxy Sniffer Web Admin Page Scanner Result Help gage EEE Refresh Close Scan Input Parameter gt Scan Statistic gt Non Processed Web Servers P Sez Search ASCII Textil E http
24. 1 2 t Ten SEN CE UV EP ee EP Ee TENE 1 0 0 0 6 0 4 0 2 0 0 48 19 15 12 33 26 46 40 45 49 52 5 36 37 url 5258036 1318 av 0 2 E 2 E 0 00 00 0 00 40 0 01 20 0 02 00 0 02 40 0 03 20 10 20 30 340 50 60 70 60 90 100 Load Test Test 1 27 Nov 2008 23 19 49 User 20 Load Test Test 1 27 Nov 2008 23 19 49 User 20 Load Test Test 1 27 Nov 2008 23 19 49 User 20 fns 19 526 ms Page 2 Special Features contains 20 URLs 19 38 Hint Click inside the diagram on the Il bars to display details Save image to disk 19 1 578 ms Most Time Consuming URL of Page GET http www proxy sniffer com 80 features html URL Response Time 188 Sampling Rate onse Time Percentile 188 Sampling Rate C19 features html C19 features html milliseconds Average 90 Percentile O Most Time Consuming URL per Page Response Time per Media Type Response Time per Media Type 0 E E 5 E TIRE 0 00 00 0 00 40 0 01 20 0 02 00 0 02 40 0 03 20 10 20 30 40 50 60 70 0xX 90 100 Load Test Test 1 27 Nov 2008 23 19 49 User 20 Load Test Test 1 27 Nov 2008 23 19 49 User 20 139 4476 ms Page 3 Download contains 4 URLs 40 43 application pdf 165 590 3978717699 E al 1589ms Most Time Consuming URL of Page GET http www topshareware com 80amages logo topshareware 883 1 gif image gif 7 848 2 797 041 test html 1 455 183 633 image jpeg 358 86 223 application x javascript
25. Hour Minute Schedule Job Max Error Snapshots 20MB memory v Statistic Sampling Interval 15 segts Additional Sampling Rate per Page Call 100 Additional Sampling Rate per URL Call ya v Add recommended Debug Options none recommended v Additional Options SSL v2N3ITLS vw ation First Teg Mun gt gt gt Continue trecommended will be displayed as hint in Project Navigator 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 27 171 Apica ProxySniffer V5 0 User s Guide English Edition PRX Job 1332 Mozilla Firefox Sel PRX Job 1332 Mozilla Firefox FO o http 127 0 0 1 7990 dfischer webadmininterface PopupDirectoryNavigatorDisplayLoadTestWeblet filePathB64 QzpcRG9rdw tIbnRIIHVuzCBFaWSzdc E http 127 0 0 1 7990 dfischerfwebadmininterface PopupDirectoryMavigatorDisplayLoadTestweblet filePathB64 02pcRG9rdWlbnRIIHYuZCBFaW5zdc 1 oe oe rox nitter 1 i SS pis rox niffer Web Admin Project Navigator Execute Load Test SHE MM Project Navigator Execute Load Test E eX Web Admin Help Jobs Refresh Close Local Exec Agent Job 1332 Real Time Comment Local Exec Agent Job 1332 Real Time Comment Job Parameter TestO1 u 10 d 240 t 60 sdelay 200 maxloops O sampling 15 percpage 100 percurl 20 paxerrmem 20 nolog annotation First Test Run Job Parameter Test01 u 10 d 240 t 60
26. Lg Produktion Fa 3NodeCluster Content Size Time HTTP Request x 16 0 00 sec nl 45 069 bytes 408 ms B GET http Ifveww proxy ES O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 163 171 Apica ProxySniffer V5 0 User s Guide English Edition 14 Web Tools The Web tools menu can be invoked from the main menu and contains four small utilities which can be useful to examine the data exchanged between the web browser and the web server Tools Utilities Base64 Text Transformation performs a base64 transformation or its inverse operation as appropriate The base64 algorithm is often used to obfuscate the values of CGI parameters The inverse a a m t e Por han operation allows you to decode such obfuscated values A AAA IG Escape Unescape URL CGI Text Value performs a URL encoding transformation or its inverse operation as appropriate This algorithm PESEE EENE is often used to mask special characters within the values of CGI E EREETRE parameters and is also used when HTML form parameters are go 4 ex transmitted to the web server P Webadmin Web Tools Hep N HE Refresh Close Examine SSL Configuration of HTTPS Server Encryption et Protocols and Algorithms examines the SSL configuration of an Baseb4 Text VHuhbnNimb3utIHRoaXMgvGV4da HTTPS web server from outside and displays hints about SSL amm
27. Resets the user related SSL session cache per loop default no reset per loop ssicmode Applies SSL https compatibility workarounds for buggy SSL servers You may try this option if you consistently receive the error message Network Connection aborted by Server for all https calls when executing the load test tz lt timezone gt Allows you to set another time zone to be used during the load test For a list of supported time zones see the Application Reference Manual Chapter 6 Xbootclasspath a lt path gt Specify for the load test job a path of JAR archives and ZIP archives to append to the default bootstrap class path Xbootclasspath p lt path gt Specify for the load test job a path of JAR archives and ZIP archives to prepend in front of the default bootstrap class path SSL specifies which HT TPS SSL protocol version should be used O O O O v2 v3 TLS allows you to detect the best SSL protocol version automatically The TLS protocol is preferred however if it is not supported by the web server SSLv3 is used or if SSLv3 is not supported then SSLv2 is used This is standard behavior implemented by many commercial web browser products TLS sets the SSL protocol version to TLS v3 sets the SSL protocol version to SSLv3 v2 sets the SSL protocol version to SSLv2 not recommended out dated SSL protocol version Annotation here you should enter a short comment about the test run such as purpose
28. Setup and Report Branding Mozilla Firefox http 127 0 0 1 7990 dfischer webadmininterf ace PopupDirectoryNavigator JavaSetupWeblet eee Proxy Sniffer gt 26 T web kde Project Navigator Java Setup and Report Branding ces Java Compiler Java Compiler Invocation Path CADokumente und EinstellungenimutongiProxySniffenjreibinijavac CLASSPATH CADokumente und EinstellungenimutongiProxySniffenprsniff jar C ADokumente und Einstellungenimul Compiler Options Java Interpreter used by local Exec Agent Java Interpreter Invocation Path CADokumente und Einstellungenimutong ProxySnifferijreibin jaya CLASSPATH 4CADokumente und EinstellungenimutongiProxySniffenprsniff jar CADokumente und Einstellungenimul Kbootclasspathip CADokumente und EinstellungenunutongYProxySniffenprxsniff jar C Dokumente und Einstellungenimutc Xmx Megabytes Interpreter Options Local Exec Agent IP Addresses 192 16 4 35 192 16 4 37 192 16 4 38 Auto Detect Y Increase this value if you get a java lang OutOfMemonr Exception when executing a load test Default leave blank Enter only values if the local Exec Agent can use multiple own client IP addresses when executing a load test Format ip addr 17 ip addr 27 ip addr 3 gt Start in such a case the load test program with the additional option multihomed Sour
29. Test Passed Failed AVTime lt 90 AV Size URL 22 4 657 3 1 391 ms 5330 ms 9 654 bytes POST https ef testix post ch 443 efisecureshtml login 2a 4 657 0 30 ms 28 ms 31 058 bytes GET https Neftestixpost ch 443 ef publiciccipicsfidpcd_anleitung_pk_de cif Total 4 657 3 1 921 ms 5 364 ms 40 712 bytes 2 URLs lt Done English Edition At the right upper corner inside the title of the window is the Report icon which allows you to generate a PDF report General data about the test run are shown in a yellow bar Further general data are described below Advanced Test Parameter An extract of the most important test input parameters Measured Results per Single User per Loop AV Session Time per Loop average time of a single loop per user repetition of a web surfing session AV Response Time per Page average response time per web page calculated over all web pages and users Network Throughput per User average network throughput per user Overall Test Results Web Transaction Rate number of successfully executed URL calls per second hits per second measured over all users aS an average over the entire test duration This value reflects the throughput of the web server Session Failure Rate percentage of failed loops error rate measured over all users By clicking on the percentage value if not zero the error snapshots can be displayed see Chapter
30. accurately measured Total Failed URL Calle 243 which means that they do not depend on the value chosen for the Additional Sampling Rate per Average Size Req Resp 31797 bytes URL Call lu a BIS UE TS e Total Failed URL Calls total number of failed calls for this URL TCP Socket Connect Time 100 68 ms e Average Size Req Resp the average size of the transmitted received data per Request Transmit Time O mz Average URL call Response Header Mar Time F23 mz Response Time i Response Header Receive Time 0 ms 814 ma e Max Response Time the maximum response time ever measured Response Content Receive Time 3 ms e Min Response Time the minimum response time ever measured e Av TCP Socket Connect Time the average time to open a new network connection to the web server measured for this URL instead of a value means that never a new network connection was opened for this URL because HTTP Keep Alive re using of cached network connections was always successful The additional percentage value shown in brackets at the left hand displays the percentage about how often a new network connection was opened to the web server in comparison to how often this was not necessary This percentage value is also called the reverse keep alive efficiency e Av Request Transmit Time the average time to transmit the HTTP request header optionally the HTTP request content data form data or file upload data to the web server me
31. core Servletirapper doService ServletWrapper java 405 META HTTP EQUIV CONTENT TYPE CONTENT text html charsetziso 8858 1 lt TITLE gt Proxy Sniffer Project Master Directory Browser lt TITLE gt H 0h O 2012 2013 Ingenieurb ro David Fischer AG Switzerland HTTP Response Content 19803 Bytes search Display in Web Browser Download Content HTML gt HEAD gt at org apache tomcat core Handler service Handler java 287 at org apache tomcat core Servletirapper service ServletWrapper java 372 at org apache tomcat core ContextManager internalService ContextManager java 812 at org apache tomcat core ContextManager service ContextManager java 758 at org apache tomcat service http HttpConnectionHandler processConnection HttpConnectionHanc at org apache tomcat service TcpWorkerThread runIt PoolTcpEndpoint java 416 at org apache tomcat util ThreadPool ControlRunnable run ThreadPool java 501 at java lang Thread run Thread java 479 Done All Rights Reserved Page 131 171 Apica ProxySniffer V5 0 User s Guide English Edition Shown next is the debug output of the current loop current web surfing session of simulated user This also contains information about extracted and assigned session variables based on the Var Handler definitions TOO0000 Page 4 EmpReview T000000 TOOOOOO TO0O0000 155 POST http com 50100 ir
32. files of a completed load test job Take into consideration that you must first acquire the statistic result file prxres of the job before you delete all files of a job otherwise the result data of the job are lost Displays the date and time when the job has been defined or when the job has been completed or for scheduled jobs the planned point in time when the job will be started Displays the current job state configured ready to run scheduled running or completed The state 22 means that the job data are corrupted you should delete all jobs which have the state because they delay the display of all jobs in this list Load Test Program amp Arguments Displays the name of the load test program and the arguments of the load test program see next subchapter Released from GUI IP Displays the TCP IP address remote computer from which the job has been initiated 9 3 1 Load Test Program Arguments Argument Parameter Meaning u number Number of concurrent users d seconds Planned test duration in seconds 0 unlimited t seconds Request timeout per URL call in seconds sdelay lt milliseconds gt Startup delay between creating concurrent users in milliseconds maxloops lt number gt Max number of loops repetitions of web surfing session per user 0 unlimited downlink Kbps Network bandwidth limitation per concurrent user in kilobits per second for the downlink web server to w
33. load test program The Project Navigator menu will then be displayed PRX Project Navigator Mozilla Firefox Ex E Attp 127 0 0 1 7990 dfischer webadmininterface PopupDirectoryNavigatorWeblet selectDir OzpcR G9rdWw tIbnRIIHVuz CBFawSzdGVsbHVvuz2VuxG1 1dG9uz 1x Qcm94eVNuaWZmzX Jc TXIUZXNOcw Q 7 The newly created and compiled load test program is marked with a dark blue background and can now be P proxy sniffer Project Navigator S A 2 he x started by clicking on the amp icon C Dokumente und Einstellungen mutong ProxySniffer MyTests H C3 57 eu Program Test01 class created Plugins ReportTemplates ScriptExamples imi Trash 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 73 171 Apica ProxySniffer V5 0 User s Guide English Edition 8 1 Load Test Programs with Dependent Files Executable load test programs class files which use dependent files such as Input Files or Plug Ins must first be zipped together with the dependent files into a single ZIP archive Thereafter the ZIP archive itself must be started as the load test program The GUI checks every time when a simple load test program class file is started to see if Input Files or Plug Ins are needed If so you will receive an appropriate information message with the hint that you must build a ZIP archive for the load test This can easily be done by just clicking on the gt gt ZIP and execute
34. one that is relevant to your projects It is also often useful for individual application releases or even daily test programs to be assigned their own sub directories MyTests Plugins Project A Project B L EnduranceTest Project C Projeg To create a new sub directory select an existing directory at left and then click on the Create Directory icon Note new directories can also be created via the Operating E ProjectPlugins Release1 Plugins Release 1 1 oystem for example via File Explorer under Windows or by using a console The Project Navigator menu has been designed to ensure that no discrepancies exist between the menu and the Fallback Ez Operating System view Login 2 ServerThroughput 2Server 4Server 6Server 8ServerServer Loadcurve J Release2 L JUN_01_Degradation 700VerificationsSL 10000 5 5 5 a 4 5 5 5 gt 4 5 5 3 5 5 y After the creation of a new sub directory an existing load test Project Navigator 2 5 2 h Xx program including its recorded web surfing session and Input Files can be copied by marking the corresponding checkboxes and then clicking on the Copy Selected Files icon The new sub directory can then be selected with a single click at the left side in the Project Navigator Proxy Sniffer Web Admin s Aes vat nae EnProxySnifferlMyTests Project_DiPerformanceTuningDay 3 MTests Plugins Project_A Project_B L EnduranceTest
35. or TL S Timeout of SSL cache in seconds O cache disabled Time zone see Application Reference Manual Comment about the test run 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 100 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 4 Scripting Load Test Jobs Several load test jobs can be started from the GUI at the same time However the GUI does not have the ability to automatically run sequences of load test jobs synchronize load test jobs or automatically start several jobs with a single mouse click To perform these kinds of activities you must program load test job scripts which are written in the natural scripting language of your operating system Windows bat files Unix sh ksh csh files Inside these scripts the PrxJob utility is used as the interface to the Proxy Sniffer system When the Windows version of Proxy Sniffer is installed the installation kit creates the directory ScriptExamples within the Project Navigator and this directory contains some example scripts The PrxJob utility allows you to start load test jobs on the local as well as on a remote system It also provides the capability to create cluster jobs to synchronize jobs to obtain the current state of jobs and to acquire the statistics result files of jobs More information about the PrxJob utility can be found in the Application Reference Manual Chapter 4 9 5 Rerun of Load Te
36. s Guide English Edition 7 3 1 More Hints for using User Input Fields User Input Fields are also often used to vary the user s think time Another example would be the setting of the booking date for financial transactions Note if you start a load test job optionally from a script see Application Reference Manual you must pass the User Input Field as an additional argument to the load test program The name of the program argument is the name of the variable which was created when the User Input Field was defined for example for a variable named hostname the corresponding argument specification would be java PrxJob transmitClusterJob Cluster 1 Test01 u 100 d 300 t 60 nolog hostname testsys ggjhkjg com 7 3 1 1 Example Adjustable User s Think Time The following example shows how the user s think time of the page breaks web pages can be dynamically set every time when starting a load test 1 Create a User Input Field and set a default value in this case in seconds 2 Inthe main menu assign the variable of the User Input Field to the user s think time of the first page break by using the option Apply new user s think time values for all page breaks 2 n 3 After that you can freely choose the user s think time of the web pages every time when starting the load test The value of the User Input Field is also shown in the load test result detail menu test scenario Add User Input Field EU al War Name
37. that is only options relevant to the current operation are displayed Also there is no main menu or main application window even though one of the menus has the title Main Menu That said there are however three important menus The Main Menu enables recording of web surfing sessions with any web browser as well as the editing of web surfing sessions and applying functional enhancements The sub menu Generate Load Test Program converts a recorded web surfing session into a ready to run load test program The Project Navigator allows the management of stored web surfing sessions and load test programs Furthermore load test programs can be started from this menu and the corresponding test results and measurements are then also available from this menu The Analyse Load Test Menu allows the analysis of load test results and measurements including comparisons of the results of different load test runs Of the three central menus described above only the Project Navigator deals with permanent data that is data persisted to disk The other two menus as well as most of all other menus work only with transient data stored in memory The other Proxy Sniffer menus shown in the figure above are described below 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 6 171 Apica ProxySniffer V5 0 User s Guide English Edition Page Scanner allows the automatic scanning of entire web
38. url x error y 1 write response content to file no binary data are written if testURL isAsciiContent FileOutputStream fout new FileOutputStream fileStartName response content html fout write testURL getDecompressedContent fout close 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 169 171 Apica ProxySniffer V5 0 User s Guide English Edition break case PerformanceDataRecord TYPE PAGE BREAK break default break catch Exception e e printStackTrace 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 170 171 Apica ProxySniffer V5 0 User s Guide English Edition 17 Manufacturer Ingenieurb ro David Fischer AG Switzerland A company of the Apica Group Note All menus provide context specific help text Product Web Site http www proxy sniffer com available using the Help Icon eee po Project Navigator Execute Load Test Web Admin Execute Load Test Job LOAD_20070612 re Execute Test from Cluster cluster1 y Number of Concurrent Users 1 y Load Test Duration 1 min y e a Max Loops per User unlimited Startup Delay per User 20 z Milliseconds Max Network Bandwidth per User unlimited y Downlink unlimited y Uplink Request Timeout per URL 60 Seconds Max Error Snapshots per URL 30 Y Statistic Sampling Interval 5 y Seconds Addit
39. 10 2 Total Network Throughput average network throughput during the test run If you have loaded several test results you can use the arrows in the Test Result selection box to switch between them Further details of the test run can be accessed by clicking on a title within the red framed range These measurement details are briefly explained in the following subchapters 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 109 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 1 Test Scenario Displays the test environment test inout parameters and a summary of the executed web surfing session Matistic Sampling Interval 15 sec Network Throughput per User 2 3 kBytes sec Total Network Throughput 19 66 MBit sec Total Transmitted 2 7 Diagram Response Time per Page Results per URL Call Overview Results per URL Call Details Diagram Response Time Percentiles gt Diagram Top Time Consuming URLs gt Diagram Concurrent Users Diagram Session Time Diagram Web Transaction Rate Diagram Users Waiting for Response gt Diagram Completed Loops Diagram TCP Socket Connect Time Diagram Network Throughput gt Diagram HTTP Keep Alive Efficiency gt Diagram SSL Cache Efficiency Diagram Session Failures Diagram Error Types Diagram Number of Errors per Page Diagram Number of Errors per URL gt Diagram External Measured Data
40. 127 0 0 1 7990 Proxy Sniffer Convert Page Scanner Result to Web Session Windows Internet Explorer 2 l Input Fields wasnt Project Navigator Convert Page Scanner Result to Web Session 5 5 e Filename file name of the web surfing session You WS a must enter a simple filename with no path and no file ro mnam R extension The file extension is always prxdat The Wines Bae a file will be saved in the selected Project Navigator Max URL Calls 500 y directory a Annotation e Web Pages allows you to select the scanned web Lesa session into Mem Monu _ I pages which should flow into the web surfing session Fo FE Thefleoxencion i anaye prat Example Tetsesten D All Pages means that all scanned web pages are selected Alternatively the option Page Ranges allows Saved you to select one or several ranges of page numbers If zz you use several ranges they must be separated by commas MET x Example 1 3 5 7 38 81 pe OM a e Max URL Calls limits the number of URL calls which Mg Cluster should flow into the web surfing session ma Hint it is recommended that you do not convert more than 1000 URL calls into a web surfing session Hint converted sessions can be loaded by clicking on the Load Session icons inside Project Navigator e Annotation we recommend that you enter a short comment about the web surfing session e Load Session into also loads the web surfing
41. 138 37 167 text css ea 1 751 URL Response Time Percentile 188 Sampling Rate C431 topshareware_8831 gif URL Response Time 188 Sampling Rate 431 topshareware_8831 gif milliseconds milliseconds Load Test Test 1 26 May 2611 14 33 35 User 38 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 119 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 7 Diagram Concurrent Users Shows the number of users during the test run The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Concurrent Users Sampling Interval 15 Seconds humber of concurrent users time 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 50 0 20 00 0 22 30 0 25 00 Load Test Test 1 28 Sep 2669 14 24 34 User 1666 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 120 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 8 Diagram Session Time Shows the response time per successfully executed loop repetition of a web surfing session during the test run The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Session Tine per User per Loop The accumulated user s think time of the loop is shown by a red line Sampling Interval 15 Seconds session time in seconds for successful completed sessions thi
42. 159 62 trisec 0 00 22 19 MBitisec Sechster Test BATestoi 1B Jun 2006 15 27 04 200 T04min 4093tisec 100 00 ap Hint execute the same load test program several times with a different number of concurrent users and compare the measured results Click on the magnifier for detalis Ea Done ZB 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 134 171 Apica ProxySniffer V5 0 User s Guide 10 3 1 Overall Load Curves English Edition In the right upper corner inside the title of the window you can generate a PDF report and you can also export the performance data PRX Analyse Load Tests Load Curves Mozilla Firefox http 127 0 0 1 7990 dfischer webadmininterface AnalyseloadtestResultDiagramWweblet oe Analyse Load Tests Load Curves Hint Click inside the diagrams amp 2 PDF Report Export Close on the icons to display details Help Response Time Behaviour Average Session Time per User and Loop Average Session Time per User per Loop seconds 150 200 concurrent user Stability Session Failure Rate Session Failure Rate percentage of failed sessions failed outer loops Concurrent Outstanding Requests Concurrent Outstanding Requests average number of outstanding requests Server Performance Web Transaction Rate Web Transaction Rate Hits per Second successful completed url calls per second 150 200 concurrent user Net
43. 2013 Ingenieurb ro David Fischer AG Switzerland loops oops SS seconds seconds amp If the number of iterations of an inner loop is controlled by a variable the value of such a variable can also be 0 zero A value of zero means that a simulated user does not execute enter the inner loop This can be used in combination with an Input File see chapter 7 2 whose file scope is new line per user or new line per loop and whose lines contain values of zero and one which are assigned to the variables of the iterations that is some of the users skip parts of the recorded web surfing session during the load test However to get valid statistical data it is required that at least once during the load test at least one user executes the inner loop one or more times All Rights Reserved Page 36 171 Apica ProxySniffer V5 0 User s Guide Inner Loop 1 Executed by 25 of all Users English Edition Content of Input File Line InnerLoop1 Inner Loop 2 Iterator Iterator LIT Page an Start Page users Perk Bene Lila Trink Tere seconds 295 dvd Content Swe dere MITP Request HTTP Response 391204 723008 GET 17352 06 ons mem S ms Oo LIT Page 2 login masts users Dink bme UbiThinh Tene seconds 239 ion 1 1 Senato Fee E Content Sue m Tene MITP Request HTTP Merge o N i post HO Page 14 in raton uiers irk bre Lp Thich Tiere seconds 1058 Por Content Sue wm Doo 2 10 tentem Request HTTP Reu
44. 213 ms 720 bytes GET http www proxy sniffer com 80 C8C98C8 1x5 gif 7 48 981 ms 2 190 ms 1 566 bytes GET http www proxy sniffer com 80 favicon ico Total 48 W 16 023 ms 25215 ms 140 530 bytes 17 URLs TO Aaa nal Page 2 Speciallieatures user s think time 3 0 seconds Test E Passed Faile AV Time lt 90 AV Size URL 19 47 a 1 578 ms 3 279 ms 27 144 bytes GET https proxy sniffer com 80 features html 2n 47 a T74ms 2 288 ms 64 537 bytes GET httoAwww proxy sniffer com 80 images_enfSessionRecorder gif a PRX Analyse Load Test Errors Mozilla Firefox DAR F http 127 0 0 1 7990 dfischerfwebadmininterface PopupAnalyseLoadtestErrorWeblet key 888daff2512575f5beb3ea5299499ZabtidataRecordIndex 138 backContext P2tleTO4ODhkYWZmMjUx T eee lt H rox niffer i Se ge han Load Test Result Detail Error Snapshots Help E Back Web Admin URL Error Index X Page Time Offset Error Date 7 Error Type URL A URL 13 Error1 Page 1 Start Page 1 39 min 27 Nov 2008 23 21 28 RequestTimeoutexpired GET httpviwww proxy sniffer com 80 images_en PointOfCollapse gif 4 URL 13 Error2 Page 1 Start Page 1 59 min 27 Nov 2008 23 21 49 Request Timeout expired GET httpviwww proxy sniffer com 80 images_en PointOfCollapse gif amp URL 13 Error3 Page 1 Start Page 1 59 min 27 Nov 2008 23 21 49 Request Timeout expired GET httpviwww proxy sniffer com 80 images_en PointOfCollapse gif Test Test01
45. 7997 for HTTPS In case if the technical web client uses encrypted HTTPS connections please take also a look at the Application Reference Manual chapter 3 12 CA Root Certificate Export Tool Furthermore as a second option it is also supported to create manually a text file by using any text editor which contains definitions of SOAP and or XML requests Such a file can then be converted to a web surfing session by using the import functionality of the Session Cutter see chapter 5 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 12 171 Apica ProxySniffer V5 0 User s Guide English Edition 3 1 2 Proxy Recorder Settings and GUI Settings Personal Settings Menu gt http 127 0 0 1 7990 PRX Main Menu Mozilla Firefox The Personal Settings menu allows you to configure non form File Edit View History Bookmarks Tools Help i based authentication methods NTLM Kerberos PKCS 11 and See Proxy Sniffer Main Menu ry F rm al o Ik q E ts i P Wackemin porani Eaton rip P Laer US die Bae are sexs s PKCS 12 based client certificates and some SSL options for the E ERR am m m w proxy recorder which may be necessary in order to successfully record a web surfing session Furthermore cascading the proxy recorder with another outbound proxy server of your company is also supported 3 PRX Personal Settings Mozilla Firefox A Note 1 the credentials for Basic and Digest
46. A URL 25 Error4 Page x3 sicherheitsnummer 4 19 min 28 Sep 2008 14 33 53 OK 200 Wrong HTTP Status Received z snit3 POST httpsdief testix past ch 44 3 efisecurezhtmli lodin A URL 25 Error Page x3 sicherheitsnummer 9 19 min 48 Sep 2009 14 33 53 OK 200 f vrang HTTP Status Received z snitl POST https eftestix past ch 443 efisecureihtmli lagin si URL 83 Error 2 Content Test Failure String Not Found Help Error Explanation next URL 8 3 POST httpsaifi testix post ch 443 ef securemhtml onl kdl z zkiis ez del 200 OK y RL Exec Step all done aj Error Log error string Auftrag wurde gel amp oumbscht not found inside content 200 OK TEXT HTML 3484 bytes 3669 ms Elf Display Response in web Browser HTTP Request Header 1 POST ef securehtml onl Kdl z zvis ez del HTTP 1 1 2 Host ef testix post ch v z si Done i 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 129 171 Apica ProxySniffer V5 0 User s Guide English Edition URL Exec Step The URL Exec Step reflects the internal processing state of the URL call captured at the point in time when the error has occurred Possible states are Internal Processing State of URL Call Value Meaning No Step Not Initialized 1 The URL call had not yet started Open Network Connection to Proxy O The URL call failed during the opening of a network connection to an outbound proxy server Open N
47. AG Switzerland All Rights Reserved Page 151 171 Apica ProxySniffer V5 0 User s Guide English Edition 13 1 1 Input Parameter Progress Display and Saving the Scan Result The window is divided into two parts The upper part of the window shows the progress of the scan or IG PAG the scan result when the scan has been completed The lower part of the window allows the setting of 2 3 hh e scan input parameters and the starting of a scan Help Personal Generate Analyse Refresh Pag T Project Tools Scanner Settings Navigator Load Test Load Tests Display Input Parameters PRX Page Scanner Mozilla Firefox http 127 0 0 1 7990 dFischer webadmininterf ace PopupPageScannerWeblet f e Starting Web Page URL from which the scan starts You can optionally scan e X i nasse Page Scanner b aD Refresh Close only parts of a web site by entering a deep linked URL path for example http www lt domain gt sales customers html In this case only web pages below Starting Web Page http 192 16 4 5 or at the same level of the URL path are scanned Scan Status scanning Scan started at 06 Dec 2008 00 13 17 e Char Encoding allows you to override the default value Auto Detect in case some or all web pages are wrongly coded such that the HTML header specified character set does not match the character set which is actually used within the HTML body of the web pages malformed HTML at server side You
48. Click on the Start Load Test Job button to start the job gt gt Start Load Test Job Display Real Time Statistic Schedule Job 1333for Day today Time HourMinute 22 v 50 w Schedule Job If you have de selected the checkbox Display Real Time Statistic the window will close after a few seconds however you can at any time access the real time statistic data or the result data of the job by using the Jobs menu see chapter 9 3 which can be called from the Main Menu and also from the Project Navigator Alternatively the load test program can also scheduled to be executed at a predefined time However the corresponding Exec Agent process must be available running at the predefined time because the scheduling entry is stored locally inside the Exec Agent jobs working directory which is monitored by the Exec Agent itself Especially if you have started the local Exec Agent implicitly by using the Proxy Sniffer Console AND if the scheduled job should run on that local Exec Agent you must keep the Proxy Sniffer Console Window open in order that the job will be started This restriction can be avoided by installing the Exec Agent as a Windows Service or as a Unix Daemon see Application Reference Manual Note if you close the window without clicking on the Start Load Test Job button the job remains in the state configured or scheduled Afterwards you can use the Jobs menu to start or delete the job or to schedule
49. E EL lh Check Condition after extracting variables I Break Outer Loop of Simulated User and report an Error if All Inner Loop Iterations are Completely Executed Error Message Error All iterations Add New Condition Done The Text Input Fields of the conditions can contain fixed text as well as placeholders for variables Example Dear vTitle 6vName In addition it is also supported to define a NOT condition for an absence of a text This can be done by enfolding the whole text with an exclamation mark and square brackets Example Dear vTitle vName Restrictions if nested inner loops have been defined a continue or a break action will only change the run time behavior of the deepest inner loop Breaking through several inner loop levels is not supported Further Hint for Using Variables when using variables please consider also the scope of the variables page 40 If the scope is global all simulated users will see the same value for such a variable and therefore the same condition will be become true or false for all users On the other hand if the scope of a variable is user or loop each simulated user will see a different value for such a variable and therefore the conditions will be calculated on a per user basis 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 38 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 Dynamic Session Parameters
50. ECT GMT 1 00 Berlin Bern Paris Madrid Rom Wien a SSL Session Cache enabled Number Format 123456 00 v The Web GUI part of the menu allows you to set the SSL Session Cache Timeout 10min Y Background Color hex FFFFFF default time zone and the default number format Enhanced Compatibility Mode a a E N PM which will be used by the GUI and by the load test 2e90902025 5902520 odege _ programs Cotetetess 2 209e9090 95 99957 O 290999095 gt TIA ET e edo 9e Additionally also Alert Notifications can be vor 909 configured which are send during the execution of a Cont obs ee mau JOO as Emails or as SMS messages see chapter OS Specific Library pkcs11wrapper dll Only temporarily applied until program termination see help to change these values permanently 1 2 2 Device Specific Library siecap11 dll pm ress solo Enter a simple file name without path and copy both files manually to javascript PopupHelpWith nchor nextProxyConfig 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 13 171 Apica ProxySniffer V5 0 User s Guide English Edition 3 1 2 1 Connect to Next Proxy Proxy Recorder Checkbox in Title if checked Proxy Sniffer cascades the proxy recorder with another next outbound proxy server of your company Note to execute a load test through a proxy server you must also enable the option Load T
51. Exec Agent Each time when a job is started on an Exec Agent the Exec Agent tries to read this file which means that the file can be created updated or deleted without the need of restarting the corresponding Exec Agent Copy Lo m Alert E 5 Configuration Local Exec Agent Remote Exec Agent Proxy Sniffer AlertConfig xml AlertConfig xml GUI SMTP Server INTERNET y Clickatell HTTP A SMS Gateway As a further option it is also supported to use a specific alert configuration for a particular load test program In such a case you have first to place a copy of the file AlertConfig xml inside the Project Navigator directory where the load test program is stored After that you can manually edit the copied AlertConfig xml file and then you have to ZIP it together with the compiled class of the load test program similar to the procedure which is required for using input files or using plug ins This effect that the program specific alert configuration is automatically transmitted to the Exec Agent s and that it overrides the default behavior on the Exec Agent s Note the copy of the AlertConfig xml file is stored in such a case inside the job specific directory on the Exec Agent 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 148 171 Apica ProxySniffer V5 0 User s Guide English Edition 12 2 1 Alert Conditions
52. Nov 2009 15 35 18 ECT Planned test Gication 10 00 min 2 54 min elapsed Combined Real Time Data at Cluster Level Number of Users Waiting for Response purple bars Il Total Passed URL Calls 3 Total Passed Loops A Users Waiting For Response 56 of 200 Total Failed URL Calls Total Failed Loops 87 TCP Socket Connect Time ms 1 sor Alice i E Usg i AY Network Throughput Miis Acces to Real Time Response Time Overview n Tom Error Overview Diagrams Monitoring of Exec Agents Diagrams of Exec Agents 3 09 vjasw M of Exec Agents Cluster Member 0 Trans Bytes Passed Loops Failed Loops Passed URL Calls Failed URL Local Exec Agent e 165 095 885 465 28 17 398 28 92 A Real time Data per Remote Exec Agent 177869159 494 30 19379 30 101 A Cluster Member Remote Exec Agent 12 running 112 46 178447791 494 29 19434 29 104 A Total c1 32971 521412835 87 56 811 384 Non Fatal Errors E T E yellow bars cM v http 127 0 0 1 7990 dFfischer webadmininterf ace PopupClusterStartFinalIfr ameWeblet jobId 208 amp actionzdisplayExecAgentDetails amp execAgentIds 1257085231 If you want to abort the cluster job you must do it at this level as this will also abort all Exec Agent jobs Aborting a single Exec Agent job will not interrupt the cluster job The same applies to the statistics result file prxres which must be accessed at this level 2012 2013 Ingenieurb ro David Fischer
53. O ms Average 723 mz Response Time Ome 814 mz 3ms English Edition Description displays during the load test at real time the response times of a URL and also a summary diagram about the measured errors of the URL Please consider that maybe only a fraction of the response times is shown depending on the Additional Sampling Rate per URL Call which was selected when the load test was started For example only every fifth response time is shown if the Additional Sampling Rate per URL Call was set to 20 Input Fields e Response Time drop down list Allows to select the period from the current time back to the past within the response times are shown inside the diagram e Time Bars drop down list Allows to select if the bars inside the diagram are shown as average values or as max values Please note that there if only a difference between the max values and the average values if multiple measured samples of the response time fall inside the same pixel inside the same displayed bar 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 86 171 Apica ProxySniffer V5 0 User s Guide English Edition Info Box Measured Values Measured Values Calculated Overall URL 1 Calls All values in this info box are calculated overall successful completed calls of the URL Total Passed URL Calls EM a2 measured since the load test was started These values are always
54. Page http www proaxy sniffer ch Char Encoding Auto Detect Exclude Path Patterns forums News Follow web Servers Verify External Links no Include text html bmp css gif ica img Javascript jpeg jpg js pct pict png stream stylesheet swt tga tif tiff Maz Scan Time 10 minutes Max Web Pages 100 Max Recened Bytes 40 MB Max URL Calls unlimited URL Timeout 20 seconds Max Path Depth unlimited Follow Redirections 10 Follow Path Repetitions 1 Follow CG Parameters no Browser Language none Annotation English Edition e Scan Statistic displays some additional statistical data about the scan Similar Web Pages are the number of web pages with duplicate content same content but different URL path Failed URL Calls are the number of URL calls which failed such that no HTTP status code was available no response received from web server or that the received HTTP status was an error code 400 599 Scanned Web Pages 53 Similar Web Pages 1 Followed Redirections 0 Non Followed Redirections 0 Processed URL Calls 129 Failed URL Calls 0 Received Bytes 3 666 462 O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 157 171 Apica ProxySniffer V5 0 User s Guide English Edition e Non Processed Web Servers displays a summary of all web servers which have been found in hyperlinks but whose web pages or page elements have not been scanned The
55. Requests 104 98 kbytes sec a gn HD Page 3 Support users thinctime 3 seconds 35 loop 2 10iterations em Test Offset Position Content Size Time HTTP Request HTTP Response 28 28 0 00sec NS 0660 bytes 391 ms 3 GET niti 92 16 4 5 support en html 200 C 29 pg 045see 5668byes 31ms GET http 192 16 4 5imap earth gif 200 Ol 30 30 O45sec MH 2693byes 46ms MEGEThtpur9216 4 5 map australia gif 200 1 31 31 O45sec N 1973bytes 62 ms MMGET httpirt 32 16 4 5 map german gif 200 ix 32 B3 045sec COW 235bytes 78ms MEGET nttpir192 16 4 5 map sweden gif 2000 otal 0 53 sec 24 129 bytes 5 Requests 45 36 kbytes sec x 33 331 HO Page 4 References user s thinktime 3 seconds 35 O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 35 171 Apica ProxySniffer V5 0 User s Guide English Edition 6 1 Conditional Execution of Parts of the Web Surfing Session O Insert Inner Loop Inner Loop Start Page Page 1 Web Load and Stress Testing Tool Proxy Sniffer Inner Loop End Page Page 1 Weh Load and Stress Testing Tool Proxy Sniffer Loop Iterations C fixed to 10 w variable 1 vInnerLoopCounti 4 Action if planned duration of Load Test exceeded Abort inner loop after current iteration _ Enable Pacing fixed to 860 w Min loop duration per user C5 variable 1 vinnerLooptountl Apply O 2012
56. Response Time per Page verage milliseconds 1 User 10 User 20 User 50 User 100 User 200 User Page Fi Start Page Page 2 Special Features Page 3 Download Page 4 Customers amp Parners concurrent user V e IQ q H en tm lt Page 6 Support FAGs 2012 2013 Ingenieurb ro David Fischer AG Switzerland With small loads the response times are constant and are independent of the number of users If the load is increased and thereby the maximum throughput of the server is reached measured in URL calls per second which is the web transaction rate or also called hits per second the response times will rise in an at least linear relationship with the number of users Web pages and or URL calls whose response times rise more strongly than others while under load are potential tuning candidates that is the reason for the sudden strong rise in their response times should be investigated Please note that not all web servers or web applications show a linear response time behavior if they are overloaded A web server may collapse in this situation in this the case the throughput falls after a specific load point has been exceeded infinite Response Times Throughput a Point of Collapse i i Complete Failure dead ist Number of Users All Rights Reserved Page 133 171 Apica ProxySniffer V5 0 User s Guide English Edition To produce the
57. Session Refresh Close Recorded Items 80 E I DO s Recording State STOPPED Overall one ben Recariing Pier m HTTP Request Header MA gt 107164580 Y a 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 69 171 Apica ProxySniffer V5 0 User s Guide Normally you should only have to enter the name of the load test program and to configure the Runtime Execution Behavior serial or parallel execution order of the URLs within the Web pages applied per simulated user without having to choose or modify any other options opecial options are only needed if o you have to execute the load test over an outbound proxy server see chapter 3 1 2 1 o you want to use more than one user account for Basic Authentication or if Digest Authentication is required against the web server o NTLM or Kerberos authentication was required to record the web surfing session see chapter 3 1 2 5 and 3 1 2 6 o an X509 client certificate was required to record the web surfing session see chapter 3 1 2 3 Input Fields e Java Classname desired name of the new load test program e Content Test Algorithm defines how the received content of the URL calls will be verified during the load test English Edition PRX Generate HTIP S Load Test Program Mozilla Firefox 127 0 0 1 7990 dfischer webadmininterface PopupCreateLoadtestWeblet m Apica 9 x ProxySniffer Gener
58. T 7 Total Passed Loops 3 449 Total Failed Loops 743 Users Waiting For Response 2 of 200 TCP Socket Connect Time ms 0 Users Think Time per Loop sec 20 0 AV Network Throughput Mbit s 23 30 Session Time per Loop sec 20 0 Total Transmitted Bytes 1 333 041 957 x Overview about Execution Steps of simulated users Progress 0 Remote Exec Agent Job 188 Real Time Comment Add Job Parameter Test01 u 200 d 900 t 60 sdelay 200 maxloops O sampling 15 percpage 100 percurl 100 maxerrmem 20 nolog English Edition Real time statistics shown in this window are updated every 5 seconds for as long as the load test job is running You may abort a running load test job by clicking on the Abort Job button This will take a few seconds because the job writes out the statistic result file prxres before it terminates Note closing this window will not stop the load test job If you close this window you can later acquire the load test result or return to this dialogue if the load test is still running by clicking on the Jobs icon in the Main Menu or in the Project Navigator window see chapter 9 3 e Exec Agent Name or Cluster Name The name of the Exec Agent or the name of the Exec Agent Cluster which executes the load test job see also chapter 11 Distributed Load Tests Architecture and Configuration e Job numbers Unique job ID unique per Exec Agent or unique cluster job ID
59. Test Extract Recorded Value 151047 Extracted on Line 232 Map to Var Name Y a Assign varto all request parameter with same recorded value J M Try URL Encoding m Assign varto all matching request file and request content patterns with same recorded value Extract Page 62 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 6 1 Extracting and Assigning Values of XML and SOAP Data In case that XML or SOAP data have been recorded Proxy Sniffer parses such data automatically and displays an additional XML Icon within the title of the HTTP response Content and the HTTP request content box EE Extract Var from XML Data da ls POST ipedasipkviPkvGatekeeper HTTP 1 1 gt E ContentType textil charset ut 8 gt AML Path vegas e ace 3 SOAP Action RES US java PersonEroefnenkResponse 4 User Agent Java 1 5 0 04 M GPENSIIDfEN http 127 0 0 1 7990 displayIndex 1 PRX 1 XML Response Content Windows Internet Explore gt j EFSONLDATEN 5 Host apse34 7021 K zPersonz mn bd FersondlD i E EN P D rt cl Item 1 XML Response Content A Keys nn NV Path Occ Hr E m HTTP 1 1 200 OK Exdractwhole value 3 lt soapenv mm 4 2 Date Fri 26 Oct 2007 11 14 08 GMT lt java PersonEroeffnenResponse xulns java java ch post pf fias pkvgatekeeper tyvpe gt C Extract Token from Value 3 Transfer Encoding chunked See re
60. Time 188 8686 mz C46 GET http u24 lscache3 c0 yaoutube com Sey i deoplauback sparamsezidz2b6easapireszzcipz2bCipbitsszzcitagzzcalaorithmzzCb total bytes 50 000 000 45 000 000 40 000 000 55 000 000 so 000 000 5 000 000 z0 000 000 15 000 000 10 000 000 5 000 000 o 000 4 0 0nn 60 000 30 000 100 000 Ee time to receive The response Content 50 ms 4 H Load Test x youtubedz Request Start Time 2012 2013 Ingenieurb ro David Fischer AG Switzerland 120 000 elapsed time before start of receiwing the response content 11 754 ms 50 ms 20 37 amp 6496 bytes user data Bi chunk protocol gt Mrequired throughput time milliseconds 140 000 160 000 120 000 00 000 Insufficient Throughput Rate 0 Ba 15 Apr 2611 18 36 43 All Rights Reserved Sample Ho 1 Met Con Reg Trim Vai Header Rev Content Rev Total Sample Ho 2 Met Con Reg Trim ait Header Rev Content Rev Total Sample Ho 3 gt 1 ms Ome 40 ms Ome 161 657 ms 161 698 ms 1 ms O ms 49 ms O ms 1B1 83 ms 161 833 ms lt 4 E English Edition Measured internal throughput of a video on a preset viewing time of 3 minutes 180 000 milliseconds The linear flow and the flow rate peak at the beginning of receiving the data indicates that the delivery is made by a special video server wh
61. b no pages gt delete selected gt no pages Start Stop Reset Recording Recording Recording Recorded Items 0 Recording State STOPPED Search Session Scratch Area 2 gt delete selected b no pages e gt x T gt aA Import Session Help Close Import Session from Definition File Browse file format Help javascript SessionCutterImportSession Q Data Format of Definition Files Definition Files are written in ASCII format Each line contains either a command or a URL definition Commands always begin with a hyphen URL definitions must contain at least 3 arguments 1 HTTP method GET POST 2 absolute or relative URL 3 expected HTTP response status code of the URL call 200 302 4 Argument 4 of a URL definition is optional and contains the request content All further arguments are optional and contain URL options which begin with a hyphen lt command gt lt argument 1 gt lt argument n gt lt HTTP method gt lt URL gt lt HTTP status code gt lt request content gt lt URLoption 1 gt lt URLoption n gt lt HTTP method gt lt URL gt lt HTTP status code gt lt request content gt lt command gt lt argument 1 gt lt argument n gt lt HTTP method gt lt URL gt lt HTTP status code gt lt request content gt lt URLoption 1 gt lt URLoption n gt lt URLoption 1 gt lt URLoption n gt 2012 2013 Ingen
62. case Connect to Next Proxy as well as Kerberos Authentication You have also to configure the Kerberos Domain Name and the IP address or the DNS name of the Kerberos KDC Active Directory Server inside the Kerberos configuration file krb5 conf 3 1 2 2 HTTPS Settings Proxy Recorder Allows you to adjust the HTTPS settings of the proxy recorder used when recording encrypted network connections Input Fields e SSL Version allows you to select the SSL protocol version e SSL Session Cache enabled if checked enables the SSL session cache keeping the same SSL session ID over multiple Web pages e SSL Session Cache Timeout lifetime of the SSL sessions within the session cache e Enhanced Compatibility Mode if checked enables workarounds to support poorly implemented SSL server libraries 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 14 171 Apica ProxySniffer V5 0 User s Guide English Edition e HTTPS Response Timeout response timeout per HTTPS URL call If this timeout expires the corresponding HTTPS URL call will be aborted 3 1 2 3 HTTPS Client Certificate based Authentication PKCS 12 Files Proxy Recorder Allows you to load X509 SSL TLS client certificates in PKCS 12 file format into the proxy recorder Because the proxy recorder operates as a man in the middle between the web browser and the web server the client certificate must be loaded and activated before a web surfing ses
63. ch 443 efipubliciccicssistyles_ef css 0 339 O 15ms GEThtpsieftestipostch 443 efipubliciccipicsibackground gif o 339 0 13ms GEThttps leftestixpostch 443 efipublicicc pics img pf logo de jpg 0 339 O 11 ms GET https leftestix postch 443 efipublic cc pics doc bg gif o 339 0 19ms GET https leftestix postch 443 efipublic cc pics img claim de gif 0 93 O 16ms GEThtpsieftestixpostch 443 efipubliciccipiesticons gif oC COMA ARCO me GET https i ef testix post ch 443 effpubliciccipics shadowAlpha png ERES GET https i ef testix post ch 443 favicon ico Be LEBE e N and Um T ETS GET m Hef testix m ch 443feffpublic ccipicsfidpca anleitung_pk_de gif A Page 3 sicherheitsnummer Cra E POST https i ef testix post ch 443 efisecure html login I IP 2 R lleare VERA me GET htnc irat tactiv noct ch AA YHoflcar ralhtmliinnintanl lal Innin nrncaarn v lii gt E N S ES 2 Lil O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 132 171 Apica ProxySniffer V5 0 User s Guide 10 3 Load Curve Diagrams English Edition To discover the maximum possible capability of the web server or web application you must run the same load test program several times each time with a different number of users We recommend increasing the load in each successive test run logarithmically in order to get a good overview for example suc
64. content Request Start Time 26 May 2611 14 35 21 Load Test Testi Response Content Throughput Sample Ho 2 Max Acceptable Response Time 38 BBB mz 391 GET http 4 1922 156 4 5 8B zen docz ProaauSnitfferllaerzauideld5En pdt total betes 34 330 ms 700 ms 10 000 000 3000 000 5000 000 i 7000000 e 000 ogg a 2 E a 6 S85 959 bytes sooo ood 4 000 000 3 000 000 2 000 000 1 000 000 eM user data Bi chunk protocol gt Mrequired throughput time milliseconds 4 000 3 000 12 000 is nu 20 000 E gt time to receive the response content 700 ms oo elapsed time before start of receiwing the response content 24 000 25 000 32 000 36 000 40 000 Insufficient Throughput Rate 59 04 Load Test Testai Request Start Time 26 May 2611 14 35 25 Seles o S ex Help Refresh Close Sample Ho 1 Met Con ms Reg Trim Ome Vai 62 me Header Rev O ms Content Rev 28 949 mz Total 29 011 ms Sample Ho 2 Met Con ms Reg Trim O me ait FOO me Header Rev Ome Content Rev 38330 ms Total 40 030 ms Sample Ho 3 All Rights Reserved English Edition Throughput measurement of a PDF document which should be received in 30 seconds by a linear network throughput in order that the beginning of the document can already be viewed after some few seconds The second measured sample does not meet this requirement Pa
65. deleted because the old main directory is also stored in a browser cookie After that Proxy Sniffer can then be re started and the new main directory will be active Further information about Proxy Sniffer configuration files can be found in the Application Reference Manual Chapter 7 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 106 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 7 More Hints for Executing Load Tests Please note that the underlying operating system of a single Exec Agent load injector can be overloaded if too many concurrent virtual users are executed there In most cases where a system Is overloaded the CPU s of the Exec Agent will be constantly at nearly 100 used In these cases the measured response times will not be valid because the measuring system itself is overloaded We recommend that you monitor the CPU consumption of the Exec Agent during the load test and that you use an Exec Agent Cluster Chapter 11 2 instead of a single Exec Agent when a single system does not have the necessary CPU resources to properly generate the load The CPU consumption of the load releasing system depends on the number of users more users more CPU the user s think time longer think time less CPU the response times of the stressed web server longer response times less CPU and whether the HTTP or the HTTPS protocol is used HTTPS more CPU We are therefore not
66. different load tests especially when the same load test program was executed several times with the same number of concurrent users All Rights Reserved Page 105 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 6 1 Configuration of the Project Navigator Main Directory Proxy Sniffer can be configured to have its Project Navigator Main Directory on a shared disk or a shared directory given all members of a team the same view of the data On Windows a directory Share must already exist On Unix systems the shared directory must be already mounted using NFS or mounted via Samba Proceed as follows e Windows systems the Proxy Sniffer mytests dat configuration file must be edited using a text editor such as Notepad The entry in this file must point to the directory share This directory shared must be created using Windows before the Proxy Sniffer configuration file is edited The mytests dat is located in the Proxy Sniffer installation directory e Unix systems on Unix systems the mytests dat configuration file must be manually created in the Proxy Sniffer installation directory using a text editor such as vi The only entry in this file should be the path to the new main directory Note on Unix systems which have only an Exec Agent started this file is not necessary After setting the new Project Navigator main directory the Proxy Sniffer application must be closed In addition all cookies in your Web Browser must be
67. elapsed time for all in one iteration executed page breaks and URL calls before the next Inner Loops Started on Page 3 iteration can start If the iteration is done faster than the pacing time the none user will be inactive until the pacing time has elapsed Done T 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 34 171 Apica ProxySniffer V5 0 User s Guide English Edition Inner loops are marked by black bars at the left side in the Web Admin GUI main menu Nested inner loops are also supported x 19 19 359sec 1 894 bytes 31 ms GET httpi t92 16 4 5 faviconico 200 OK IMA Total 3 62 sec 178 046 bytes 19 Requests 49 12 kbytes sec 20 20 JO Page 2 Download user s think time 3 seconds 35 SHloop1 10iterations em Test Offset Position Content Size Time HTTP Request HTTP Response 21 21 0 00sec NT 21096 bytes 406 ms GET httpirt 92 16 4 5 download en html 200 22 22 0 48 sec Lo HM 96 bytes 31ms I iGET httpJ 192 16 4 5 new gif 200 OK IMAGE 23 23 048sec DE 959Bbytes 47 ms MMGET htto 192 16 4 5 pdf bookmarks gif 200 24 24 048secc MEI 27O5byes 62ms MGET httpirto2 16 4 5 px console gif 200 01 25 25 os0osc 3 W 73bytes 62 ms GET httpJi192 16 4 5 smallinfoTransp gif 200 26 26 0 50sece MEN 27141 bytes 78ms MMGET hito192 16 4 5if recordingextension cit otal 0 58 sec 60 679 bytes 6
68. further information 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 15 171 Apica ProxySniffer V5 0 User s Guide English Edition 3 1 2 5 NTLM Authentication Proxy Recorder Checkbox in Title if checked enables NTLM authentication against Web servers during recording Note to execute a load test which uses NTLM authentication you must also enable the option NTLM Authentication in the Generate HTTP S Load Test Program menu see chapter 8 The allocation of individual NTLM accounts per simulated user is supported when generating load test programs Input Fields e Domain Windows domain name e Username username of domain account e Password password of domain account 3 1 2 6 Kerberos Authentication Proxy Recorder Checkbox in Title if checked enables Kerberos authentication during recording of web surfing sessions against web servers and against next outbound proxy servers Note to execute a load test which uses Kerberos authentication you have additionally to enable the option Kerberos Authentication in the Generate HTTP S Load Test Program menu see chapter 8 The allocation of individual Kerberos accounts per simulated user is supported when generating load test programs Input Fields e Username the username of a domain account e Password the password of a domain account e Web Server Host Name Full DNS Name Required In order that a Kerberos authentication ca
69. here If Additional Options I sa the value of the input field Max Loops per User is not set e to unlimited the test run may complete before the planned test duration elapses because all users have already recommended will be displayed as hint in Project Navigator executed their maximum number of loops Done Max Loops per User maximum number of surf session repetitions per user If the value of the input field Load Test Duration is not set to unlimited the test run may complete before the planned test duration elapses because all users have already executed their maximum number of loops 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 75 171 Apica ProxySniffer V5 0 User s Guide English Edition Startup Delay per User usually concurrent users are not started at exactly the same time because this rather unusual scenario will overload the web server immediately This parameter controls how much time will elapse before an additional user will be started ramp up of load at Start Max Network Bandwidth per User allows you to reduce the maximum network bandwidth per user in order to simulate slow network connections to the web server for example connections over DSL or modem lines The downlink and the uplink speeds can be adjusted separately to simulate asymmetric network bandwidths Request Timeout per URL timeout in seconds per single URL call If this timeout expires the U
70. j Help Navigator Overall Load Test Session Refresh Close Help Navigator Overall Load Test Session Refresh Close Help Navigator Overall Load Test Session Refresh Close Var Handler ODK Var Handler GDR Var Handler Y d 3 no vars defined Add Load Test Plug in S33 S Add Load Test Plug in Step 2 weferMinutes global var gt gt en u gt gt i A Plug in Defer Load Test Start v J Plug in 1 Defer Load Test Start z nl eta a pa si A M Ps c E Input Files Add Input File e A Class DeferLoadTestStart class v e E E de Constructed global lt Execution before at start of Input Files Adel input Flo m I a load test Plugin in Description User Input Fields Add Input Field z x none Defers the start of the load test for a configurable Input Parameter none numger or minutas Defer Load Test Minytes User Input Field v User Input Fields Add Input Field Default Value lo Load Test Plug ins Add Plug in A vDeferMinutes Add Plug in ron Load Test Plug ins Add Plug in Plug in Desciption Defers the start of the load test for a configurable number A 1 Defer Load Test Start of minutes a some Plug Ins require input parameter Therefore it may be necessary to define additional variables One option to define such variables is to create global visible stand alone variables with constant initial values see chapter 7 9 in c
71. lt random deviation gt Allows the insertion of a Page Break This command can be called multiple times before or after URL definitions The first parameter is the comment for the page break the second parameter is the user s think time in Seconds and the third parameter is the random deviation of the think time in percent 0 100 eof Stops processing of the definition file at this point This command can be used when only a part of the URL definitions should be processed URL Options requestContentFile lt file name gt Allows the use of the content of a local file as request content Argument 4 of the URL definition is not used and not required if this option is set If the command defaultRequestContentDirectory was previously called the file name is only allowed to be the simple name of a file within the default request content directory requestContentType lt content type gt Allows the setting of a specific value for the request content type for this URL call The default value used when this option is not set is that set by the command defaultRequestContentType or failing that application x www form urlencoded if the command defaultRequestContentT ype was not previously used requestHeaderField lt request header field gt Allows the setting of an additional HTTP request header field for this URL call This option can be specified several times allowing the addition of several HTTP request header fields re
72. ma l Appa y J 1 eee Proxy Sniffer 3b dui Project Navigator Execute Load Test Execute Load Test Job Test01 Load Test Input Parameter 3 Test01 xml Host Local Exec Agent w Execute Test from Number of Concurrent Users 30 v Backto previous Menu Save image to disk URL Response Time 188 Sampling Rate 48 GET http v24 1 scache3 c youtube com 887v ideop layback sparamscidZ2Cesupirez2eCipAz2Cipbitsz2eCitaaz2CalgorithmZ2 Load Test Duration 1 min v AAN pe ERI Maximum Acceptable Max Loops per User unlimited 200000 Startup Delay per User 200 w Milliseconds unlimited Downlink unlimited Uplink 60 Seconds Max Network Bandwidth per User 160000 Request Timeout per URL 140000 Max Error Snapshots 20MB memory 120000 Response Time in Milliseconds 180000 unconfigured accept any 2 passed 10 100000 5 resp time exceeded O 3 failed 0000 Acceptable 100 0 60000 Statistic Sampling Interval 15 Seconds Additional Sampling Rate per Page Call 100 w Additional Sampling Rate per URL Call 100 v Add Resp Throughput Chart per Call vP ro anamma fecommended none recomme Performance Details per Call Resp Throughput Chart per Call Debug Options Additional Options 3 40000 20000 Request Headers f Req
73. or cancel the schedule of this job gt http 127 0 0 1 7990 PRX Main Menu Mozilla Firefox File Edit View History Bookmarks Tools Help i Proxy Sniffer Main Menu T i 4 sh lis lt 7 I Page Personal Project Load Test Generate Analyse Refresh EI Web Admin Professional Edition Help Tools Scanner Settings Navigator Jobs Load Test Load Tests Display Recorded Items 40 a FH i gt omm 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 80 171 Apica ProxySniffer V5 0 User s Guide 9 1 1 Real Time Job Statistics Exec Agent Jobs gt PRX Job 188 Mozilla Firefox Seles B http 127 0 0 1 7990 dfischerwebadmininterf ace PopupDirectoryNavigatorDisplayLoadTestWeblet filePathB642QzpcRG9rdW t IbnRIIHVuz CBFaWsSa e Proxy Sniffer el 2 f Web Admin Project Navigator Execute Load Test Help Jobs Refresh Close Re Response Time Overview Time Commen Error Overview d Statistical Overview EE jd Diagrams delay 200 md Diagrams 00 percurl100 md Diagrams Test started at 13 Oct 2009 20 18 48 ECT Paned test duration 15 00 min 7 38 min elapsed Server Throughput URL calls per second NUT I Total Passed URL Calls 89 094 Total Failed URL Calls 1 254 HTTP Keep Alive Efficiency X 95 5 AV Web Trans Rate callsisecy 194 7 Abort Job 188 Number of Users Non Fatal Errors Waiting for Response yellow bars Fatal Errors red bars pon x Er m
74. rhinkTime Var Label Text users Think Time Default value E Var Scope global var required Create Hint User Input Fields are freely configurable load test 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 50 171 Apica ProxySniffer V5 0 User s Guide English Edition E http 127 0 0 1 7990 index 0 PRX Item 0 Manage Page Windows Internet Explorer ne nus i o x e p Sniff Wabdmin Item 0 Manage Page Help Close Item Offset Position Content Size Time HTTP Request HTTP Ec x 1000sec MN 44741 bytes 156 ms GET hitto chia proses n38eacr 1 nes AR ms FS GET Witrr iman rr Modify Page Break Page Description Start Page User s Think Time C fixed to o second variable thinkTime seconds usp rink Time Randomness 35 y v Apply new user s think time values for all page breaks 2 n http 127 0 0 1 7990 7filePathB64 0zpcUHI Z3Ihb5BGaWrlc1x0Ocm94eYNuaWZmzXJIcTXIUZZNOc1xGaXIzdFRA Windows Internet Explorer ioj x b rox nitter gt gt u uii nA Project Navigator Execute Load Test RUE Refresh x Execute Load Test Job Test01 Test xml p Test Scenario Diagram Response Time per Fage P Results Execute Testfrom Cluster cl gt User s Think Time Diagram Response Time Percentiles Diagram Top Time Consuming URLs Diagrar H
75. sampling 15 percpage 100 percurl 20 7 the job lt gt Done Input Fields e Display Cluster Jobs shows all Exec Agent Cluster jobs e Display Exec Agent Jobs of allows to select the Exec Agent from which a list of all load test jobs is displayed e Clean Up Delete All Non Running Jobs deletes all jobs except running and scheduled jobs Note all jobs can be deleted after they have been acquired the test results will not be lost because the load test result data prxres file are transferred into the corresponding Project Navigator directory from which the load test has been started We recommend that you delete all old jobs at regular intervals e Clean Up Delete Old Completed Jobs deletes all completed jobs except the newest one This button is only shown if at minimum two jobs have been completed 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 98 171 Apica ProxySniffer V5 0 User s Guide English Edition Columns of the job list Job Each job has its unique ID which was automatically assigned when the job was defined However the ID is only unique per Exec Agent Cluster jobs have an own separate ID own enumeration counter A Allows to acquire the statistic result file prxres of an already completed load test job or reconnects to the temporary statistic of the load test job if the job is still running or allows to cancel the schedule of the job Deletes all data
76. sites including all web pages contained therein similar to a Web Spider or a Web Crawler The result of the scan can be converted directly into a web surfing session out of which a ready to run load test program can be created This is a fast and convenient alternative instead of recording web surfing sessions manually using the Main Menu However this option is Suitable only for testing relatively simple web sites In general real world web applications can only be tested using manually recorded web browser sessions via the Main Menu Var Finder provides a convenient overview of all CGI and form parameters passed between client and server in a complete web surfing session Using this menu dynamic session parameters such as the NET VIEWSTATE parameter can be managed with a single mouse click URL Details Var Handler URL Details displays all recorded details about a URL The Var Handler allows Input Files to be defined with URL parameter allocations useful in situations such as logging in to web applications using different user accounts The Var Handler also allows many additional load test program options to be dynamically changed or example changing the name of the target web application server Response Verification In addition to checking only status HTTP codes during a load test Proxy Sniffer also checks the received content of web pages by an automatically applied heuristic algorithm designed to exclude false positive res
77. the actual number of non fatal errors yellow bars as well as the number of fatal errors red bars failed sessions counted overall simulated users By clicking on this diagram the Error Overview Diagrams are shown see chapter 9 1 1 3 e Total Passed Loops The total number of passed loops repetitions of web surfing sessions since the load test was started e Total Failed Loops The total number of failed loops repetitions of web surfing sessions since the load test was Total Passed Loops 30440 started Total Failed Loops r43 E User s Think Time per Loop sec 20 0 e 2 User s Think Time per Loop sec The total user s think time in seconds for one loop per simulated user Session Time per Loop secs 20 0 e Session Time per Loop sec The average session time for one loop per simulated user This value is the sum of the average response time of all URLs and all user s think times per successful completed loop The Number of Users Waiting Users Diagram shows the total number of currently simulated users red bars as well as the actual number of users which are waiting for response from the web server purple bars The users waiting for response is a subset of the currently simulated users By clicking on this diagram the Statistical Overview Diagrams are shown see chapter 9 1 1 4 e Users Waiting For Response the actual number of users which are waiting for response from the web server compared to of the tot
78. the load test program is generated again You should be sure that you have already made all Var Handler definitions such as defining Input Files and User Input Fields before you start modifying the program code All special classes and methods used by the load test programs are fully described in the Proxy Sniffer Java API documentation in order to enable you to understand how the program works On Windows systems the Proxy Sniffer Java API documentation is accessible from the Start gt Programs gt ProxySniffer menu The inner structure of a load test program is organized as follows Measured Web Server Load Test Program class Number of Users Number of created Instances 2012 2013 Ingenieurb ro David Fischer AG Switzerland The main method which is marked by a blue background at the bottom on the image first reads all input data After that the structure of the statistics data is created Then an own instance of the load test program itself is created for each emulated user The main method starts these users instances in a loop and then waits until all users have completed their work Finally the statistics result file is written and the load test program is terminated The method run which is the main method of a single thread or user controls the number of loops and or the elapsed time and terminates the activity of the user if one of these values has been exceeded The method execute
79. the response times for all URLs of the web page Also these response times are either average values or max values depending on the selection in the Time Bars drop down list However these values are calculated since the load test was started and always accurately measured which means that they do not depend on the value chosen for the Additional Sampling Rate per Page Call You can click on a URL response time to show the corresponding URL Response Time Diagram see chapter 9 1 1 2 At the left side inside the diagram the average response time of the web page is shown as red colored text calculated since the load test was started But depending on the selected period this value may not be displayed in every case At the right side inside the diagram the last measured value is shown Page 81 Besponske Tire is Sampling Bate Remote Exec Agent Job ide Page 21 Heb Los and Stress Testing T Roser Up lus 19 Get 2004 2 29 10 10 29 min elacted r sp ntia Tim in millidecond D pL D0 0 00 54 Diz DiDD 4 2 00 385 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 85 171 Apica ProxySniffer V5 0 User s Guide 9 1 1 2 URL Response Time Diagram Real Time I PRX Job 188 URL 1 Overview Mozilla Firefox be qe Proxy Sniffer Web Admin URL 1 Overview Real Time Remote Exec Agent Job 188 Response Time Last 5 Minutes Time Bars Average Val
80. the transient memory of a particular cluster member and also because the server side SSL cache is usually handled by the cluster members and not by the load balancer This load balancer functionality is called IP stickiness which represents the recording of client IP addresses inside the load balancer algorithms This term has nothing to do with the sticky bit of Unix file systems If you encounter this situation the load will appear on only one web server and will not be distributed across all web server cluster members The solution to this load balancer behavior is to have the Exec Agent use multiple client IP addresses during the load test therefore each concurrent user will have its own IP address or if more concurrent users are running than available local IP addresses the local IP addresses will be averaged across the concurrent users 1 The first step to enable multiple IP addresses for an Exec Agent is to reconfigure the underlying Windows or Unix operating system such that multiple local IP addresses are available This can be done by assigning additional IP addresses to the same physical network interface 2 he second step is to assign these multiple IP addresses to the Exec Agent configuration For the local host where the Web Admin GUI is running the second step can be done by invoking the Setup menu inside the Project Navigator gear wheel icon in the top navigation For remote Exec Agents you must edit the file javaSe
81. time of the error is shown The URL in which the error occurred is marked with a pink background T000279 error string Bitte geben Sie not found inside content 200 OK TEXT HTML 6278 bytes 342 ms Thread Statistic at Error Time on Cluster Member z snit 1 Index NoofUsers Passed Failed AV Time Thread Step Page 0 15Users 15000ms Page 1 startseite 1 1User 940 O 361 ms GEThttpsllef testixpost ch 443 efisecureihtmli login amp resetlogin amp p spr cd 1 URL mjo 8940 0 33ms GEThttps fef testix post ch 443 efipublicicc js jquery 1 2 6 min js Ur 0 940 O 13ms GEThtpsieftestixpostch 443 efipubliciccissiiquery bgiframe packjs o 0 O 22 mss GET hittps eFtestix post ch 443 efipubliciccijsitabbed js 0 9S940 O 1 ms GET hitps ettestix post ch 443 efipubliciccijsiariajs o 40 O 13ms GEThtpsieftestixpostch 443 efipubliciccisitooltip2 js 0 0 0 16ms GEThtpsifeftestixpostch 443 efipublicicciisiwtjs 0 940 O 12ms GEThttps leftestixpostch 443 efipublic cc jslef base js User 939 0 10ms GEThtpsieftesticpostch 443 efipubliciccicssistyles css o gt 938 0 29ms GET https leftestix postch 443 efipublicicc css elements css o 938 0 17ms GEThtpsieftestixpostch 443 efipubliciccicssiframework css o 93 0 18ms GEThtpsieftesticpostch 443 efipubliciccicssielements_form css o 339 O 14ms GET https leftestix post
82. up and run on several computers in which case the load releasing computers are combined into a virtual application cluster The configuration is very simple and only requires that an Exec Agent process be installed on the involved load releasing systems This is implied in the case where the product has been installed and started on several computers as each system already will contain an Exec Agent Alternatively individual Exec Agent processes can be installed separately as a Windows service and or a Unix daemon see the Application Reference Manual The communication between the Web Admin GUI and the remote Exec Agent processes usually uses raw TCP IP network connections to port 7993 however this port number can be freely chosen if the Exec Agent process is installed separately The communication can also be made over HTTP or HTTPS connections tunneling and also supports outbound HTTP S proxy servers The support of outbound HTTP S proxy server means in this case that load tests can be started from a protected corporate network and then transmitted over the proxy server of the corporation to any load releasing system on the internet all without the need for ordering new firewall rules The computers of a load releasing cluster the cluster members may also be heterogeneous that is Windows and Unix systems as well as strong and weak systems can be mixed within the same cluster The individual cluster members can be placed in differen
83. web surfing sessions loops per minute measured over all concurrent users The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Completed Loops per Minute Sampling Interval 15 Seconds number of successful completed loops per minute sessions per minute 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 30 0 20 00 0 22 30 0 25 00 Load Test Test 1 28 Sep 2669 14 24 34 User 1600 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 122 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 12 Diagram TCP Socket Connect Time Shows the time to open a new network connection to the web server before data are sent socket open time The number of data points depends on the Statistic Sampling Interval which was set when the test run was started TCP Socket Connect Time Sampling Interval 15 Seconds time to establish a tcp network connection to the web server in milliseconds 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 50 0 20 00 0 22 30 0 25 00 Load Test Test 1 28 Sep 2669 14 24 34 User 1600 10 1 13 Diagram Network Throughput Shows the total network throughput of the test run measured over all user
84. writes out all transmitted and received HTTP headers to standard output debug content amp loops includes the above option debug loops and in addition writes out all transmitted and received HTTP content data to standard output however this option only writes out data which has been transmitted or received in ASCII format such as HTML form parameters and HTML XML SOAP or CSS style sheet data but no binary data such as images debug cookies amp loops includes the above option debug loops and in addition writes out all received and transmitted cookies to standard output debug keep alive amp loops includes the above option debug loops and in addition writes out additional debug information about re used network connections to standard output debug SSL handshake amp loops includes the above option debug loops and in addition writes out additional debug information about SSL handshakes to standard output Additional Options these options allow you to enter special options All special options keywords begin with a minus sign Several options can also be combined separated by space characters O multihomed Forces the Exec Agent s to use multiple local IP addresses when executing the load test This option is only used by the Exec Agent s if multiple IP addresses are configured at the operating system level and are assigned to the Exec Agent configuration see Chapter 12 The effect of this option is
85. 01 java Excellence in Cloud Performance y Response Verification Summary 000 2011 A a clear improvement of the test quality is achieved URL Execution G 01Mar2011 x86 E V size 15 1740 bytes no failure action Serial Executed 19 02Feb2011 6 18 A size 15 12200 bytes no failure action Parallel Executed 82 B 14 14 A size 5 680 bytes no failure action Threads pee 6 02Mar2011 19 19 A size 556 770890 bytes no failure action PE 03Feb2011 22 A size 15 177150 bytes no failure action Bou E 03002011 04Mar2011 040ct2011 050ct2011 06Sep2011 34 34 amp The number of virtual users is counted locally on the Ex 07Mar2011 36 36 size 15 205830 bytes no failure action 088ep2011 095ep2011 40 30 A our license price and renting the Amazon Cloud servers 10Aug2011 11Aug2011 46 46 amp To learn more about Apica s web performance services p lt lt gt 26 26 amp Download Proxy Sniffer Installation Kits 27 27 A size 5 56680 bytes no failure action 30 30 amp Fixed license fee per year for local load generators Overwrite amp Compile All Rights Reserved Page 26 171 Apica ProxySniffer V5 0 User s Guide English Edition After that the load test program can be started It is recommended that you choose for the first test run only a few number of simulated users and a short execution time PRX Project Navigator Mozilla Firefox SEE B http 127 0 0 1 7
86. 09 20 27 04 Elapsed Time 8 15 min User 266 T Page at y eres iud fe eae ate eges Uo ale number of errors which did occur on ne Bra the URLs of the web page You can 3 s lo click on a error counter to show the NT error detail information error bioo oo oioo so oono oasa owsio owsao owswo Gosso Gosto oTo 0 0720 meo snapshots for the corresponding URL Humber of Errors on Page Hz Remote Exec Agent Job 1883 Page 2 Web Lasttest und Stresstest All failed URL Calls 61 Errors 17 m number of errors 1 Current Tine inn First Error Snapshots Displays a list about errors which did occur at first at the start of the load test By clicking on a magnifier icon the corresponding error detail information error snapshot is shown 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 88 171 Apica ProxySniffer V5 0 User s Guide English Edition Latest Error Snapshots Displays a list about the latest newest errors By clicking on a magnifier icon the corresponding error detail information error snapshot is shown Failure Disgrams E FintEsor Brapstols RER Er Al feed URL Cabs C geston Faluras uuy Ema Brapshal Memory 22 used Latest 100 Erro tepals Lure Time 13 C 8 2058 20438 JH Bb Tie U m n Ma bakes ina Tene Geer Type Par AL ides Erie UM 2g 2 am Gonet Teri Fakt Bring Hok Fara me FI ie Lied ara Sir Tarima T na Al ATA 1
87. 127 0 0 1 7990 dfischer webadmininterface PopupDirectoryNavigatorDisplayLiveStatisticsCurvesWeblet joblId 188kexecAgentId 12403061524068FilePathB64 QzpcRGordw IbnRIIHvuzCBFaWszdavsbHvuz2Vuxal 1dG9uZ1xQcm94 eVNuawZmZxXIcTXIUZXNOc1 xXUZXNOMDEUY2xhc3M 40 amp e OV e rvi ew d j a g r am S at r e al ti m e eee gee Proxy sniffer Statistical Overview Real Time Remote Exec Agent Job 188 A about a load test job View O one column two columns C Auto Update every 10 seconds coco ac mug pp MP pc canc Note the values shown in the mgr mer et ree et mgr wos etn TOE aw ara TEE diagrams are captured at regular MM E te UN EN cera A intervals depending on the Statistic Sampling Interval which was selected when the load test was Session Failures 15 Sec Sampling Interval Remote Exec Agent Job 188 Session Time per User per Loop 15 Sec Sampling Interval Remote E Job 188 Test 1 started at 13 Oct 2889 20 18 48 11 48 min elapsed Test 1 started at 13 Oct 2009 20 18 48 11 48 min elapsed starte number of failed loops failed sessions DUE EE session time per loop in seconds 1 ent nce Diagrams e Concurrent Users The total Web Transaction poi 15 Sec Sampling Interval Remote Exec Agent Job 188 Completed Loops per te 15 Sec Sampling Interval Remote Exec Ag Job 188 number of simulated users est l started at 13 Oct 2009 20 18 48 11 46 min elapsed est l started n ie Oct 2669 26 18 48 11 40 min elap
88. 163283 first 94153 demand 451047 449647 http 127 0 0 1 7990 Proxy Sniffer Var Finder Mozilla Firefox ini x eee Hint try out at first to apply automated dynamical handling by clicking on the V icon of a parameter If this is not Q eee Proxy Sniffer possible click on the magnifier icon find out the first occurrence of the parameter value extract it and then assign it F s ip h Refresh Cl Web Admin Var inder to parameters of succeeding requests elp earch Retres ose Potentially dynamic exchanged Session Parameters Extract of all HTTP Requests Find parameter values with min 4 characters where min 0 of all characters are in ASCII HEX range O F Include File Paths Find Dynamical handling of parameter VIEVVSTATE successfully accomplished Statusi ins step22 ddStartYears 2000 Status1 ins_step22 btDepotValue 25000 Q Statust ins_step22 ddDepotYears 2005 W type tt31 54 Form Param 54 Form Param 54 Form Param 55 CGI Param Ar gt 57 Form Param VIEWSTATE_3 _VIEWSTATE 57 Form Param dD wtMTgOMjMwNDc403G802w8aTwxPjs O2wS8dDw7bDxpPDE O2k8NT47PjtsPHQ8O Q Status1 ins_step31 ddPremiumindex kein CTS Var Name Parameter Name Recorded Value A 1 CGI Param Q type 163283 amp 18 COGlParam W act first 2 gt 18 CGIParam levid Y levid 94153 a 30 CGI Param W agenda demand amp 33 CGI Param id 451047 amp 41 CGlParam
89. 2 HELP AL AAA TET 64 7 8 Overview of most commonly used Extract and Assign Options ccccceeccceeccceeeceuceceeeceuceceueecaueceusessuessueessueesuessueessueeseeessuseseeessueessueeseeessneens 65 7 9 Directly Defined Variables stand alone VariableS cccccsccccscccesecceeeceuceceeeccueesecesseeseueessueesseesesessueesseesaueeseeeseueesueesseessueessesseueessussseeeseatess 66 GAO S2EE UNIS UM E UU rate leinlanedsertaee ieee laneiserrae E O 67 8 Generaling Load SS PRO OMS a o TESTEN 69 81 Load Test Programs witht Dependent FIIeS oer ettet eo eoe egauntanadtuedadesquehasedatennt ee 74 9 Executing Lodd SSI diet cic OT D ooo ee 75 Pal Sta nd EXC ACEN Sl A 80 9 1 1 Real Time Job Stalisties EXeCsAGeNt JODS a a ee MM a e O a e tI D Mos Dea i To ed 81 9 1 1 1 Response Time Overview Diagrams Real Time ccccscccsscccssecceeeceucecuecceeeceuseseeeaeeeseueessueseueessueeseeessueeseeeesaeessusesseesseeessussseeesaaeess 84 9 1 1 2 URL Response Time Diagram Real Time chase vec e a 86 9 1 1 3 Err r Overview Diagrams Real TIME se uico urs en 88 9 1 1 4 statistical Overview Diagrams Real Time a EHRE 90 9 1 1 5 Real dime OTI e A Cone me annotatio an cet bun eC du aD uM ME EM aM M E AM EE Ad EE E 91 94 2 Lodno me stalstes mc ies 93 92 Sarg el Eee RE ERNEUT ee neh Bat as ee re ee ee en 94 9 2 1 plealdime Gl sterJob alls ilCS ltda dada ee tddi 95 9 2 2 Loading the Statistics File of Cluster Jobs nite H
90. 2 Reviewing the Automatically Applied Content Test Avoid executing load tests without controlling the received content of URL calls by comparing them to the originally recorded data Many errors from web server applications are embedded inside valid HTTP 200 responses Therefore the content of the responses must be also be checked to detect content errors under load For this reason Proxy Sniffer examines the content of all recorded URL calls and automatically applies a content check per each URL call using a heuristic algorithm This algorithm performs content checks by searching for an ASCII text string inside the received content however if this seems to be impossible or if this doesn t seem to make sense the received content is checked by its size content length instead of by searching an ASCII text string After clicking on the View icon inside the Web Admin GUI main menu the display of the recorded web surfing session changes and the automatically applied content test methods are displayed for the URL calls at right Binary data such as images are checked by their size this is fast and works well in most cases You should always review content tests where an ASCII text fragment is searched for inside HTML data web pages and check whether the pre configured search text makes sense x E 0 Page 1 Web Load and Stress Testing Tool Proxy Sniffer user s think time 5 seconds 135 Test HTTP Request HTTP Response Resnonse
91. 24 0 00 18 0 00 12 0 00 06 0 00 Remote Exec Agent Job 188 13 Oct 2669 26 29 18 18 29 min elapsed 18 29 min 0 24 0 00 18 0 00 12 0 00 06 0 00 Response Time Overview Real Time Remote Exec Agent Job 188 e 191 ims 11 3ms All Rights Reserved English Edition Description displays during the load test at real time a diagram per web page about the measured response times Please consider that maybe only a fraction of the response times is shown depending on the Additional Sampling Rate per Page Call which was selected when the load test was started For example only every fifth response time is shown if the Additional Sampling Rate per Page Call was set to 20 Input Fields e Response Time drop down list Allows to select the period from the current time back to the past within the response times are shown in the diagrams e Time Bars drop down list Allows to select if the bars inside the diagrams are shown as average values or as max values Please note that there if only a difference between the max values and the average values if multiple measured samples of the response time fall inside the same pixel inside the same displayed bar Page 84 171 Apica ProxySniffer V5 0 User s Guide English Edition Max Value Max Value Average Value Average Value i 1Pixel 1Pixel width width The tables at the right side of the diagrams contain
92. 30 bytes no failure action 40 30 amp our license price and renting the Amazon Cloud server 46 46 amp To learn more about Apica s web performance services Save Load Test Program Overwrite amp Compile English Edition On the left hand side you can choose the Project Navigator directory in which the load test program will be stored The current directory is marked in blue You can also create new subdirectories by clicking on the Ei icon On the right hand side the title Display Load Test Program is shown This allows you to view examine the automatically generated load test program before it is stored Directly below this the Response Verification Summary is shown This contains an extract of the automatically applied content test configuration The overview contains only URLs a whose received content is verified by a search string text fragment or b whose content test configuration was manually modified for example a disabled content test configuration for a particular GIF image because it was a rotating banner advertisement Here you can again modify the content test configuration by clicking on the corresponding magnifier icons It is recommended that you save the web session after you have made any changes This can be done by clicking on the 3 icon Enable the checkbox Overwrite amp Compile and then click on the Save Load Test Program button to store and compile the automatically generated
93. 4 660 19 ms 17 ms 19871 bytes GET https Vet testix pum ch A43refipublicicc jsiwtjs B 4 660 0 15 ms 13ms 3 840 bytes GET https ef testix post ch 443 efipublicicc js ef base js Bl 4 660 0 12 ms 12 ms 899 bytes GET https eftestix post ch 443 ef publiciccicss styles css 4 660 0 30 ms 24ms 57517 bytes GET https Meftestixpostch 443 ef publiciccicss elements css mu 4 660 0 23 ms 18ms 25 715 bytes GET https eftestixpost ch 443 ef publiciccicssiframework css MEN 4 660 0 19 ms 18ms 15127 bytes GET https ettestix post ch 443 efipubliciccicss elements_form css na 4 660 0 15 ms 12 ms 2 484 bytes GET https feftestix post ch 443 ef publiciccicss styles_efcss na 4 660 0 12 ms 11 ms 865 bytes GET https Vet testix post ch 443 effpublicicc pics background gif EEI 4 660 D 16 ms 13ms 6 532 bytes GET https ef testix post ch 443 etipublic cc picsfimg_pf_logo_de jpq 116 4 660 0 14 ms 11 ms 955 bytes GET https feftestix post ch 443 ef publicicc ipicsidoc_bd qif nz 4 660 15ms 12ms 1 368 bytes GET https et testix post ch 443 efipublic cc pics img claim de gif ell 4 660 14ms 12ms 4437 bytes Jfet testi iciccipicsii i na 4 660 16 ms 12 ms 4 156 bytes GET https et testix pu ch ada pna 20 4 560 14 ms 11 ms 1 656 bytes GET https eftestixpost ch 443 favicon ico Total 4660 728 ms 1 364 ms 229 624 bytes 20 URLs mm Oe cC gnl Page 2 login maske user s think time 15 0 seconds
94. 4 ms 31 ms 218 ms T Oms 3910 ms 16 047 ms 10 843 kbytes sec p ms Ims 42ms 41 ms 16ms Oms 102ms 2187 ms 51 000 khytesisec Test consecutively numbered Clicking on a number displays the 3 ms 1ms 108 ms 17 ms ms Oms 128 ms 1 485 ms 4 812 kbytes sec 4 MS 1 ms 42 ms O ms MS O ms 43 ms 969 ms 15 558 kbytes sec U RL overview Menu 5 MS 1ms 121 ms 1ms D ms 0 ms 124ms 1469 ms 14 460 kbytes sec i 8 ms 0 ms 80 ms 4 ms ms 0ms 65ms 1110ms 12 492 kbytesisec Av Net Con average time per URL call required to open a Loue L E Se ceres d tes lem AL SUS ESE network connection to the web server before HTTP data are send 8 ms 3ms 55 ms 73ms 0 ms Oms 133ms 1 672 ms 57 714 kbytesisec i 10 ms 2ms 48 ms 24 ms lt TO MEETS EST PANES supported by the web server this time is only measured for some Total Oms 4 833ms 30 907 ms 69 291 kbytes sec URL calls and not on all because the network connections have 11 Page 2 Login Form user s think time 3 0 seconds been reused Test gt Av Net Con gt Av Req Trm gt Av Wait Av Header Rew Av Content Rev Min Time Av Time Max Time Av Throughput Pra srme 15ms El 334 ms 4912 ms m 6 ma 5247 me 14236 m6 ERE E EIER NES Av Req Trm average time per URL call to transmit the HTTP Total 15ms 13003ms 67875ms 0303 kbytes sec request data to the web server measured after the networ
95. 5 22 55 18 iz res unprotected 16 Jul 2003 20 04 51 frei fre unprotected 03 Apr 2002 18 35 51 ri unprotected 14 May 2002 18 29 31 A E unprotected 16 May 2005 21 22 26 A unprotected 24 Apr 2005 20 30 44 Il E unprotected 29 Mar 2005 15 40 07 iz java lang IllegalArgumentException null source at java util EventObject Event bject java 32 at org jboss pool PoolEvent PoolEvent java 49 at org jboss pool jdbc xa wrapper XACLientConnection setLastUsed XaClientConnection je at org jboss pool jdbc StatementInPool setLastUsed StatementInPool java 39 at org jboss pool jdbc ResultSetInPool setLastUsed ResultSetInPool java 43 at org jboss pool jdbc ResultSetInPool next ResultSetInPool java 877 at dfischer db DbLib count DbLib java 119 at dfischer db PrxFile directorylsEmpty PrxFile java 221 at dfischer webadmin WebMainMenu execute WebMainMenu java 2032 at dfischer appserver GenericPrxauthHttpServlet execute GenericPrx uthHttpServlet jav at dfischer appserver GenericPrxauthServlet execute GenericPrxAuthServlet java 9 at dfischer appserver GenericPrxServlet doGet GenericPrxServlet java 91 at dfischer appserver GenericPrxServlet doPost GenericPrxServlet java 43 unprotected 24 Apr 2003 01 26 03 at javax servlet http HttpServlet service HttpServlet java 760 at javax servlet http HttpServlet service HttpServlet java 853 at org apache tomcat
96. 563 563 mmm B1 AV Time AV Page Size Passed 130 015 bytes 64502 bytes 27 201 bytes 22 529 bytes 135691 bytes Passed Failed A 2 1 1 A a 24 0 100 553 BI 2 550 10 534 TA 484 2941 amp 452 20 X Auto Refresh Apply Refresh GET http 1 32 16 4 5 80 format czs GET http 1 92 16 4580 XX RR git GET http 1 32 16 4 5 50 flag Serman git GET http 92 16 4 5 60 AlagEngland cif 3 p AAA c GET http 32 15 45 80 arrovv red 12x49 git GET http 7 32 15 45 G0 pdf icon 16x16 gif GET http 32 15 45 80 screenshots 1 p gif GET http 4192 16 4 5 80 screenshots 3 p gif GET http 011 32 15 4 S S0 responsetime git Tr OET SA 47 04 Fm ee gt O 2012 2013 Ingenieurb ro David Fischer AG Switzerland Disable Detailed Statistic LILI 4 All Rights Reserved By clicking on the magnifier icon of a page the most relevant measured values of the URLs are shown for the selected page Using this menu you can also display and analyze error snapshots by clicking on the magnifier icon next to the failure counter see Chapter 10 2 In this way you can begin analyzing errors immediately as they occur during the running load test By clicking on a URL the corresponding URL Response Time Diagram is shown see chapter 9 1 1 2 All of these detail data including all error data are also stored inside the final result file prxres which can be accessed when the load t
97. 564 rra e hbri 1 icer Cone De dure String Hot Found Page RI Vue Lond and Bra Tetra T 5 2 G T FR ker ars hi A ie Cordel Tesi Fabre rr Hot Ford Page de Lond and Saeed Testing 19 ET GE 180 armo es hien oy BEL Conterk Terri Fabs rr Ho Found me FI ls Loud end Sir Tarima T 15 ATRAS 154 Seeder nbi Input Fields e All failed URL Calls effects that all errors about failed URL calls are shown non fatal and fatal errors e Session Failures only effects that only fatal errors about failed URL calls are shown session failures Error Snapshot Memory used By clicking on the plus sign you can increase the amount of memory available to store error snapshots Please note when the memory is already 5096 or more used no additional error snapshots for non fatal errors are captured This means that increasing the memory may also re enable capturing for non fatal errors ne Error Overview Real Time Local Exec Agent Job 1312 E Fan Diagrams ho Firge rear Sog tart EB Lai Ero nee 0 A ages EL Cars C Saxsien Fabre gne eir Gr emer WG used Max Error Snapshot Memory und 18d Emi ph Men 1 wi Ire eaae Error enapzshot Memory A 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 89 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 1 1 4 Statistical Overview Diagrams Real Time gt PRX Job 188 Statistical Overview Mozilla Firefox Description displays statistical i http
98. 70 171 Apica ProxySniffer V5 0 User s Guide English Edition Character Encoding defines which character set is used to search for strings within the received content and for data read from Input Files Usually you can use the default option OS Default which means that the default character set of the local operating system is used however if you execute remote tests on other operating systems different from your local OS Windows gt Unix it is recommended that you use the character set ISO 8859 1 to avoid problems with special characters such as umlauts HTTP Protocol Version usually the HTTP protocol version 1 1 should be used for load tests This protocol version is supported by all newer web browser and web server products and allows the re use of network connections over several URL calls HTTP keep alive option If HTTP protocol version 1 0 is chosen the network connections cannot be re used and a new network connection is opened and closed for each URL call Allow Keep Alive the re use of network connections can also be disabled for HTTP protocol version 1 1 using this option however this is not recommended Strip Referer Header Field the HTTP referer header field is not commonly used by web applications and therefore often dropped by local internet security tools Enabling this option reduces the data transfer and makes the load test program smaller Accept Header Field the HTTP accept header field is not common
99. 901 bytes no failure action Ite c lo Je J de dem dE Odes de l npe ko s x x x x x x x x x E x x x EN x 15 15 Page 2 Web Load and Stress Testing Tool Proxy Sniffer usege think time 5 seconds 35 Item Test HTTP Request HTTP Response MiEnonse Verification x18 186 I cet httpi 192 15 4 5 purchase en html 200 TEXT HTMLB The number of virtual users is counted locally on the Exec Agent x ds ih iB GET htto 4192 16 4 5 StarterPackage gif 200 IMAGE GIF amp size 5 15254 bytes no failure action The content test configuration can be modified by clicking on the magnifier icon 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 20 171 Apica ProxySniffer V5 0 User s Guide English Edition During the execution of a load test program the HTTP response code and the received MIME type of each URL call is always compared with the originally recorded response from the web surfing session if not disabled manually The response verification menu allows the specification of how received content is to be tested PRX Response Verification amp Failure Action for Item 16 Test Index 16 Mozilla Firefox F http 127 0 0 1 7990 dfischerfwebadmininterface PopupModifyContentTestWeblet id 12537458467818 displayIndex 168calledFromMainRecorderlframe 1 p ae HTTP Response Verification amp Failure Action for Item 16
100. 990 dfischer webadmininterface PopupDirectoryNavigatorWeblet selectDir QzpcRG9rdw 1lbnRIIHYuZCBFawWSzdGYsbHYuZ2WuxXG1 1dG9uz1xQcm94evNuaWZmzx Jc TXIUZXNOcwG Ly i Proxy Sniffer e bi E hs x 3 Web Admin Project Navigator Help Setup Network Jobs Analyse Refresh Close C Dokumente und Einstellungen mutong ProxySniffer MyTests NOA ME Plugins ReportTemplates ScriptExamples i Trash pr PRX Execute Load Test Mozilla Firefox FU httpi 127 0 0 1 7990 dfischer webadmininterf ace PopupDirectoryNavigatorStartLoadTestWeblet filePathB64 QzpcRG9rdW1IbnRIIHVuz CBFaWSzdGVsl gt 9 eee H i H H gt rox niffer A P Ine Project Navigator Execute Load Test So e dica Acl Start Job 1331 on Local Exec Agent EO JS merci dices Execute Load Test Job Test01 Job State configured Test Test01 Execute Test from Test Arguments 3 u10 d 240 t 60 sdelay 200 maxloops 0 sampling 15 percpage 100 percurl 20 maxerrmem 20 nolog annotation First Test Run Number of Concurrent Users Concurrent Users 10 Load Test Duration Planned Test Duration 4 00 min Max Loops per User unlimiig Max Loops per User unlimited v Startup Delay per User 200 w Milliseconds i Start Load Test Job Display Real Time Statistic Max Network Bandwidth per User unlimited Downlink unt ed Uplink Request Timeout per URL 60 w Seconds 9 Schedule Job 1331 for Day Time
101. AG Switzerland All Rights Reserved Page 95 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 2 2 Loading the Statistics File of Cluster Jobs The statistics result file of a cluster job contains the consolidated merged measurements for all cluster members The calculations for merging the results are extensive therefore it may take up to 60 seconds for the result file to be shown The individual measurements of the Exec Agents are embedded separately inside the same consolidated result file http 127 0 0 1 7990 Proxy Sniffer Cluster Job 6 Execute Load Test Mozilla Firefox ni x The consolidated statistics result file is marked with a blue background and eee gee proxy sniffer Cluster Job 6 Execute Load Test a x is already selected for you Test01 Test C leted Cluster Cluster 1 i q nudi By clicking on the magnifier icon you have access to the out and err File Size Modified files of the corresponding Exec Agent jobs Test01 class 34 454 16 Jun 2006 01 15 07 userAccounts txt 93 16 Jun 2006 01 15 07 4 Exec AgentLocal Exec Agent 43 users 16 Jun 2006 15 48 47 4 Exec AgentTest PC II 64 users 16 Jun 2006 15 48 47 A Exec Agent Sun Fire V240 33 users 16 Jun 2006 15 48 47 job_6 in 287 16 Jun 2006 01 15 07 job_6 status 229 16 Jun 2006 15 48 48 job_6 ziplist 32 16 Jun 2006 01 15 07 m v Acquire S
102. After a web surfing session has been recorded the load test program can be generated see chapter 8 However it is often desirable or even required that the recorded web surfing session must first be edited Some possible cases are o The web application contains HTML form based authentication and it is required that each user use an own username and password to login into the web application see example in chapter 7 2 o You wish to make a parameter of an URL call variable in order to set the value of the parameter each time before starting the load test For example a booking date of a flight see example in chapter 7 3 o The recorded session contains dynamically exchanged session parameters which must be extracted at run time from the web pages and then assigned to succeeding URL calls in order that the load test program runs successfully see chapter 7 4 All of these tasks and many more can be performed by using the central variable handler menu called Var Handler which manages all dynamically applied modifications to web surfing sessions The process involves two steps 1 First a variable must be defined or extracted and then 2 The variable must be assigned In other words a variable must first be extracted before it can be assigned however some of the most commonly used dialogs also support making automatic and or global assignments The process of extracting variables is completely independent from assignment thus many co
103. B 122 10 1 11 DDISgrar m oGompisiedioO Ssuateiitiaieu Code ee it tada 122 101 12 Diagram LET desto ie Role M X 123 TOS TS Diagram Ne Work INFO Han DUE essetis NE 123 10 1 14 Diagram HTTP Keep Alive Econ M 124 10145 DiagrameseLGache EMiclenCy 25 oii teo ER ee see ae a a 124 DOSE Diagram Se SslO dimi Sa UT 125 TOA M Bie lA e on 126 10 1 18 Diagram Number Or Emors Del Padierna ira drid a 127 102119 Diagram NUMBER Ol Enor PEU ion 127 10 2 AEROSOL A 128 103 Load GUNS DATOS do 133 10 81 Overall LO ad GUN ES A A A A A AA AAA ac ones 135 10 32 Response Mme per T Age sintieran 136 1059 Response TIME per bt sui pot ISPN a ER i pei ISPN aaa uranio 137 10 94 ES AA m 138 10 4 GOMpPAlSOn DIAGN a cisnes 139 TO RESPONSE MIMO c Un eeiam Et imp Ruin ERE ime cacaos caes 139 1072 P HO NOCO V ON estas estu aide diim iau editio ditm evite tt tada 140 DU SCR ep I m PUu 140 11 Distributed Load Tests Architecture and COntigilballQIsss ee Oe Puce icut he Do uu t ndo soe 141 11 1 Configuring Additional Load Releasing Systems Exec Agents ccccccscccceececeeeeeceneeseeeeeseeeseeeeeseusesseeeeseueeseaeeesaueeseaeessauseseueesseeeesausesseeeesees 142 T1 2 Gontguting boad Heleasing GIBSIBIS ans eee e 143 1 1 3 Staino Distributed Load MIU 144 12 Using Multiple Client IP Addresses per Load Releasing
104. ER a RUE 96 A A yq 5 98 9 3 1 Bola E o OS 99 9 4 SCN LOA ES ct aS ee mE 101 9 5 Rerun of Load Tests Jobs Job Templates irepl orbe aka onan bandola canaletas bei 101 920 Project NAVIG ALON RETE ON 104 9 6 1 Configuration of the Project Navigator Main Directory cccccccccsecccceeeeceeeeeeeeeseeeseeeeeseeeseeeeeseeeeseaeeeseueessaeesseueeseaeessaeeeseusessneeeseneetaneeeas 106 ea ca e Y odd AAA O A MSMDMMIDEMI Me EIM 107 10 ANALYZING Measurement Resulta a ina 108 A oo nn M 109 10 141 Test Scotia m 110 10 1 2 Diagram Response Time Der Pages 111 10 1 3 Results per URE Call OVGIWIOW usada alla E aan 112 10 1 3 1 Response Content Throughput In Depth Measurement of HTTP S Response Streams cccoccccocnccncnconcccncnconcnnnnnnnncncnncnnnnncnncnnnnonos 113 TOTA Results per BELA BIBIT a es da ts o e o E 117 10 55 Diagram Response TIMePSrcenuleS asin naaa cada dla adi 118 1046 Diagram TOD Hine Consuimlitig URES accutane ad iii 119 10 1 7 Diagram Concurrent USAS e e e ee ie eo 120 10 18 Diagram Sess cm E 121 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 3 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 9 Diagram Web Transaccional aiii 121 10 01 10 Diagram Users Waiting Tor ES ON a ae E
105. FF8AADFEBCEDDABS gt lt j lii Done 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 61 171 Apica ProxySniffer V5 0 User s Guide 7 6 Replacing Text Patterns English Edition In rare cases the name of an HTTP request parameter is variable instead of the parameter value being variable Even rarer are cases where a file path of a URL call contains variable parts You can handle such cases as follows 1 Use the text pattern based variable extractor as described in the previous sub chapter 2 At the last step use the checkbox Assign var to all matching request file and request content patterns with same recorded value instead of Assign var to all request parameter with same recorded value There are also other rare cases in which a text pattern must be extracted from an HTTP response header because a variable HTTP redirection occurs on which only a part of the URL file path or a part of a CGI parameter is variable This is also supported if two extractor icons are present you simply use the Second one HTTP Response Header 1 HTTPM 1 302 Found E Date Fri 07 Apr 2006 14 11 01 GMT L 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Extract Var from Text Line Pattern Text Line contains text Search Text ima id atiReportingcare Extract Var an sestedine y Token Delimiters Po Extract Taken Nr 7 El
106. Ingenieurb ro David Fischer AG A Company of the Apica Group www proxy sniffer com 9 Apica ProxySniffer Apica ProxySniffer User s Guide Version 5 0 English Edition O 2012 2013 Ingenieurb ro David Fischer AG F May 15 2013 All Rights Reserved Apica ProxySniffer V5 0 User s Guide English Edition Table of Contents E MOU ION ee idilio 6 1 1 Mena and Bi Is dferciie Rad RTT en es et ee inet ie ee eis nel 6 2 Recording Web Surfing Sessions without using the Firefox Recoding Extension u024s002n0nenn none nenn onnnn nenne nennnnnne nennen nnne nnne nnne rna rn nnn nns 9 21 Recordio the First Web Page E EU m TN 9 2 2 Hecorama subsequent WEB Pages une Ri MEN MEME I EE EE serie 11 3 Further Hints for Recording Web Surfing S SSIONS c0cccc ccecccseceseeeecceeeecsecescseeccueecscdacsceesecceueseuaeseseeeccsucceusessodeeccucenoseesceareceusescseesceeveccueetoeeeees 12 3 1 1 Support of Technical Client Programs and Web Services SOAP and XML Data Communication over HT TP S 0024002240 220000 12 3 1 2 Proxy Recorder Settings and GUI Settings Personal Settings Menu 2u00222002220002000 nenne nnnnn nenne nnnnn nenne nnnnn nenne nenne nnnnne nenne nnnnnenn 13 3 1 2 1 Connect to NexcPIOXy Pix Recorder n 232 5 re er ee a II UII re a cit DM o 14 3 1 2 2 ATTES Settings PrO CU o 14 3 1 2 3 HTTPS Client Certificate based Authentication PKCS 12 Files Proxy Recorder
107. JOB SFR exceeded e At the End of a Job If the Session Failure Rate IS Atthe End of a Job if the Average Response Time per Page is greater than 5 seconds prx exceeded Y ENDJOB RTP exceeded Atthe End of a Job ifthe Average Response Time ofthe Slowest Page is greater than 10 seconds Y prx exceeded Y ENDJOB RTS exceeded greater than a threshold in percent e At the End of a Job If the Average Response Time Display Alert Configuration of AII Exec Agents Y Copy local Alert Configuration to All Exec Agents v copy Delete Alert Configuration on All Exec Agents V Delete per Page is greater than a threshold in seconds e Atthe End of a Job If the Average Response Time of the Slowest Page is greater than a threshold in seconds The values for periodically checked alert conditions are calculated from the measurements collected within one interval Repeated alerts are suppressed A cancel notification is released if the measurement is later less than the threshold Done 2 he values for periodically checked alert conditions are calculated from the measurements collected within one interval Repeated alerts are suppressed A cancel notification is released if the measurement is later less than the threshold 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 149 171 Apica ProxySniffer V5 0 User s Guide Engl
108. L Call captures the measured response time of a URL each time when a simulated user calls a URL event based sampling Used to display the response time diagrams at real time as well as in the Analyse Load Test Details menu For endurance tests over several hours it is strongly recommended that the sampling rate for URL calls is disabled or set to 196 or 296 For shorter tests 10096 sampling rate is recommended In addition to capturing the response time of the URL calls further data can be captured by using one of the following Add options o recommended no additional data are captured o Performance Details per Call additionally collects the network connect time the request transmit time the response header wait time the response header receive time and the response content receive time of the URL calls o Request Headers additionally collects the request headers of the URL calls o Request Content Form Data additionally collects the request content form data of the URL calls o Req Headers amp Content additionally collects the request headers and request content form data of the URL calls o Response Headers additionally collects the response headers of the URL calls o Resp Headers amp Content additionally collects the response headers and the response content of the URL calls 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 76 171 Apica ProxySniffer V5 0 User s Guide English Ed
109. Load Test Start Date Users TestDuration Web Trans Sess Failures Net Throughput Annotation response times of several test runs This is commonly amp ProySniferd 06 May 2006 19 45 08 10 215min 26 15 trisec 0 00 3 64 MBitisec Erster Test used to visualize tuning efforts that is before and 3 amp ProxySniffer01 06 May 2006 19 49 46 20 217 min 51 50trisec 0 00 7 16 MBitisec ZQeiter Test after tuning of the web server In contrast to load curve diagrams these comparison of test runs can be ma d e wi th th e same nu mb ero f user S h oweve r th i S S Hint execute the same load test program severa times with a different number of concurrent users and compare the measured results Click on the magnifier for details not mandatory You can compare any test runs as long as all test runs have used the same name for the web pages same text for all page break comments 10 4 1 Response Time PRX Analyse Load Tests Test Result Comparison Mozilla Firefox You can generate a PDF report in the upper right corner of the window The diagram in the upper part of the window shows the response time comparison of all web pages Response Time Comparison per Page Average Page 1 6 The diagram in the lower part of the window shows the response time comparison of the URLs within a particular web page by default the first web page Clicking on the diagram bars in the upper diagram displays a comparison of URL calls for any o
110. ML and SOAP Request Data Java Classname 3 Content Test Algorithm gt 4096 Bytes v URL Execution G Serial Executed 18 Parallel Executed 82 Threads p User B Switch to Serial Exec swit Parallel Exec required enter a simple classname for the load test program with no path and no file extension HTTP Protocol Options E HTTP Protocol Version Strip Referer Header Field 1 1 vw Allow Keep Alive Strip Accept Header Field to Load Test over HTTP S Proxy Apply next proxy configuration from personal settings HTTP SSL Authentication Options EI Basic Authentication O Apply individual Basic Authentication per user from input file basicauth bd O O Apply individual Digest Authentication per user from input file digestauth bd Password Digest Authentication use common Username NTLM Authentication O use common NTLM account from Personal Settings menu v Kerberos Authentication CO use common Kerberos account from Personal Settings menu v HTTPS Client Certificates O apply individual PKCS 1 2 certificate per user from input file pkcs1 2auth bi v Program Description recommended wii be displayed as hint in Project Navigator 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 24 171 Apica ProxySniffer V5 0 User s Guide English Edition Depending if all URLs of a page are executed in serial order or som
111. NS stack on the load generators will communicate with the default configured DNS servers of the operating system but without considering the hosts file o mtpu number Allows to configure how many threads per simulated user are used to process URLs in parallel simultaneously Note this value applies only for URLs which have been configured to be executed in parallel 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 78 171 Apica ProxySniffer V5 0 User s Guide English Edition O nosdelayCluster Effects for Cluster Jobs that the Startup Delay per User is applied per Exec Agent Job instead of applying it overall simulated users of the Cluster Job Thus a faster ramp up of load can be achieved setuseragent lt text gt Replaces the recorded value of the HTTP request header field User Agent with a new value The new value is applied for all executed URL calls collect measuring agent host gt port lt measuring agent host gt port example collect measuringhost1 measuringhost2 Forces the load test program to collect additional data from external measuring agents Such data contain for example system operating values like CPU usage and memory consumption of the Web server and the database server ssicache lt seconds gt Alters the timeout of the user related SSL session cache The default value is 300 seconds A value of 0 zero indicates that the cache is disabled sscreset
112. New Cluster Name Cluster 1 Cluster A 3 Cluster 1 Load Factor Members 2 00 2 Cluster Cluster 1 Local Exec Agent 1 00 50 096 amp TestPC II 11 00 50 0 amp 3 Sun Fire 240 yl Web Admin GUI Exec Agent on Inter Exec Agent Cluster Exec Agent 1 pi w P 4 QR c 7 B u HTTPS V M P Wu a Web Admin GUI Y EN 40 Exec Agent 2 I rn Load Test l N _ i L a 14 gt Y Load Test EN Exec Agen Web Applic f i oad Tes d Done To get a suggestion for the load factor of a particular Exec Agent you can click on the m icon within the list of all defined Exec Agents It is however recommended that you click several times on the B icon in order to get a stable result Even so this result may not accurately reflect the power of the computer system 2012 2013 Ingenieurb ro David Fischer AG Switzerland Exec Agent Load Factor 1 13 After an arbitrary name of the cluster has been entered the cluster members Exec Agents can be added to the cluster by clicking on the blue arrows in the list of Available Exec Agents By clicking on the magnifier icon of a cluster member the Load Factor of this member can be modified The load factor controls how many users will be assigned to this cluster member when the load test is distributed across the cluster members The load factor by itself is an abstract value meaning that the distribution of the users is made based on the ratio between the loa
113. P Addresses to an Exec Agent On the local system where the Web Admin GUI is running assigning multiple IP addresses to the local Exec Agent can be done by clicking on the Setup icon in the Project Navigator Inside the setup menu you must enter all IP addresses in the input field Local Exec Agent IP Addresses separated by comma characters Alternatively there is an Auto Detect checkbox available which assigns all IP addresses configured at the operating system level automatically ee ee ee a AN loj x 6 4 Eh e x Help Setup Network Jobs Analyse Refresh Close nang On external Exec Agents where no Web Admin GUI is available you can assign the IP addresses to the Exec Agent by editing the file javaSetup dat with a text editor javaOptions javaVirtuallpAddresses 192 16 4 5 192 16 4 6 192 16 4 7 javaEditor The file javaSetup dat is located inside the Proxy Sniffer installation Directory Important Note when you start a load test you must use the additional option multihomed to specify that multiple IP addresses are to be used by the Exec Agents FOL ULL di dilipliiit ALE Hal L Age a dli uu aul Add multihorned SSL vziv3ITLS Additional Sampling Rate per URL Call Debug Options Additional Options SS2s Done trecommended will be displayed as hint in Project Navigator 2012 2013 Ingenieurb ro David Fischer AG Switzerland PRX Project Navigator Java
114. PRO 100 VE Network Conne this capability Otherwise unu need ta ask your network administrator Far IP address SENE the appropriate IP settings 132 15 4 35 255 255 255 0 Th ti le 1932 16 4 37 255 255 255 0 Dr AST al Obtain an IP address automatically 132 15 4 38 um ntemet Protocol TCP ZIF me IF address Subnet mask 255 255 255 0 SINE US BLUES Gateway Metric Description Transmission Control Protocol nterret Protocol The default ie ees PENAL wide area network protocol that provides communication Use the following DNS server addresses across diverse interconnected networks Hi I fm Preferred ONS server Show icon in notification area when connected E Automatic metric Alternate DNS server a Notify me when this connection has limited ar na cannectivity a IER 1 12 1 1 2 Unix like Systems You can configure multiple virtual IP addresses for the same network interface by executing the ifconfig command The specific arguments for the ifconfig command depend on the Unix variant and operating system version Linux Solaris Mac OS X Please refer to your operating system manual to find out how to define virtual IP addresses on your system 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 146 171 Apica ProxySniffer V5 0 User s Guide English Edition 12 1 2 Step 2 Assigning Multiple I
115. Page Diagram Mur Test Scenario Test Start Dale 18 Oct 2008 15 54 57 test aborted by remote comma c Load Test Program Testi class Load Source Host dynatest 192 16 4 30 Load Source OS Windows XP Target Host 182 16 4 5 80 fa Applied HTTP Version 1 1 Test Input Parameter Concurrent Users j Planned Test Duration Planned Loops per User unlimited Startup Delay per User 200 millisec Request Timeout per URL 60 sec Statistic Sampling Interval 15 sec Additional Sampling Rate per wib Page Call 10096 Additional Sampling Rate per URL Call 2095 Real Time Comments modify 19 Oct 2009 15 56 45 1 48 min server shutdown 18 Oct 2008 15 58 45 3 48 min server restart Test Sequence 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 92 171 Apica ProxySniffer V5 0 User s Guide 9 1 2 Loading the Statistics File After the load test job has completed the statistic results file is stored in the job directory of the local or remote Exec Agent In order to access this results file you must transfer it back to the local Project Navigator directory from which the load test program was started English Edition Proxy Sniffer Main Menu Mozilla Firefox http 127 0 0 1 7990 Proxy Sniffer Project Navigator Execute Load Test Mozilla Firefox File Edit View Go Bookmarks Tools Help eee a 2 x A aoe Proxy Sniffe
116. Project C Project D EJ Day2 ProjectPlugins Release1 DO DO LOTO DOT _ 8 Si lt 0 Nu 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 104 171 Apica ProxySniffer V5 0 User s Guide http 127 0 0 1 7990 PRX Project Navigator Windows Internet Explorer eee qe Proxy Sniffer Web Admin Help Project Navigator Setup Network Jobs Analyse Refresh Close Haag E ProxySniffer MyTests Project D PerformanceTuningDay Day2 File AuthenticationUseCase java renamed to Authentication UseCase2 java source code successfully readjusted MyTests Plugins Project A Project B LEJ EnduranceTest Project C Project D PerformanceTuningDay L ProjectPlugins Release1 http 127 0 0 1 7990 PRX Project Navigator Windows Internet Explorer gt i LE 3 199 des 21e eee gee Proxy Sniffer Web Admin Help Setup Network Jobs Analyse Refresh Close Project Navigator E ProxySniffer MyTests Project D PerformanceTuningDay Day2 gt Delete selected Files no abort MyTests Plugins Project A Project B LO EnduranceTest Project C Project D PerformanceTuningDay L ProjectPlugins Release http 127 0 0 1 7990 selectDir CRTpcUHJveHITbmlmZmVyXE15VGVzdHNCUHJvamVjdF9EXFBlcemzZvcm1hbmNIVH Windows Internet Explorer Help Proxy Sniffer Web Admin Project Navigator Setup Network Jobs Analyse Refresh Close EsProxySnifferlMyTests Proj
117. RL call will be reported as failed no response from web server and the emulated user will abort the current loop and continue with the next loop Max Error Snapshots limits the maximum number of error snapshots taken during load test execution see chapter 10 2 Either the maximum memory used to store error snapshots can be configured recommended for cluster jobs value overall cluster members or alternatively the maximum number of error snapshots per URL can be configured not recommended for cluster jobs value per Exec Agent Statistic Sampling Interval statistic sampling interval during the load test in seconds interval based sampling Used for time based overall diagrams like for example the measured network throughput If you run a load test over several hours it is required that you increase the statistic sampling interval up to 10 minutes 600 seconds to save memory If the load test runs only some minutes you may decrease the statistic sampling interval Additional Sampling Rate per Page Call captures the measured response time of a web page each time when a simulated user calls a web page event based sampling Used to display the response time diagrams at real time as well as in the Analyse Load Test Details menu For endurance tests over several hours it is strongly recommended that the sampling rate for web pages is set between 1 and 5 For shorter tests 100 sampling rate is recommended Additional Sampling Rate per UR
118. SYSteM ccccsccccsececeeeeeeeeceeeeseeeeeeeeseeeeseeseueeseeeesaeeseueesseesseeseueeseessuseseeeseeeeeeeseeeeaes 145 12 1 1 Step1 Configuring Multiple IP Addresses at the Operating System Level occconncnoncncccnncconcnccncnconnnnnnncnnnnnnnonnrnnnnnnnnnnnnnnnnnnnnnrnnnnnnnnnaninnns 146 Tol e a LM LM a a EAGMU Mu ey ee ee 146 T2 1292 npe SV Ste IIS ea ash Ae Ne rai e oo a tq ELENCO uates ee ae eh 146 12 1 2 Step 2 Assigning Multiple IP Addresses to an Exec Agent cccccceeccceececeeeteeeeeseeeeceeeesseeeeseeeesseeeeseeessaeeeseusessueesseueessaeesseeesesesseeeseas 147 12 2 Sending Email and SMS Alert Notifications during Test Execution lsseessseesssessssssseeeee nnne nnnnn n nnns n sna rn nsn nsns an nsn sre a anres 148 Lr ME om Seele RR Ere 149 T2 2 MWessagedieagliBie5 assise siuiipndSPv ar EUu ns ISP v a pM pel EUN a p ERN pA MERE S cuiu RM IUOS cuu neas arx a a 150 13 A no M UM P EM M LEM MEC EE E AE d aM EE 151 13 1 1 Input Parameter Progress Display and Saving the Scan Result cece cecceeecceeeeceeeeeeeeeeeeeseeeeeeeeseueeseeeseeeeseueesaeeseueesseessueeseueesseesaneesaes 152 13 2 Analyzinarthesse ame METTE 155 CN MEE x cU REC ee ee ee rer 157 13 1 3 Gonverting a Scan Result into a Web Surfing Session Pee i n 162 ta ee o A ee 164 O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 4 171 Apica ProxySniffer V5 0 User s Guide English Edition
119. TP Hin Content dd Cookies the first URL from which the Session i 24 AN HTTP Request el parameter should be extracted Item 23 GET https ef testix post ch ef securehtmlAogin onl kdl login proceed Found in Response Header Line 8 Position 53 proceed jcurz59A46991027551FF8AADFEBCEDDABS Item 24 GET https ef testix post ch ef securehtml onl kdl sess proceed jcur 59AA6991027551FF8AADFEBCEDDABS5 Paste th e val U e Or th e n ame of th e gt Found in Request Header Line 1 Position 47 proceed jcur 59A46991027551FFSAADFEBCEDDABS session param eter into th e In put field Found in Response Content Line 12 Position 61 nz p sess id 59A46991027551FFSAADFEBCEDDABS name logo Found in Response Content Line 13 Position 125 Fp sess id 3D53AA6991027551FF8AADFEBCEDDABS5 26p menu idt and extract It from the response In Found in Response Content Line 13 Position 253 Fp sess id 3D53AA6931027551FF8AADFEBCEDDAB5 amp amp p sess i Found in Response Content Line 13 Position 288 mp p sess id 59AA6991027551FFOAADFEBCEDDABS E wh ich the fi rst OCCU rre nce IS fou nd Found in Response Content Line 15 Position 67 nz p sess id 59AA6921027551FFS8AADFEBCEDDABS name f2 t Item 25 GET https ef testix post ch ef securehtml onl kdl sess logo anz p sess id 59AA6991027551FF8AADFEBCEDDAB5 gt Found in Request Header Line 1 Position 53 nz p sess id 59AA6921027551FFS8AADFEBCEDDABS Found in Response Content Line 121 Position 45 s id value 59AA6921027551
120. TPSprosySnifferCH G2 Feb 2007 20 49 16 User 1 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 124 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 16 Diagram Session Failures Shows the number of failed web surfing session failed loops which occurred during the test run The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Session Failures Total 79 Sampling Interval 15 Seconds number of failed loops failed sessions 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 30 0 20 00 0 22 30 0 25 00 Load Test Test 1 28 Sep 2669 14 24 34 User 1600 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 125 171 Apica ProxySniffer V5 0 User s Guide 10 1 17 Diagram Error Types English Edition Shows a compilation of the most frequently occurring error types Note this basic error information is accurately measured also in case when not enough memory was left to capture error snapshots for all occurred errors Al failed URL Calls Session Failures only Time Profile of Error Types ALL failed URL Calls number of errors per error type M User Specific Test Failed 57 Errors M Content Test Failure String Not Found 16 Errors oros PEE ETEA ee dest reese nears MAN IO M mane HW sneer Seen eer renee M e mop O ok 288 Wrong HTTP Status Received 6 Er
121. Test Index 16 sd x em Sonnac o e T I AN Verify HTTP Status Code 200 OK v OR M L Approve secondary status code for all URLs with same primary status code Verify Content Type TEXT HTML _ Disable Content Type Verification for all URLs with with content type text html v Disable Content Verification L Disable content verification for all URLs with HTTP status code 200 OK AND content type text html v Verify Content by Size O Apply size deviation for all URLs with HTTP status code 2000K v AND contenttype text html Size 38274 Bytes Deviation 5 v Display Web Page Quality String 1 00 The number of virtual users is counted locally on the Exec Agent 1 00 Each virtual user emulates a own web browser in such a way that a web application can not see 1 00 You can use our shop to purchase or rent 1 00 Unused support points will not expire as long a valid upgrade bug fix option remains without 0 96 over all concurrently executed load tests jobs irrespective of whether the load test jobs are 0 96 Each Starter Package contains a single GUI license as well as a single Exec Agent license 0 96 which contains additional new functionalities All of our customers who have a valid 0 96 This means that all employees are sharing the same Exec Agents in parallel 0 96 Depending on your license an Exec Agent limits the number of virtual users Apply Failure Action E
122. Verification 5 GET hitpirt82 15 4 5 200 TEXT HTML 4 professional web load tests and weh stress tests 2 GET http192 16 4 5iformatess 200 TEXTICSS 4 size 5 2970 bytes no failure action 3 GET hite 192 16 4 5000000 gif 200 IMAGE GIF size 5 43 bytes no failure action 4 igi GET httpJ 192 15 4 5 flagGerman gif 200 IMAGE GIF size 596 174 bytes no failure action 5 igi GET http 182 15 4 5 fTagEngland gif 200 IMAGE GIF A size 5 1 220 bytes no failure adtion 6 igi GET http 82 16 4 5 arraw red 12x8 gif 200 IMAGE GIF A size 5 88 bytes nn failure actiofi Hi iB GET http 82 16 4 5 pdf icon 1616 gif 200 IMAGENGIF A size 5 287 bytes no failure acti 8 MB GET hitpirt82 15 4 5 screenshots 1 p gif 200 IMAGE GIF A size 5 7 909 bytes no failure adtion El GET hito1192 16 4 Siscreenshots 3 p gif 200 IMAGE GIF size 5 7 603 bytes no failure adtion m 10 MB GET http 0192 16 4 Siscreenshots 7 paif 200 IMAGE GIF amp size 5 6 481 bytes no failure adtion 11 O GET httni 192 16 4 5 screenshots 8 p uif 200 IMAGE GIF A size 5 21 831 bytes no failure 4ction 12 igi GET http r1827 15 4 5 bullet circle gif 200 IMAGE GIF size 596 161 bytes no failure action 13 Bi GET httpi 182 15 4 5 respansetime gif 200 IMAGE GIF size 596 32 950 bytes no failure action 14 O GET http 92 164 Sevels enjpg 200 IMAGENPEG size 596 41
123. WindoewItemDetails displavIndex 15 E 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 45 171 Apica ProxySniffer V5 0 User s Guide English Edition After the login URL call has been found the variables username and password can be assigned to the form parameters HTTP Request Content by clicking on the corresponding Y icons http 127 0 0 1 7990 Proxy Sniffer URL Details Var Handler Mozilla Firefox Assignment Options s a 9 amp ex Proxy Sniffer y e Project Search Generate Save Web Admin URL Details Var Handler Hep fest gearoh Generate Save Refresh Close Assign from Var select the variable which should be assigned Var value conversion 1 POST Y prxtool servietWebMainMenu HTTP 1 Y LoginFlag 1 SIM assign Var to Parameter e none the value of the variable will be assigned 2 Hast www d fischer com 8080 D Fusername fischer S username unchanged 3 User Agent Mozilla 5 0 Windows U Windows NT d al i l 4 repete codPeduesuu E en e encode the value of the variable Will first be URL 8 Accept Language en us E s encoded and then assigned for example Z rich 4 gt x oft Varvalue conversion none 70 T A HB will be transformed to Z FCrich HB This is the appropriate option when the value of the variable m
124. _ Start Date 27 Nov 2008 23 19 49 User 20 Test Duration 3 08 min File Test01_27NovO _231949_20u preres URL 13 Error 1 Request Timeout expired Page 1 Start Page Error Date 27 Noy 2008 23 21 28 1 39 min after start date Current Thread T000018 URL 13 GET http www proxy sniffer com 80images_en PointOfCollapse gf 2 Request Timeout expired URL Exec Step wait for server response 3 Error Log error expected HTTP status 200 lt gt received 2 Request Timeout expired No Content Type bytes Failed at Wait for Server Response gt Help Error Explanation HTTP Request Header GET http www proxy sniffer com 80 iimages_en PointOfCollapse gif HTTP 1 1 Host www proxy sniffer com User Agent Mozilla 5 0 Windows U Windows NT 5 1 en US rv 1 9 0 4 Gecko 2008102920 Firefow3 0 4 Accept j Accept Language en us en q 0 5 Accept Encoding gzip deflate Accept Charset I80 8858 1 utf 8 q 0 7 q 0 7 Keep Alive 300 Connection Keep Alive Proxy Connection Keep Alive Proxy Authorization Basic ZmlzY2hlcjphbGV4YW5kcmlh Pragma no cache CON Om Cn amp oo ht HTTP Response Header no response header data received HTTP Response Content Inn resnnnse content data receiver t top of page Done Page 138 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 4 Comparison Diagrams Comparison diagrams allow you to compare the
125. able All jr 54 amp HTTP Request Content Paramete Extract a variable from fl a form parameter iulii 1 HTTP 1 1 302 Found t Content Paramete 2 Date Fri 07 Apr 2006 14 11 01 GMT i POST L statu ype ttl amp id 449647 SEE 3 Server Microsoft IIS 5 0 RADIO Tatus1 ins_step1 subtype rbRate 19 amp Text Line Pattern 4 X Powered By ASP NET RADIO E Status1 ins_step1 subtype rbAtonce gt 33 amp HTTP Request URI 6 Location S Jeveltaccountlist11414 RADIO Statust ins_step1 type rbLeben 35 amp HTTP Request URI KI ai ll 38 amp HTTP Request URI Extract a text pattern from the response ES a content HTML or XML data into a variable 3 lbodys none Extract variable from a CGI Extract a text pattern from a parameter of a HTTP redirection HTTP redirection into a variable none U1 level istatus aspx type tt22 amp tid D amp changed False amp id 449647 B5 Extract variable from a CGI parameter of a hyperlink 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved English Edition Page 65 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 9 Directly Defined Variables stand alone Variables Variables are usually defined implicitly by creating Input Files User Input Fields usi
126. able to give you a general hint as to how many users can be emulated by a single load releasing system you will have to experiment We recommend that you first run a load test with only a few users and then estimate how much CPU power in total will be necessary to generate the required load After that you can decide if an Exec Agent Cluster should be used and how may systems need to be part of this cluster Furthermore we recommend that you tune the TCP IP parameters of load releasing systems see Application Reference Manual chapter 5 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 107 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 Analyzing Measurement Results Measurement results can be analyzed using the Analyse Load Tests menu into which the statistics result files can be loaded Loading result files occurs either implicitly during the acquisition of the job statistics result file this file is also stored inside the Project Navigator directory or explicitly by clicking on the corresponding llb icon of a statistics result file within the Project Navigator The loaded data inside the Analyse Load Tests menu are stored inside a volatile memory cache therefore if you delete some results here they will only be removed from the memory cache but the corresponding files inside the Project Navigator will not be deleted Load Test name of the load You can also invoke this menu from th
127. age 9 171 Apica ProxySniffer V5 0 User s Guide English Edition First Web Browser Window Web Admin GUI PRX Main Menu Windows Internet Explorer Afk _ DRM pc c AA A A A AX Favoriten B PRX Main Menu e proxy sniffer Main Menu NS Td A 9 HN gt ih m Web Page Personal Project Load Test Generate Analyse Refresh Web Admin Professional Edition Help Tools Scanner Settings Navigator Jobs oad Test Load Tests Display Es Recorded Items 27 Q El i gt O amp el inser J Recording State STARTED en 38322 esting Resting Rewarding Apply Fiter x 0 0 HO Page 1 Start Page user s think time 0 seconds 0 item Test Offset Position Content Size Time HTTP Request HTTP Response Second Web Browser Window Web Application x 1 oo0scl J 207 bytes 782ms SPGET http www safearea com au 301 Moved Permanently TEXT HTML x 2 a 0 84 sec 4 554 bytes 1344 ms I GET ite ESE ae sessio PEPE 17 TE E ER x 3 3 2 33 sec 201 bytes 843 ms A GET hitpiiveww safearea com auihtmlrportietijournal m x 4 4 2 33 sec 450 bytes 875 ms D GET httpuiwww safearea com auihtmliportletannoun r Y ttp an safe ca com au ka X al x 5 5 2 33sec 54 013 bytes 2 875 ms GET http twww safearea com aulhtmllis barebone js x 6 6 2 33sec 155 bytes 875ms 3 GET htpwww safearea com au htmiithemesiclassi War Favoriten
128. al Helvetica sans serif color FFOO54 4 tit font size 30px font family Verdana Arial Helvetica sans serif text decoration none color FFCC33 5 6 subtit font size 1 2px font family Verdana Arial Helvetica sans serif text decoration none color FFCC33 subtitwhite font size 1 2px font family Verdana Arial Helvetica sans serif text decoration none color FFFFFF T monilinl f tavt edlaenratian nana enlnrswhita 1 ya gt Bur Ce f 2012 2013 Ingenieurb ro David Fischer AG Switzerland Assign Var gt Assign Var to Protocol Host Port m ssign To Recorded Value Protocol http Host 192 16 4 5 C Port 80 Assign from Var hostname Y v Assign var to all requests with same protocol hast and part Assign hostname global var amp User Input Field Add File none Add Field hostname All Rights Reserved Input Fields Assign to whether the variable should be assigned to the protocol http https to the host name or to the TCP IP port In case you want to make more than one of these items variable you must create additional User Input Fields Assign from Var select the variable which was created when the User Input Field was defined Assign var to all requests with same protocol host and port when checked the variable will be assigned to all URL calls which use the sa
129. al number of currently simulated users Users Waiting For Response z of 200 E Uh ee e CP Socket Connect Time ms The time in milliseconds per URL call to open a new network connection to Total Transmitted Bytes 0233041057 the web server e AV Network Throughput Mbit s The total network traffic which is generated by this load test job measured in megabits per second This floating average value is calculated since the load test job was started 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 82 171 Apica ProxySniffer V5 0 User s Guide English Edition e Total Transmitted Bytes The total number of transmitted bytes measured since the load test job was started More actual measurement details are available by clicking on the Detailed Statistic button Especially an overview about the current execution steps of the simulated users is shown Abort Job 4 Progress 0 User Page JT f nat Start Page 1185 ms ISI Page 2 Download 13 me NS Page 2 Support Sms ea Page 4 References ms Fi TA Page 5 About Us 4 ms Page 1 Start Page O sec think time User Test AV Time AV Size 1 1 033 ms 31 911 bytes 2 14ms 3592 bytes gt 10 ms BBS bytes 4 a ms 395 bytes 5 ams 1547 bytes 5 J mz T16 bytes 7 B ms 917 bytes 3 ams 8 542 bytes 3 aims 5 236 bytes 10 11ms 33580 bytes m Tia a 569 abr 566 564 563 563 563 563
130. an be sorted by URL index or by error time Clicking on a magnifier icon displays the detail data of the corresponding error snapshot in the lower part of the window The title in the lower part of the window contains the URL index a consecutive error number relative to the URL and a short summary description of the error Clicking on the Error Explanation displays a hint about why the error was occurred 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 128 171 Apica ProxySniffer V5 0 User s Guide English Edition iz I PRX Error Snapshots Mozilla Firefox Ed E http 11127 0 0 1 7990 dfischer webadmininterface PopupAnalyseloadtestErrorWeblet key c47ce4cF 1 96bb8eade2d309aa5bee0 ltselectAll 18sortByDate 1 T Load Test Result Detail Error Snapshots Sorted by Date amp Time a XK Page Time Offset Date Error Type Cluster Member URL A URL 83 Error Page 13 loeschen 49 19 min 26 Sep 2009 14 33 53 Content TestFailure String Mat Faund z snit POST https def testix past ch 443 ef secure html anl kdl zzvis ez del A URL 51 Error Page 6 1 eingabemaske 9 19 min 28 Sep 2009 14 33 53 User Specific Test Failed z shit3 POST https ef testix past ch 443 efisecurezhtml anl kdl zinl zinl ta plaus 2 LURL 51 Errori Page 6 1 eingabemaske 49 19 min 26 Sep 2009 14 33 53 User Specific Test Failed z shitl POST https Wettesti past ch 44 3 ef secure html anl kel zinl zinl ta plaus
131. andler English Edition The variable handler can be invoked by clicking on any recorded URL call in the main menu At the left side of the window all details of the URL call which change from call to call are displayed On the right side of the window the Variable Handler is displayed and shows a summary of all extracted and assigned variables This right hand side part of the window remains constant static for all URL calls RNT IZ 10 4 IMAGES C iM 7 069 bytes 3 0 92 sec Fer 7 3 Aly Fite 37 SE IT STIUTC TUSTET Preview EU seca nj Ju JU UF 203 ms IM GET http 1 192 16 4 5 imas_ en ScreenShotWebAdmintPreview gif 200 OK IMAGE GIF 2111see ll BB 9977 bytes 219ms MBIGET http 192 16 4 Simalles_en ScreenShotFinalResultPreview gif 200 OK IMAGE GIF 10 1 11sec BN 19774byes 250ms MRGET ntipirt 92164 5 m3 les en ScreenshotRealtimePreview jpg 200 OK IMAGE JPEG Total 1 36 sec 106 944 bytes 10 Requests 78 64 kbytes 9 ic 11 Item Offset Position Content Size Time HTTP Request HTTP Res pnse 12 0 00sec NT 2193 bytes 594 ms 3 GET http www d fischer f m 8080 pntoolise 13072sc 3 M 625 bytes 94ms MIGET hitpiiiwww d fischeyP m 8080 pndooliLogoFische Total 0 81 sec 2 818 bytes 2 Requests 3 17 kbytes sec 2012 2013 Ingenieurb ro David Fischer AG Switzerland NebMainMenu 200 OK TEXT HTML 200 OK IMAGE GIF http 127 0 0 1 7990 P
132. ark Usually a Java application server supports both session cookies and URL rewriting however only one of these procedures is applied on a per user basis to build the session context The inner algorithm of the application server works as follows 1 When a web browser requests any page from the server for the first time the server does not know if the web browser supports session cookies For this reason the server sends a session cookie to web the browser and performs additionally URL rewriting for all hyperlinks and forms for the first web page 2 When the web browser requests a second page from the server and transmits the received session cookie back to the server the server will then know that the browser supports cookies For the current and all succeeding web pages URL rewriting will no longer be done 3 lf on the second page request the web browser does not send back the cookie or if the application server is configured to disable the use of session cookies in which case an initial cookie will not have been sent anyway the web server notes the absence of the session cookie and does URL rewriting for the current web page and all succeeding web pages You do not usually have to do anything special in this case because most Java application servers support Var Handler HR session cookies However if session cookies are disabled you must first enable the support of URL rewriting inside the Var Handler before the load test
133. ase that only constant values are required as Plug In input parameters see chapter 7 9 Of course such additional variables can also be extracted from other sources for example from Input Files or User Input Fields or from responses of previous URL calls Furthermore it is also possible to develop and add self written Plug Ins You will find the corresponding documentation in the Load Test Plug In Developer Handbook PDF and in the Proxy Sniffer Java API Documentation which both are included in the installation kit 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 53 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 5 Dynamically Exchanged Session Parameters The HTTP protocol by itself is stateless there is no memory from URL call to URL call however most web applications require state information such as the stage in a process that a user has reached before login after login placed an order and so on Usually cookies are used to keep state information Cookies are set by the web server as additional HTTP response header fields and sent by the web browser back to the web server along with the HTTP requests of succeeding URL calls This is normally not a problem because the correct handling of cookies is automatically done by the load test program However some web applications use as a special session context dynamically generated CGI or form parameter values which are exc
134. asured after the network connection was already established e Av Response Header Wait Time the average time for waiting for the first byte of the web server response header measured since the request has completely transmitted to the web server e Av Response Header Receive Time the average time for receiving the remaining data of the HTTP response header measured since the first byte of the response header was received e Av Response Content Receive Time the average time for receiving the response content data for example HTML data or the data of a GIF image e Average Response Time the average response time for this URL This value is calculated as reverse keep alive efficiency 100 Av TCP socket Connect Time Av Request Transmit Time Av Response Header Wait Time Av Response Header Receive Time Av Response Content Receive Time Av Response Content Receive Time URL Errors Real Time Profile of Error Types This diagram shows an overview about what kind of errors did occur for the URL at which time measured since the load test was started This basic error information is always accurately measured independently of the value chosen for the Additional Sampling Rate per URL Call and captured in every case also if no more memory is left to store full error snapshots 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 87 171 Apica ProxySniffer V5 0 User s Guide Englis
135. ate HTTP S Load Test Program Help POET Refresh Close A Apica xcellence in Cloud Performance Java Classname gt MTEST 01 Java Code Model large v apply heuristic methods from recorded session to check received content v 180 8858 1 v Generate External Files for XML and SOAP Request Data Content Test Algorithm Character Encoding gt 4096 Bytes v URL Execution i Serial Executed 18 Parallel Executed 82 Threads p User B Switch to Serial Exec Switch to Parallel Exec required enter a simple classname for the load test program with no path and no file extension HTTP Protocol Options 8 HTTP Protocol Version Allow Keep Alive Strip Referer Header Field Strip Accept Header Field to Load Test over HTTP S Proxy C Apply next proxy configuration from personal settings HTTP SSL Authentication Options E Basic Authentication O Apply individual Basic Authentication per user from input file basicauth td Digest Authentication O Apply individual Digest Authentication per user from input file digestauth txt use common Username Password NTLM Authentication O use common NTLM account from Personal Settings menu v Kerberos Authentication O use common Kerberos account from Personal Settings menu v HTTPS Client Certificates O apply individual PKCS 12 certificate per user from input file pkcs12auth bd v Pr
136. atus normal completed Scan started at 06 Dec 2008 20 31 25 required Enter a simple filename with no path and no file extension The file extension is always prxscn Example PageScan 01 recommended will be displayed as hint in Project Navigator A saved Page Scanner result can be restored and loaded back into the Page Scanner by clicking on the corresponding Load Page Scan icon inside Project Navigator Saved Page Scanner F Results LE ZeroThinkTime LPF Cluster El Hint saved page scanner results can be restored by clicking on the Load Page Scan icons R inside Project Navigator 13 1 2 Analyzing the Scan Result PRX Page Scanner Mozilla Firefox SEE E http 127 0 0 1 7990 dfischerlwebadmininterFace PopupPageScannerWeblet Y eee N See Proxy Sniffer e X 3 Web Admin Page Scanner Help Navigator Refresh Close car Abort Scan Display Result Convert to Session Save Reset Starting Web Page http www proxy sniffer com i Scan Status normal completed Scan started at 06 Dec 2008 20 31 25 E 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 155 171 Apica ProxySniffer V5 0 User s Guide English Edition PRX Page Scanner Result Mozilla Firefox Sele B http 127 0 0 1 7990 dfischer webadmininterFace PopupPageScannerResultWweblet f N 5 a ex e Proxy Sniffer j P Project Web Admin Page S
137. ay contain spaces or special characters Assign var to all request parameters with same recorded value 1 HTTP A 0 200 OK Formio mutisee 0 00 e e 2 Content Type TEXT HTML GET ipndaol servlet WebMainMenu jsessionid Assign e decode this is the reverse of encode This option IS 3 Expires 0 CHECKBOX Qf id 40 1012 normally not used 4 Cache Control no cache must revalidate CHECKBOX Br idit39 1011 password uservar l 5 Pragma no cache CHECKBOX amp f_id 38 1010 Z A Rp mnm username user var Assign var to all request parameters with same recorded value by enabling this option all URL calls in the recorded web surfing session are searched to see if any other URL calls use the same recorded value If so the variable will also be assigned to the other URL calls resulting in the global replacement of recorded parameter values irrespective of the parameter name amp Input File Input Files Aid File userAccounts bd Dn amp File Scope new line per user HTML ES 1 2 lt HEAD gt 3 lt META HTTP EQUIV CONTENT TYPE CONTENT texthtml charset iso 8859 1 4 lt TITLE gt Proxy Sniffer Project Master Directory Browser lt TITLE gt 5 6 1 Extract Var STYLE TYPE texticss l hady n a illi tabla tr td form innit calact antion f fontcita Ant fant famihr Arial Halvaties 1 o
138. be read each time an simulated user executes an inner loop The new lines are distributed over all users loops and inner loops Line Order Controls whether the lines are read in sequential or randomized order Comment Tag Defines a start character or a start string for commented out lines Such lines will be ignored during the load test Var Delimiter Defines the variable delimiter character which separates values contained on the same line several values variables can be extracted from the same line Trim Extracted Values Controls whether blank characters white spaces are removed from the start and the end of the extracted variables Controls the behavior when all lines from the Input File have already been read when a new line is requested reopen file the file is re opened If a randomized line order was set the lines continue to be randomly read in a new order stop load test the load test will be immediately aborted This option can be used to avoid duplicate logins with the same username password in the case where fewer lines are available than users which should be simulated Note that EOF can also become true for a randomized line order because the lines are first mixed during opening the file and then read 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 43 171 Apica ProxySniffer V5 0 User s Guide English Edition Next you should test to ensure that the parsing of t
139. button in the HTTP response content The http 127 0 0 1 7990 Proxy Sniffer URL Details Yar Handler Mozilla Firefox 3 i go l Se N 2 qe search results are marked in red er admin URL Details Var Handler gt E N a Refresh Because automatic handling failed you probably cannot extract the value by using the form parser HTTP Response Content Forms HTTP Request Header gt fxcontroltestat 443 Qj Extract or by using the hyperlink parser HTTP Response Content a en B MEWSTATE 1 loop var is ccept image gif images xbitmap image jpeg im orm Parameter Unique Hyperlinks Extract therefore you must use the text pattern 3 m eee TUE z si tlt unionis NM based token extractor E TN zu 5 SEE IEEE E zu 3 54 amp HTTP Request Content Paramete Proceed as follows VIEWSTATE_3 loop var HTTP ResponseHeader HTTP Response Content Forms Extract 1 Form 55 HTML Form Parameter 1 HTTP 1 1 200 OK Formp seting I gt 57 amp HTTP Request Content Paramete 1 Scroll left to the beginning of the line where the result is found AT m LLLI levid loop var erver Microsoft i inpSearchName 2 HTML Hyperlink and memorize the line number in this case 232 4 X Powered By ASP NET HIDDEN _ VIEWSTATE dDwWINzM3NDASMIMy103c 19 amp HTTP R
140. can be executed successfully You will recognize the need for this Enable dynamic URL rewriting m when you review the recorded URLs in the main menu if the URL rewriting parameter is found in all URL or calls in the majority of web pages you will have to enable URL rewriting support in the Var Handler Rewrite Parameter isessionid Map to var Name isessionid To do this proceed as follows Dynamic Handling automatically v 1 Click in any URL detail menu on the URL rewriting icon Wb inside the Var Handler required recommendation use default 2 Enter the name of the URL rewriting parameter in the field Rewrite Parameter HS 3 Enter an arbitrary variable name in the field Map to Var Name Note dynamic URL rewriting is only needed for 4 Use the option automatically for the field Dynamic Handling load test against J2EE servers if session cookies have been disabled at server side 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 67 171 Apica ProxySniffer V5 0 User s Guide English Edition After URL rewriting has been enabled the Var Handler shows only the first extraction of the URL rewriting parameter but not its assignment This is normal behavior because the assignment in succeeding URL calls will be done automatically later in the load test without the need for additional configuration Var Handler Ga db G E Note the URL rewriting parameter may also have a name othe
141. can try ISO 8859 1 or UTF as a workaround if Page Scanner is unable to Page Scanner input Porei jger 0 extract hyperlinks succeeding web pages from the starting web page Starting Web Page gt 16 4 Char Encoding Auto Detect v Exclude Path Patterns Eu _ Follow Web Servers Verify External Links e Exclude Path Patterns allows you to exclude one or more URL path patterns rad LAT m ren from scanning The path patterns are separated by commas Include Options y ML A a N mr e Follow Web Servers allows you to include content and web pages from other Follow Redirections 10 Follow Path Repetitions 1 Y O Follow Col Parameters web servers within the scan for example this option can be used when images Authentication ame Poser U embedded in the web pages are located on another web server You can enter E ads MI Pi sat Version 6 iB several additional web servers separated by commas Not recommended epion pp configuration of Personal Setihas Example http www lt domain1 gt https imgsrv lt domain2 gt 444 The protocol a 0 http or https the host name usually www the domain and the TCP IP port Be a are considered but URL paths are NOT considered e Verify External Links allows you to verify all external links to all other web servers This is commonly used to detect broken hyperlinks to other web servers e Include affects which sets of e
142. canner Result Help ps Navigator Refresh Close Convert to Session Search ASCII Text wv Remove Scan Input Parameter gt Scan Statistic Broken Links Non Processed Web Servers gt Scan Result per Web Page Duplicated Content gt Largest Web Pages Slowest Web Pages Remove URLs with content MIMEtype applicationipaf any none AND HTTP status code 200 ok x Page 1 http www proxy sniffer com Web Load and Stress Testing Tool Proxy amp 1 0 2000k 168 ms 43 779 bytes http www proxy sniffer corm x i 1 20800k 777 z1ms 40517 bytes formatcs To 4 72 AS Y Ret Poo UK Riu The most important statistical data about the scan are shown in the overview marked in orange near the top of the window Below the orange marked overview various scan result details can be selected The search form on the right side near the scan result detail selection allows you to search for an ASCII text fragment over all web pages of the scan result By default the text fragment is searched for within all HTTP request headers all HTTP response headers and all HTTP response content data The remove URL form which is shown below the scan result detail selection allows you to remove specific sets of URLs from the scan result The set of removed URLs is selected by the received MIME type examples IMAGE GIF APPLICATION PDF and linked with a logical AND con
143. ce Code Editor Java Editor Path CAProgrammeiTextPad 5iTextPad exe PDF Report Branding Logo gif 350 x 300 pixels gt http 27 0 0 1 7980 dfischerwebadmininterface htdocs ZebraLogoP df2 gif Footer on First Page Proxy Sniffer amp trade Web Load and Stress Testing Tool pnVersion Ingenieurb amp uuml ro David Fisch Detail Report Title Proxy Sniffer Detail Report Load Curves Report Title Proxy Sniffer Load Curves Report Comparison Report Title Proxy Sniffer Comparison Report 3 Enter URL for image Windows example to set a local image file WC images LoadTestReportLogo gif Hint default values are applied by entering an empty blank value into the Input fields Done All Rights Reserved Page 147 171 Apica ProxySniffer V5 0 User s Guide English Edition 12 2 Sending Email and SMS Alert Notifications during Test Execution The Exec Agents can be configured in such a way that Email and SMS Alert Notifications are released during the execution of a load test job The corresponding Alert Configuration Menu can be called from the Personal Settings Menu The Alert Configuration Menu will create a file named AlertConfig xml which is located in the Proxy Sniffer installation directory and which contains the configuration data for all alert devices and for all alert notifications If no AlertConfig xml file exists on an Exec Agent no alerts are released from this
144. cessive test runs with 1 2 5 10 20 50 100 200 500 1000 users The results of these test runs can be combined to produce load curves which will provide an excellent overview of the response time behavior the throughput and the stability of the web server or web application and how they vary depending on the number of users PRX Analyse Load Tests Load Curves Mozilla Firefox B http 127 0 0 1 7990 dFischerwebadmininterFace AnalyseLoadtestResultDiagramWeblet action display amp key 0 1b9a41107a26c3e9cf4a3a0dcd976dOb amp key 1 122f44947d813bb03780b7ba5604b4948 amp kev eee A ni far Hint Click inside the diagrams SS E t eal a bd Analyse Load Tests Load Curves on the icons to display details Help c a Close O Average O Session Failures Max Server Capacity reached Response Time Behavid ance Web Transaction Rate ansaction Rate Hits per Second successful complet url calls per second concurrent user concurrent user Stability Session Failure Rate Network Influence Average Network Connect Time PRX Analyse Load Tests Load Curves Mozilla Firefox B http 127 0 0 1 7990 dfischer webadmininterf ace AnalyseLoadtestResultDiagr amWeblet eee seo p Sniff Hint Click inside the diagrams SS e s it Analyse Load Tests Load Curves on the icons to display details Help PDF Report Export Close OOA 5 O sensn Free Page 1 to 6 Server Sided Tuning recommended
145. contains all URL calls and page breaks and is repeatedly called from the method run This structure has a direct influence on how variables must be declared within the program static global variables are shared between users that is all users see the same value If a static variable is not a primitive data type integer boolean etc then modifications to the value must be protected by a synchronized statement in order to avoid data corruption common local variables have a per user value even if they have been defined only once The values of these variables are set by the constructor or during the execution of the methods execute and or run All Rights Reserved Page 165 171 Apica ProxySniffer V5 0 User s Guide English Edition For debug purposes an empty log vector is created before the method execute is called The reason for doing this is that inside the execute method any console and log output should not be written by calling the Java method System out println as later on it would be nearly impossible to check what has happened inside a thread because all output data of all threads would be mixed The method log exists for this purpose This method collects all output data of a loop until the loop has been terminated After loop termination the log data of the loop are synchronized and written to standard output inside the method run During the development of your own program extensions you can force
146. css 3 216 128 ms 218 ms 616 bytes GET http 32 1 5 4 5 0 gif 4 216 43ms 125 ms 669 bytes GET http 32 15 4 5 80 arraw red 12x8 gif 5 216 0 124 ms 390 ms 1 783 bytes GET http 32 15 4 5 8 0 flagEngland gif 6 216 0 65 ms 125 ms 812 bytes GET http 1 92 16 4 5 50 NagGerman gif E 216 0 104 ms 390 ms 22735 bytes GET http1192 16 4 5 B80fmages_enfscreenshotClusterPrewjievw jog 8 216 0 133ms 375 ms TETE bytes GET httpjr182 15 4 5 80 mages eniScreenshotevebAdrmintP review dqif 8 216 0 135 ms 266 ms 10 586 bytes GET http1192 16 4 5 B0fmages enfScreenShotFinalResultPreview gif 10 D 216 gy ms 141ms 20382 bytes GET http 92 16 4 5 804mages_ enfScreenShotRealtimerreview jpg 216 4 533 ms 10406 ms 112866 bytes 10 URLS E 11 Page 2 Login Form users think time 3 0 seconds Test Passed Falled Av Time lt 90 AV Size URL 12 153 0 7755ms 12844 ms 2 592 bytes GET httpirt82 15 4 33 8080 prtaol servietvebMainMenu 13 154 0 5 247 ms 8797 ms 1227 bytes GET htto192 16 4 33 8080 prtoolLogoFischer gif Total 154 13 003 ms 20 642 ms 4118 bytes 2 URLS user s think time 3 0 seconds AM Size URL 44 903 bytes POST http 32 15 4 33 8080 prxtaol servet vebainhenu 791 bytes GET http 92 16 4 33 5080 ptoolEkit gif 884 bytes GET http 01192 16 4 33 8080 prdool Home qif 6 354 ms 14641 ms 743 bytes GET http 91 92 16 4353 8080 pretooliMavigation git 2 704 m
147. cts all received content data of error snapshots taken malformed web pages and stores them in files so they can be displayed later in the web browser import java io FileOutputStream import dfischer utils PerformanceData import dfischer utils PerformanceDataRecord import dfischer utils PerformanceDataRecordFailureInfo import dfischer utils HttpTestURL Writes the response content of all error snapshots to files if they contain ASCII HTML XML data Program Argument name of the result file prxres public class ExtractErrors public static void main String args try read result file from disk PerformanceData performanceData new PerformanceData performanceData readObjectFromFile args 0 loop over all measured url calls and page breaks PerformanceDataRecord performanceDataRecord performanceData getPerformanceDataRecord for int x 0 x lt performanceDataRecord length x switch performanceDataRecord x getDataType case PerformanceDataRecord TYPE PERFORMANCE DATA loop over all error snapshots per url call PerformanceDataRecordFailureInfo failureInfo performanceDataRecord x getFailureInfo for int y 0 y lt failureInfo length y get data of failed url call HttpTestURL testURL performanceDataRecord x getFailedUrl failurelnfo yl if testURL null now we have access to all frozen url data String fileStartName
148. current web server configuration and so on This annotation will be displayed on the result diagrams 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 79 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 1 Starting Exec Agent Jobs If you have specified that the load test program be executed by a single Exec Agent but not by an Exec Agent Cluster see Chapter 11 the load test program is transmitted to the local or remote Exec Agent and a corresponding load test job with a job number is created locally within the Exec Agent The job is now in the state configured that is ready to run but the job is not yet started gt PRX Start Job 1333 Mozilla Firefox Hint each Exec Agent always executes load test jobs as separate http 127 0 0 1 7990 dFischer webadmininterface PopupExecAgentStartConfigWeblet execAgentId 1256761045921 amp jobld 1333 h backg rou nd processes and IS also able to execute more than one job Pe ei een Start Job 1333 on Local Exec Agent Ea x at the same time The option Display Real Time Statistic only means Im that the GUI opens an additional network connection to the Exec Agent S which reads the real time data directly from the memory space of the dE 3 era 1288 1 80 adlay 200 TER D sampling 15 percpage 100 percurl 20 maxerrmem 20 nolog annotation First test run untuned corresponding executed load test program Concurrent Users 100 Saa ng
149. d BERNER LEREALE nn server shutdown D 0 n0 0 00 40 0 01 20 0 02 00 0 02 40 0 03 20 0 04 00 0 04 4 Losd Testi Testi 19 Ger 2009 15 54 57 Useri 0 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 91 171 Apica ProxySniffer V5 0 User s Guide English Edition You can also modify delete or add real time comments before you generate the PDF report However all retroactively entered real time comments are not permanently stored inside the result data PRX Result Detail Mozilla Firefox J Mtpu127 0 0 1 7990 dfischer webadmininterf ace PopupAnalyseLoadtestDetalsWeblet key 4348854 cb6D4Fe ded Pe m Load Test Result Detail Statistics and Di Advanced Test Parameter Measured Results per Single User per Loop Overall Startup Delay per User 200 ms AV Session Time per Loop 20 62 secloop Web Tra Request Timeout per URL 60 sec AY Response Time per Page 0 16 secipage Sessio Statistic Sampling Interval Network Throughput per User 16 6 kBytes sec Total Me Warnin Diagram Response Time per Page Results per U Diagram Response Time Percentiles Diagram Top Time Consuming URLs Diagram Con Diagram Web Transaction Rate Diagram Users Waiting for Response Diagram Cor Diagram Network Throughput Diagram HTTP Keep Alive Efficiency gt Diagram E Diagram Error Types Diagram Number of Errors per
150. d factors If you mix strong and weak systems within the same cluster it is recommended that you give a higher load to the stronger systems than to the weaker systems It is not necessary that all cluster members have the same operating system time Fach time a cluster job is started the cluster job controller automatically measures the time differences between the cluster members These measured time differences will be automatically accounted for when the consolidated statistics data are merged Testing Network Connection ta Exec Agent Test PCI Successful Connected to 192 16 1 35 7993 Exec Agent OS Type Windows lt P 5 1 Java 1 3 1 11 Proxy Sniffer 4 0 4 Exec Agent Os Time Offset 294 Seconds in the past Update Load Factor for Exec Agent Test PC I amp Yes Y Update Cluster Members No Continue All Rights Reserved Page 143 171 Apica ProxySniffer V5 0 User s Guide 11 3 Starting Distributed Load Tests English Edition If additional Exec Agents and or clusters have been defined you can select when starting the test run from which system or cluster the load test is to be released input field Execute Test from The succeeding steps inside the Web Admin GUI are then the same as for executing the load test locally http 127 0 0 1 7990 Proxy Sniffer Project Navigator Execute Load Test Mozilla Firefox E Pale xl SS ae ipe Project Navigator Execu
151. d test is not exceeded ouch failures are also named fatal errors e None Continue Loop means that the simulated user continues to execute the current loop repetition of web surfing session Such failures are also named non fatal errors This option should only be used if no variables have to be extracted from the response of the URL call or in other words only if the succeeding URL calls do not depend on the response of this URL Reset Settings By clicking on the Apply Default Settings button at the bottom of the window you can undo your changes and the default settings are reapplied 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 22 171 Apica ProxySniffer V5 0 User s Guide English Edition 4 2 3 Configuring Parallel or Serial URL Execution with Web Pages This function allows to configure the Runtime Execution Behavior serial or parallel execution order for one URL or for a group of URLs or for all URLs which will be applied per simulated user during the execution of the load test Normally the first URL of a standard Web page should always be executed serial analog to the behavior of a normal Web browser Additionally any redirects located at the start of a Web page should also be executed serial Subsequently following URLs of a Web page such as images can then be executed in parallel The synchronization point for all in parallel executed URLs is always at the end of the page
152. dition with the received HTTP status code for the URLs 200 302 or with a Page Scanner error code such as network connection failed e with content MIME type selects a specific MIME type see also http www iana org assignments media types The input field is case insensitive upper and lower case characters will be processed as identical any means that all MIME types are selected independent of their value none means that only URL calls whose HTTP response header does NOT contain MIME type information HTTP response header field Content Type not set will be selected e HTTP status code selects an HTTP status code or a Page Scanner error code Note A few URLs with excluded content types may flow into the scan result not selected by scan input parameter You can use the remove URL form to clean up the scan result and to remove any unwanted URLs The most common case is to remove PDF documents from the scan result 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 156 171 Apica ProxySniffer V5 0 User s Guide 13 1 2 1 Scan Result Details gt Scan Input Parameter gt Broken Links e Scan Input Parameter displays all inout parameters for the scan without authentication data a d Scan Input Parameter E Scan Statistic E Duplicated Content gt Mon Processed Web Servers Scan Result per Web Page gt Largest Web Pages gt Slowest Web Pages starting Web
153. do File Edit View History Bookmarks Tools Help s a A ea i httpifwww proxy sniffer com LU j ABP ti Web Load and Stress Testing Tool P Proxy Sniffer Console Don t miss to return back to this user s manual We recommend that you read at least chapter 7 completely inclusive all subchapters in particular subchapter 7 5 because the usage of dynamic variables is a little bit tricky If you are using an Internet Explorer or a Safari web browser for recording of web surfing sessions you should also read the next chapter 2 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 8 171 Apica ProxySniffer V5 0 User s Guide English Edition 2 Recording Web Surfing Sessions without using the Firefox Recoding Extension Hint you can skip this chapter 2 if you use a Firefox web browser AND have also installed the Proxy Settings E 3 xj Firefox Recording Extension m Tvpe Proxy address to use Port Load tests against web servers or web applications are usually based on recorded web surfing sessions en i A This means that you usually first record a web surfing session before you can execute a load test In E simple cases when no login is required and no HTML forms need to be submitted you may secure 127 0 0 1 sr alternatively use the Page Scanner tool described in chapter 12 2 instead of recording a web surfing E session manually Gopher Recording of web surfing se
154. e English Edition 7 5 1 Automated Handling of Dynamically Exchanged Session Parameters Var Finder The Var Finder menu provides an overview of all URL request parameters and their values used anywhere in the entire recorded web surfing session In this view a parameter name value pair is shown only once even if the same name value pair is used by more than one URL call If the same parameter name is used with different values it will be shown multiple times once for each distinct value Proceed as follows 1 First review the recorded values and try to judge which values could be dynamically exchanged session parameters If the value contains a long number or is a cryptic hexadecimal string the value has a good chance of being a dynamically exchanged session parameter In the example at left levid id and VIEWSTATE are dynamically exchanged session parameters But type and Status1 ins_step22 txtPolicyNumber are not because their values have been entered manually into forms during the recording of the web surfing session 2 Try next to perform an automated handling of the dynamically exchanged session parameters This succeeds in approximately 50 of all cases To do this click on the BY icons which are shown at the left of the parameter names If you receive a success message there is nothing more to do for this parameter Dynamical handling of parameter lewd successfully accomplished The corresponding definition
155. e main menu and from the Project Navigator without loading result files test program Start Date date and time the gt http 127 0 0 1 7990 PRX Main Menu Mozilla Firefox File Edit View History Bookmarks Tools Help test run was started DM Proxy Sniffer Main Menu E a p a ha L s t E lh A i e Web Admin Professional Edition Help Tools Scanner Settings Navigator NL a Load Test load asts Display User number of simulated a users Recorded ignes pr Q H un l MILL Test Duration duration of PRX Analyse Load Tests Select and Compare Results Mozilla Firefox AE lt the ie van le EB http fr 127 0 0 1 7990 dfFischerwebadmininterFace Popup nalyseLeadtestweblet zlaadFileB64 GzpcRa8Srdw 1IbnRIIHVugzcBFawszdavsbHwuzezWuxal1dasuz1xQcma4eviyuawzmzsJcTXILEzXNOC 1 SRT Ir i 5 q 45 E Web Trans number of Ld dd rk Jobs Analyse Refresh Close ES roues l BE a xl successfully executed URL SU X RB Analyse Load Tests Select and Compare Results Help Clear Refresh Close pana calls per second hits per second that is the web E Uriarte Browse Pe extension poes C ES ES server throughput Sess Failures percentage of failed loops 223 Yond Youd Juyoad Test Start Date User Test Duration Web Trans Sess Failures URL Error Rate Net Throughput Annotation 238 A Test 28 Sep 2009 13 41 02 800 19 01 min 174 8 callsiser 0 0 0 0 12 28 MBitisec P B Te
156. e of the URLs are executed in parallel Proxy Sniffer measures the response time of a page in different ways If all URLs are executed in serial order the response time of the page is calculated as the simple sum of all response times of the URLs without considering any internal overhead time between the URLs On the other hand if some of the URLs are executed in parallel the response time of the page is measured as the time difference between the start of the page and the end of the page and includes also the internal overhead time All Serial Execution Order eng mel Time clu 2 Time Qe URL Time n Time nasa eben Response Time URL 1 Time URL 2 Time URL n Time Mixed Serial Parallel Execution Order Page Start Time Page End Time URL 1 S v URL 2 P URL 1 Time ly URL 2 Time m URL n Time Page Response Time Page End Time Page Start Time 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 25 171 Apica ProxySniffer V5 0 User s Guide 4 3 Executing a First Load Test English Edition You can now execute a first try of the load test if your recorded web surfing session does not contain dynamically exchanged session parameters see also chapter 7 5 For this here only a short overview is shown More detailed information about executing load tests is documented in the chapters 8 9 and 10 First of all you have to convert the recorded web surfing session
157. e option Assign var automatically to all HTTP requests which contain form or CGI parameters with the same recorded Serer PETE E SetCookie SURZc3kKbLLagqobblfqmblphwipjesbkdi 5 Content Type textiplain v HIDDEN Di 3 HIDDEN bp sess id 59 6981027551FF8AADFEBCE ye i TS gt A lii Extract Value of Form Parameter sess id Select Form by Form Index 3 Recorded Value 53 85881027551FF8AADFEBCEDI b iefisecure html onl kdl sess menu anz p sess id 58AA B881027551FFBAADFEBCEDDABS5 amp p menu id un t luz fefisecure htmifonl kdl kta salda anz p sess id 59446991027551FFSAaDFEBCEDDABS Map to Var Name ivi Assign var automatically ta all HTTP requests which contain farm ar CGI parameters with the same recorded value Try URL encoding 3 PRX Search Overall Visible Items Mozilla Firefox El The Search Overall menu gives you PUO http 127 0 0 1 7990 dfischerfwebadmininterface PopupSearchRecordedDatawWeblet L an excellent ove rview if you know ae Search Overall Visible Items x already the name or the value of a f Dynamically Exchanged Session Jeur loop var Search ASCII Text 59446991027551FFSAADFEBCEDDABS Match Case v Include URL Encoded Values Parameter Thus it is easy to determine 23 A HTTP Redirection La nside HTTP Request Header HTTP Request Content ot Inside Referer ues HTTP HD Header HT
158. e token containing the desired variable text fragment into the field Extract Token Nr and then click again on Test Extract Tl Extract Token Nr Microsoft Interne Help Extract Token Hr Token Recorded value TD class mainField widthz imu id dtl tingca d ctl img amp ccaunteubtipe onmouseove ShowLayve divSubtjpeLaye 451047 2 All Rights Reserved Page 59 171 Apica ProxySniffer V5 0 User s Guide Var Handler Ga db Ed XE P Extract Var from Text Line Pattern Text Line contains text Search Text Img id ctlReportingcard Extract Var on located line Token Delimiters lr Extract Token Hr 7 Test Extract A Recorded Value 451047 Extracted an Line 232 Map to far Name w Assign varto all request parameter with same recorded value v Tre URL Encoding Assign var to all matching request file and request content patterns with Same recorded value Extract 2012 2013 Ingenieurb ro David Fischer AG Switzerland English Edition 8 Check to see if the blue marked value is exactly the same as the recorded value of the parameter which should be extracted that is if it is the same as the variable test fragment search result 9 Finally enter an arbitrary variable name into the field Map to Var Name In this example the name id 1 is chosen because the parameter must be extracted twice once each into two different variables as was sho
159. eb browser uplink Kbps Network bandwidth limitation per concurrent user in kilobits per second for the uplink web browser to web server sampling seconds otatistic sampling interval in seconds interval based sampling Used for time based overall diagrams like for example the measured network throughput percpage percent Additional sampling rate in percent for response times of web pages event based sampling each time when a web page is 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 99 171 Apica ProxySniffer V5 0 User s Guide percurl lt percent gt maxerrsnap lt number gt maxerrmem lt megabytes gt nolog dssl multihomed Ss version ssicache seconds tz value annotation text called English Edition Additional sampling rate in percent for response times of URL calls event based sampling each time when a URL is called Max number of error snapshots per URL per Exec Agent 0 unlimited Max memory in megabytes which can be used to store error snapshots 1 unlimited Additional separate log file disabled Debug loops Debug headers amp loops Debug content amp loops Debug cookies amp loops Debug keep alive for re used network connections amp loops Debug information about the SSL protocol and the SSL handshake amp loops Forces the Exec Agent s to use multiple client IP addresses Use fixed SSL protocol version v2 v3
160. ect_DiPerformanceTuningDay eu MyTests Plugins Project A Project B Lg EnduranceTest Project_C Project_D Dod Yond Yond Your Yond Release Plugins Release_1_1 aem NA oso m ggg 2012 2013 Ingenieurb ro David Fischer AG Switzerland English Edition Individual Java load test programs can also be renamed or copied to a new name This can only be done using the Project Navigator that is it cannot be done using the Operating System This is because the Java program contains references in the source code to its own name The Project Navigator handles this requirement and will automatically make the appropriate adjustments when copying or renaming a Java load test program Note compiled Java programs class files can never be renamed only source files java can be renamed Note also that the Project Navigator will require confirmation when overwriting or deleting files using a red shaded status row Whenever a red shaded status row appears you should review the action before approving it An example is given at left for deleting files Clicking on the EI Icon in the Project Navigator will provide a preview of the measurements in the statistics files including the description associated with the corresponding test run The description of the recorded web surfing sessions and the load test programs will also be displayed if available This feature allow you to quickly compare statistics files of
161. ecute Load Test window at any time and you can check later Job State Date E E eg the result or the actual effort of all load test jobs by using this 1339 X running part of clusterjob 01 Nov 2009 16 33 41 Test01 u 66 d 600 t 60 sdelay 600 maxloops O sampling 15 percpage 100 percurl 100 menu 1338 X 3 configured 01 Nov 2009 16 33 22 Test01 u 200 d 600 t 60 sdelay 200 maxloops O sampling 15 percpage 100 percurl 10 1337 X 3 scheduled 01 Nov 2009 20 00 00 Test01 u 200 d 600 t 60 sdelay 200 maxloops O sampling 15 percpage 100 percurl 10 1336 X X completed part of clusterjob 01 Nov 2009 15 40 48 Test01 u 66 d 600 t 60 sdelay 600 maxloops 0 sampling 15 percpage 100 percurl 100 If a load test job has completed you are disposed to acquire the 1335 X 3 configured part of clusterjob 01 Nov 2009 15 31 46 Test01 u 66 d 360 t 60 sdelay 600 maxloops 0 sampling 10 percpage 100 percurl 100 i cca i 1334 amp X configured partofclusterjob 01 Nov 2009 15 30 49 Test01 u 134 d 1200 t 60 sdelay 600 maxloops 0 sampling 15 percpage 100 percurl 2 corresponding statistic result file prxres If a load test job is still 1333 4 3 configured 30 Oct 2009 22 50 41 Test01 u 100 d 1200 t 60 sdelay 200 maxloops 0 sampling 15 percpage 100 percurl 2 running you are disposed to the temporary live statistic window of 1332 X X completed 28 Oct 2009 23 23 40 Test01 u 10 d 240 t 60 sdelay 200 maxloops O
162. eite Page Response Time 188 Sampling Rate Page H1 startseite milliseconds AAA AAA AAA A A A A A AA 0 00 00 0 05 00 0 10 00 0 15 00 0 20 00 0 25 00 Load Test Test 1 28 Sep 28809 14 24 34 User 1600 Page 2 login maske Page Response Time 188 Sampling Rate Page 2 login maske milliseconds 0 0 00 00 0 05 00 0 10 00 0 15 00 0 20 00 0 25 00 Load Test Test61 28 Sep 2889 14 24 34 User 1666 Page 3 sicherheitsnummer English Edition Page Response Time Percentile 188 Sampling Rate Page H1 startseite milliseconds 1423 IFEF 727 av 0x 10 20 30 40 50 60 70 50 90 100 Load Test Test 1 28 Sep 28809 14 24 34 User 1606 Page Response Time Percentile 188 Sampling Rate Page 2 login maske milliseconds ed rosas 1320 av 0 10 20 30 40 50 60 70 50 30 100 Load Test Test 1 28 Sep 2889 14 24 34 User 1666 Page Response Time 188 Sampling Rate Page H3 sicherheitsnummer All Rights Reserved Page Response Time Percentile 188 Sampling Rate Page H3 sicherheitsnummer Page 111 171 Apica ProxySniffer V5 0 User s Guide 10 1 3 Results per URL Call Overview Displays statistics about all URL calls 0 Page ZT Start Page user s think time 0 0 seconds Test Passed Failed AV Time lt 90 3 AV Size URL 1 216 0 3410 ms 8 031 mms 42 395 bytes GET http 92 16 4 6 80 2 216 0 102 ms 157 ms 5202 bytes GET httpjr182 15 4 5 8 farmat
163. elected Files IV Load prxres File on Analyse Load Test Menu M Close window after acquire Cluster Job Directory CADOKUME 1imutongiLOKALE 1iTempiPrmcClusterJobsijob_6 Project Navigator Directory 3 C Programme ProxySnifferiMyTests Usually you would work inside the Analyse Load Tests menu with the consolidated measurement results only However it is also possible to expand the measurement results to access the results of each individual Exec Agent job A Test01 16 Jun 2006 15 27 04 1 03 min 7 94 trisec A Test01 16 Jun 2006 15 27 04 1 04 min 14 09 trisec Test01 16 Jun 2006 15 27 05 1 03 min 19 38 trisec 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 96 171 Apica ProxySniffer V5 0 User s Guide English Edition This feature can be used to check if all cluster members have measured approximately the same response times however variations in a range of 20 or more may be normal amp Test01 16 Jun 2006 15 27 04 43 1 03 min 7 94 trisec 0 65 MBit sec amp Test01 16 Jun 2006 15 27 04 64 1 04 min 14 09 tr sec 1 1 13 MBit sec 4 Test 16Jun 2006 15 27 05 93 1 03 min 19 38trsee 100 00 1 62 MBitisec qe NON E LN Analyse Load Tests Page amp Url Response Time Comparison gt 3 x Response Time Comparison per Page Page 1 Start Page Page 2 Login Form Page 3 Login Main Menu 3 TestCil Test i Users 43 Start 16 Jun 2006 15 27 64 Dura
164. equest URI 5 AspNat Version 1 1 4322 HIDDEN _ EVENTTARGET btnSearch j TD class mainField width gt lt img id dtiReportin Duo A exu CU 233 div id divSubtypeLayer451047 style position abst syortS1047 civespostion absolute dh 220 fist dipteymonere e A border 0 cellspacing 3 cellpadding 5 bgcolor eeeeee z none 2 Locate a unique text pattern near the place where the variable itlReportingcard__ctl2_IblAccountSubtypeText bestehender FX Control Service Vertrag ispan gt E i text fragment the search result should be extracted This text 44 pattern can also be located on a preceding or succeeding line T U1 accountlist aspx act first amp levid 94153 Please note that the variable text fragment itself must not be U2 SS IE ENFIN part of the text pattern If the text pattern is not on the same line 5 Wb accountiistaspx agenda demand E as the text fragment to be extracted you must also memorize the negative or positive line offset 1 2 1 2 3 Mark the unique text pattern and click on the var extractor icon Eb 231 lt TR gt 32 lt TD class mainField width gib EI Sa d PA A gt Ly 233 lt div id divSubtypeLayer15104 7 style position absolute width 220 left 50 dis 4 Wait 3 seconds The selected text pattern will be copied into the Var Handler input form field Search Text At this point if a negative or positive line offset is needed select the li
165. equired e Additional Settings krb5LoginModule conf normally there is no need to make any modifications in this file This file contains some generic Kerberos settings If you modify this file you must restart the Proxy Sniffer GUI restart the Proxy Sniffer Console Additional Requirements for Kerberos Authentication the Kerberos tickets are acquired by the web clients by using DNS as communication protocol to the KDC This means that all computers on which Proxy Sniffer is running the GUI and the Exec Agents and which are using Kerberos must be able to open a DNS network connection to the KDC This is the only additional requirement It is not required that the computers are registered as a member of the domain and it is also not required to configure any other things inside the operating system Kerberos authentication against a web server located in a Windows domain is also supported for all non Windows operating systems like for example Mac OS X Linux and Solaris 3 1 2 7 GUI Settings Input Fields e Time Zone allows you to set the default time zone to be used by the load test programs and by the GUI e Number Format allows you to set the default decimal grouping separator character for numbers for example 123 456 00 or 123 456 00 e Background Color allows you to choose your desired background color for all windows only temporarily applied until program termination for Windows Mac OS X and Ubuntu systems modify the
166. er s Guide English Edition 10 3 2 Response Time per Page This menu option displays the load curves of all web pages average response times and 90 percentile value of the response times Again you can click within the diagrams on the red rhombuses to display the detailed results of the corresponding test run PRX Analyse Load Tests Load Curves Mozilla Firefox B http 127 0 0 1 7990 dfischer webadmininterface AnalyssLoadtestResultDiagramweblet LY H l Hint Click inside the diagrams 3 a6 P Sniff wal b A d bs Mai An alyse LoamTests Load Curves on the icons to display details Help PDF Report Export Close Page 1 to 6 Response Time per Page Average milliseconds 1 User 10 User 20 User 200 User 12885 ms Frage as ouy 37ms ems 48ms 136ms sssms 1250 ms Come Stme Toms 1r ms sims 1554ms 120 140 160 150 200 concurrent user Page 6 Support FAQS 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 136 171 Apica ProxySniffer V5 0 User s Guide 10 3 3 Response Time per URL English Edition This menu option displays the load curves of all URL calls average response times and 90 percentile value of the response times Again you can click within the diagrams on the red rhombuses to display the detailed results of the corresponding test run PRX Analyse Load Tests Load Curves Mozilla Firefox E oy
167. erent number of concurrent users and compare the measured results Click on the magnifier for clei x z E Done A n z http 127 0 0 1 7990 Proxy Sniffer Project Navigator Mozilla Firefox 2 ge E X S Lea N Project Navigator Help Setup Network Jobs Analyse Refresh Clame This menu shows all files of the load test job however only the CiProgramme Proxys Bong Statistics results file is usually needed and this is already selected The out file contains debug information and the err file is either empty or contains internal error messages from the load test program itself Ha ScriptExamples Trash By clicking on the Acquire Selected Files button all selected files are transferred back to the local Project Navigator directory PASPA EUN mm m m m If the checkbox Load prxres File on Analyse Load Test Menu is selected the statistics results file is also loaded into the memory area of the Analyse Load Tests menu where the statistics and diagrams of the measured data can be shown analyzed and compared with results of previous test runs 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 93 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 2 Starting Cluster Jobs If you have specified that the load test program be executed by an Exec Agent Cluster see Chapter 11 2 the load test program is transmitted to the local cluster job controller
168. essibility of an Exec Agent by clicking on the B icon within the list of Exec Agents functional ping of Exec Agent 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 142 171 Apica ProxySniffer V5 0 User s Guide 11 2 Configuring Load Releasing Clusters English Edition If several Exec Agents have been defined they can be combined to form a load releasing cluster You can also define more than one cluster by using some of the same Exec Agents in several different clusters http 127 0 0 1 7990 Proxy Sniffer Project Navigator Exec Agent Network Configuration Mozilla Firefox iBi x eee gee Proxy Sniffer Web Admin Exec Agent Description Load Factor A B Local Exec Agent 1 00 Q 3 B TestPC II 1 00 A 36 B SunFire 240 1 00 Host 127 0 0 1 7993 plain 192 16 4 35 7993 plain 192 16 4 78 7993 plain direct network connection direct network connection direct network connection Add New Exec Agent Description Host Port 7993 Protocol plain Username Password Indirect Network Connection through HTTP S Proxy Proxy Host Proxy Port Proxy Username Add New Exec Agent optional only for http and https protocol supported Proxy Password Project Navigator Exec Agent Network Configuration Port Protocol ProxyHost ProxyPort Proxy User Auth He Ip 4 Refresh Close Exec Agent Clusters Create
169. est job has completed Page 83 171 Apica ProxySniffer V5 0 User s Guide 9 1 1 1 Response Time Overview Diagrams Real Time PRX Job 188 Response Time Overview Mozilla Firefox P http 127 0 0 1 7390jdfischer webadmininterFacePopupDireckoryMavigatorDisplayLiveStatisticsOverviewlWebleE jobId 188 amp execAgentId 12403061524068filePathB64 OQzpcRaardw1IbnRIIHvu 5 7 Ld dil 2 Proxy Sniffer Web Admin Response Time Last Minute Time Bars Average values Ww Page 1 Response Time 186 Sampling Rate Page 1 Web Load and Stress Testing T Average Walues response time in milliseconds Page 2 Response Time 188087 Sampling Fate Page 2 Web Lasttest und Stresstest Average Values response time in millizecondz Page 3 Response Time 1HBH Sampling Fate Page 2 Web Load and Stress Testing T Average Values response time in milliseconds Page H4 Response Time 166 Sampling Rate Page 4 Web Load and Stress Testing T Average Values response time in milliseconds ni1 0 O 00 54 0 4 0 00 42 O 00 56 0 00 530 2012 2013 Ingenieurb ro David Fischer AG Switzerland Elapsed Time 10 29 min Update Display Remote Exec Agent Job 188 13 Oct 2469 26 29 18 18 29 min elapsed Remote Exec Agent Job 1883 13 Oct 2669 26 29 18 18 29 min elapsed Remote Exec Agent Job 1883 12 Oct 266 26 29 12 18 29 min elapsed 00
170. est over HT TP S Proxy in the Generate HTTP S Load Test Program menu see chapter 8 Input Fields e Next Proxy HTTP Host DNS hostname or TCP IP address of the next proxy server for unencrypted connections e Next Proxy HTTP Port HTTP TCP IP port number of the next proxy server for unencrypted connections e Next Proxy HTTP Cache disabled if checked request the next proxy server to disable its internal cache e Next Proxy HTTPS Host DNS hostname or TCP IP address of the next proxy server for encrypted connections e Next Proxy HTTPS Port HTTPS secure TCP IP port number of the next proxy server for encrypted connections e Next Proxy Auth Username basic authentication username used for proxy authentication on the next proxy server e Next Proxy Auth Password basic authentication password used for proxy authentication on the next proxy server e No Next Proxy for Host Domain allows you to set a list of hosts or domain names for which the proxy settings must not be applied The entries must be separated by commas or semicolons For Kerberos Authentication against an outbound proxy server Add additionally the full DNS name of the next proxy server to the host names in the Kerberos Configuration and set as alias the IP Address or the simple DNS name of the proxy server which is used in the Next Proxy HT TP S Host input fields Set the username and the password in the Kerberos configuration Enable both checkboxes in such a
171. eter ULE oe misconfigurations Test HTTP S Request executes URL calls whose input data can be iem entered manually Can be used to examine the HTTP protocol Unescaped Text A MODO Examine SS Coran HITS Sarnen Protocois d YS HTTPS Server https SSL v2N3 TLS v use Proxy Apply next proxy configuration from Personal Settings import URL HTTP Method GET HTTP Protocol Version 1 1 v Browser Type Firefox 3 0 v Browser Language English v C Basic Authorization Username Passwort NTLM Authentication Apply NTLM configuration from Personal Settings PKCS3H 2 Client Cert Apply active PKCS 12 client certificate from Personal Settings use Proxy Apply next proxy configuration from Personal Settings Transmit Cookie CI POST Data Apply veb Tool Done 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 164 171 Apica ProxySniffer V5 0 User s Guide English Edition 15 an dak Load Test FIM Proxy Sniffer follows the philosophy that almost all functionality can be done by using the GUI without requiring programming knowledge Nevertheless itis also possible to modify the automatically generated load test programs manually You can freely modify the program on this second level according to your needs however you should remember that the modifications are not protected from being overwritten when
172. etwork Connection 1 The URL call failed during the opening of a network connection to the web server Transmit HTTP Request 2 The URL call failed during the transmission of the HTTP request data Wait for Server Response 3 The URL call failed while waiting for the first byte of the HTTP response data from the web server Receive HTTP Header 4 The URL call failed while receiving the HTTP response header from the web server Receive Content 5 The URL call failed while receiving the HTTP response content from the web server HTML data images Close Network Connection 6 The URL call failed while closing the network connection to the web server All Done 7 The URL call itself completed successfully all data transmitted and received but the received HTTP status code was incorrect or the received MIME type text html image gif was incorrect or an error was detected inside the received content data Enhanced HTTP Status Codes In addition to the normal HTTP status codes range from 100 599 the Proxy Sniffer load test program generates some additional HTTP status codes in error situations that are not directly related to the HTTP protocol These additional HTTP status codes have negative values Enhanced HTTP Status Code Meaning 98 An internal network error occurred at the client side load test resource problem There are commonly not enough free TCP client sockets available on the Exec Agent and you have to tune the s
173. etwork traffic which is generated by this load test job measured in megabits per second 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 90 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 1 1 5 Real Time Comments Description supports to enter comments during the load test execution Real time comments are notes or hints which you can enter during the load test execution Y PRX Job 205 Mozilla Firefox oe Project Navigator Execute Load Test Job Parameter Tezt1 u 60 d 2700 4 60 Selen 200 maxioops O amping 15 percpage 100 percurl 20 maxemmem 20 Teri Aanad a 19 Det 2004 22 23 49 ECT 7 Planned tezi duration 500 min FS min elaered server PELLI Tdi These comments are later displayed inside all time based diagrams of the load test result detail menu see chapter 10 1 statistic Sampling Interval 15 sec Network Throughput per User 16 8 kBytes sec Test Scenario Warning Diagram Response Time per Page Re Diagram Response Time Percentiles Diagram Top Time Consuming URLs gt Dia gt Diagram Web Transaction Rale Diagram Users Waiting for Response gt Dia gt Diagram Network Throughput gt Diagram HTTP Keep Alive Efficiency gt gt Diagram Error Types Diagram Number of Errors per Page Dia Heb Transaction Rate Serpling Interval 15 Seeands number of successful completed url calls per second it er secon
174. f Add Field b none mini lut WebUserSettings jsessionid bu3fy0bbj1 backContext L3Bye HRvb 2wvc 2VydmxldC 9X ZWJNYWIuTWvudo oa U2 ipntooliservlet WebMainMenu jsessionid bu3fyObbj1 action navigate amp currentDir 1 U3 WebPnmRecordDataMain jsessionid bu3fyObbj1 f_id 10368f_version_nrO0 amp genericBackContext L3ByeHRvb After this the complete extract and assignment definition appears as follows 4 a an password User var A Input File 15 4 HTTP Request Content Paramete http 127 0 0 1 7990 dfischer webadmininterFace DataRecordVarHandlerIframeWeblet action assignV ar Menu amp var AssignerType 218 amp assignParamName username amp recordedValueB64 zmlzY 2hlcg 77 username user var A Input File 15 amp HTTP Request Content Paramete 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 46 171 Apica ProxySniffer V5 0 User s Guide 7 2 1 More Hints for using Input Files English Edition Because the extraction of variables from Input Files is completely independent from their assignments there are many other scenarios where Input Files are useful for example o Testing search forms where the search text is read from the first variable of a line and the response content test of the search result is compared to a second variable on the same line o Toset the emulated user s think ti
175. far Parallel URL Runtime Execution Total Parallel Executed URLS 82 Total Number of Pages 119 General Load Test Configuration Max Parallel Threads per User 6 recommended Apply Apica ProxySniffer English Edition It might be necessary to consider variables which are assigned or extracted to or from URLs meaning that a variable cannot be extracted from a parallel executed URL and then assigned to another succeeding URL which is also executed in parallel on the same page Therefore to avoid unexpected runtime errors during a load test we recommend that you always use the Auto Configuration for Parallel URL Runtime Execution which considers almost all aspects In principle you can configure the Runtime Execution Behavior for each URL separately However such a manual configuration may be overwritten when you invoke later the Auto Configuration To avoid this behavior you can protect a manual configured URL by enabling the checkbox Disallow Auto Configuration for this URL The configuration for protected URLs is shown in the Main Menu in bold letters gt PRX Generate HTIP S Load Test Program Mozilla Firefox 127 0 0 1 7990 dfischer webadmininterface PopupCreateLoadtest Weblet Generate HTTP S Load Test Program TEST_01 Java Code Model large w apply heuristic methods from recorded session to check received content v Character Encoding 150 8859 1 v Generate External Files for X
176. flate Accept Charset ISO 8859 1 utf 8 q 0 7 qz 0 7 Keep Alive 300 9 Content Type applicationi www form urlencoded 10 Content Length 48 11 Connection Keep Alive 12 Cookie JSESSIONID a06wox8D0r8 ola mi a w m HTTP Request Content 1 LoginFlag 1 2 username fischer 3 password HTTP Response Header 1 HTTP 1 0 200 OK 2 Content Type TEXT HTML 3 Expires 0 4 Cache Control no cache mustrevalidate 5 Pragma no cache Py B Set Cookie2 JSESSIONIDzI44hjn80w2 Versionz1 Discard P ath prxtool of 7 Set Cookie JSESSIONIDzI44hjn80w2 Path prxtool S 8 ServletEngine Tomcat Web Server 3 2 3 JSP 1 1 Servlet 2 2 Java 1 3 1 08 Windows XP 5 1 x86 java v or Sungfrosystems Inc t top of page Proxy Sniffer Project Master Directory Browser Mozilla Firefox x File Edit View Go Bookmarks Tools Help A gt E 8 A E http 127 0 0 1 7990 dfischer webadmininterface AnalyseLoadtestDownloadContentWeblet key 30dd9F8F 1 fe957e4d24c645dbe01cc77 amp dataRecordIndex 158failureInfoIndex 0 amp direct 1 F v s e m s pm m Logout Setup Navigate Home Back Reload o F Root Directory Filename Version Type Size Bytes Owner Date EB unprotected 06 Dec 2002 22 45 3 ri i unprotected 14 Feb 2006 17 17 51 cl E unprotected 06 Aug 200
177. ge 115 171 Apica ProxySniffer V5 0 User s Guide PRX Response Content Throughput Charts Mozilla Firefox DER as http 12 7 0 0 1 7990 dfFischerwebadmininterFace Popup nalyseLaadtestDisplavResponseThraughputaraphsweblet key 5323Fd41994fFb728c21aab f 1db95ds1 amp EargetIndex z5 amp data ABP M It 2 Praxy Sniffer Web Admin Response Content Throughput Charts Maximum Acceptable Response Time in Milizeconds O 347 uncontfiqured accept any Apply Response Content Throughput Sample Ho 1 25 GET https www post inance ch 443 en priw prod html Number of Loaded Charts Average Insufficient Throughput Rate Maximum Acceptable Response Time Exceeded total bytes 5 5 ms 44 mz soo 0 0 0 1 6 2 4 3 2 4 0 4 8 5 6 time to receive the response content mz o elspsed time before start of receiving the response content Load Test PF TESTHI Request Start Time 13 Apr 2611 28 69 38 Response Content Throughput Sample Ho 2 25 GET https www post i nance ch 443 en pr iveprod html total bytes 154 ms 45 mz 40 B an 100 120 140 Er time to receive the response content 45 ms Aa elapsed time before start of Pecejwina the response content Load Test PF_TEST amp 1 Request Start Time 13 Apr 2611 26 89 31 2012 2013 Ingenieurb ro David Fischer AG Switzerland S ex Help Refresh Close 10 e 0 0 Sample Ho 1 Met Con Reg Trim
178. h Edition 9 1 1 3 Error Overview Diagrams Real Time Y PRX Job 188 Error Overview Mozilla Firefox Sele Description displays during the B http 127 0 0 1 7990 dfischer webadmininterFace PopupDireckoryMavigatorDisplayLiveErrorOverviewWeblet jobId 1888execAgentId 124030861524068filePathB64 QzpcRa9rdWw 1IbnRIIHvuzCBFawSzdavsbHyuzzwu 7 load test at real time an overview ax ara ier a Error Overview Real Time Remote Exec Agent Job 188 He Hereby Close about all occurred errors Failure Diagrams The first diagram shows an Real Time Profile 2S Eonar Tce cado Gace Agent Job 188 overview about what kind of errors D RM NL Md Current Time did occur at which time counted 3 3 overall URLs and measured since the load test was started This basic error information is always captured in every case also if no more memory is left to store full error snapshots jagrams First Error Snapshots Latest Error Snapshots All failed URL Calls Session Failures only Error Snapshot Memon 20 used Hot Found 464 667 Errors Hetwork Connection failed 377 Errore Content Test Failure String Not Found 351 Errors Hetwork Connection aborted by Server 15 Errors vates scolis ti Sanani The succeeding diagrams which are un shown per web page provide only BEENDEN information at which time errors did occur The tables at the right side of Y the diagrams are showing the Load Test Test 13 Oct 20
179. hanged between the web application and the web browser in such a way that if you repeat the same web surfing session the values of these parameters are changed by a more or less random algorithm If you use during a load test these burned in values of dynamically generated server side CGI or form parameters the load test will fail A good example of this is the VIEWSTATE parameter used by Microsoft web servers The solution to this problem is that the values of these dynamically exchanged session parameters must be extracted at runtime during the load test and then assigned to the corresponding parameters of succeeding URL calls To make this task easier Proxy Sniffer provides the Var Finder menu You can invoke the Var Finder either from the main menu or from the Var Handler PRX Main Menu Mozilla Firefox al Ioj x File Edit View History Bookmarks Tools Help M n CE S a amp amp etx pC GY CF ntedi27 00 1 7990 Y Gr PD gt Help wipe Overall Lend Test Sesion Refresh Close N o gt Main Menu amp OO i we Proxy Sniffer at Web Page Personal Project Generate Analyse Refresh Web Admin Professional Edition Help Tools Scammer Settings Navigator Load Test Load tests Display n ESO 73 no vars defined EJ Add File Ll m c T 0 Add Field 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 54 171 Apica ProxySniffer V5 0 User s Guid
180. he Freese Project Navigator e ssX web server you have create the ZIP archive manually by using the ZIP GersomtsProsStriyTesstia zp secante fen ena nasa functionality of the Project Navigator The corresponding instructions are displayed in the lower part of the window I Gy A og Peg mamak Hint if the date of one of the files which has been added to the ZIP archive is newer than the date of archive itself you will be asked at the start of the load test if the archive should be automatically re zipped This means that you only have to create the ZIP archive once afterwards you can just start C RepartTemplates I 8criptExamples iil Trash 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 74 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 Executing Load Test Programs After the load test program has been called by the Project Navigator you must enter the test input parameters for the test run a single execution of the load test program is also called test run The most important parameters are Number of Concurrent Users and Load Test Duration You should also enter a small comment about the test run into the input field Annotation Input Fields save as template stores all load test input parameters PRX Execute Load Test Mozilla Firefox additionally inside a XML tem plate see chapter 9 5 Later ES http 127 0 0 1 7990 dfischer webadmininte
181. he Input File works correctly This can be done by clicking on the B icon for an Input File definition dee l eX ry u kan Var Handler Test Input File Help roject_ Refresh Close no vars defined get Var3 Vara Vars Var amp Var vars Var8 Var10 Vart1 Var 2 Var13 Var14 Var 5 Varl6 Var17 ya 8 varig Var20 Add File hjgzg8gqgiu Pi ee LL J Lo abc0987 geheim higzg8gggiu maria 3 abc0887 geheim hijgzg8gggiu maria abc0887 geheim higzg8gggiu maria abc0887 geheim hjgzg8gggiu maria Speyer bunts td Gm 4 File Scope new line per user A A M DD ee A ee Afterwards you can extract variables from the Input File by clicking one or more times on the variable extractor icon al abc0987 Extract New Var from Input File first 20 input lines tested EOF Action reopen file userAccounts txt Dome File Scope new line per user Line Column x 1 Input Fields Var Name username l l l l Line Column the column number of a line from which the variable is extracted 1 2 3 Var Scope user var required Var Name any new variable name but with the following naming restrictions Zu The name can only contain the characters A Z a z 0 9 and _ Spaces are not permitted The name must not start with an u
182. header for example a 302 Found in Response Content Line 30 Position 511 list aspxiid 451047 gt Agenda Arna Found in Response Content Line 59 Position 100 _det aspx id 451047 amp aid laam redirection with URL CGI parameters or from the HTTP response content for example HTML or XML data Item 35 GET htips f control test atilevel agenda det aspx id 451047 amp aid 1 amp c 1 A helpful hint is displayed near the arrows in this Found in Request Header Line 1 Position 31 det aspx id 451047 amp aid l amp c l example Eound in Response content Found in Response Content Line 8 Position 61 Het aspx id 451047 amp amp aid l am l Found in Response Content Line 29 Position 511 list aspx id 451047 gt Agenda Arna l s Found in Response Content Line 29 Fosition 580 det aspx ids45l047 amp aid lac l Zz As can be seen In this example the parameter must be Found in Response Content Line 67 Position 84 tepl aspx id 451047 document extracted from URL 19 and assigned to the URLs 33 35 and 38 Item 38 GET https f control test atlevel agenda_det aspx id 451047 2a1d 14c 1 solid 1 Found in Request Header Line 1 Position 31 det aspx id 45l047 amp aid l amp c la amp so Cw 2012 2013 Ingenieurb ro David Fischer AG Switzerland Done 2 Apica ProxySniffer V5 0 User s Guide English Edition Click next on the first blue arrow Then the URL Details Var Handler menu is displayed Click on the search
183. ich prevents on the one hand network peaks and ensures on the other hand that no jerky video playback occurs Page 114 171 2012 2013 Ingenieurb ro David Fischer AG Switzerland Apica ProxySniffer V5 0 User s Guide PRX Response Content Throughput Charts Mozilla Firefox PR http 1127 0 0 1 7990 dfischer webadmininterFace Popup4nalyseLoadtestDisplayResponse ThroughputoraphsWeblet qee Proxy Sniffer Web Admin Response Content Throughput Charts 38 GET http 92 16 4 5 80 endoc ProxySnifferUsersGuideV45En pdt Maximum Acceptable Response Time in Milizeconds 30000 Q uncontfiqured accept any Apply Number of Loaded Charts 30 66 0 Maximum Acceptable Response Time Exceeded 83 3 Average Insufficient Throughput Rate Response Content Throughput Sample Ho 1 Max Acceptable Response Time 38 BBB mz 391 GET http 4 192 156 4 5 8B z en docz PFroaasuSnifferllaerziuideld5En pdt total bytes 28 95 0 mz 62 ms 10 000 000 4000000 000 000 E a 000 000 cree 5 000 000 4000000 3000000 000000 11000000 Muser data Mchunk protocol gt BPrequired throughput o time milliseconds 000 2 000 12 000 ir 000 20 000 2z4 nuu 2 000 32 000 36 000 yo 000 time to receive the response content Insufficient Throughput Rate B BE B 2 mz A elapsed time before start of receiwing the response
184. ieurb ro David Fischer AG Switzerland All Rights Reserved Page 31 171 Apica ProxySniffer V5 0 User s Guide English Edition Comments at the start or within a line are supported and begin with a hash character All values can be also be optionally enclosed with double quotes Example default settings defaultURL http www d fischer com autoPageBreak 4 3 50 POST search 200 query address amp x 5 responseContentCheck phone number responseContentType text html GET http www proxy sniffer ch clients html 200 GET hotlinks index html 200 GET jobs 301 GET http www proxy sniffer com logo gif 200 responseContentType image gif Commands e userAgent lt browser type gt Allows the setting of a new web browser identifier to be applied for all URL calls The default value is Mozilla 4 0 compatible MSIE 7 0 Windows NT 5 1 NET CLR 1 1 4322 NET CLR 2 0 50727 e defaultURL lt URL gt Allows the setting of a default absolute URL to be used as the basis for all following URL definitions which are specified in relative format Only the protocol the host and the TCP IP port of the absolute URL specified are used in building the full URL in combination with the relative URL e defaultRequestContentDirectory lt directory gt Allows the setting of a default local directory from which request content files are read This command can be used in combination with the URL option requestContentFile e defaultReq
185. ieurb ro David Fischer AG Switzerland All Rights Reserved Page 47 171 Apica ProxySniffer V5 0 User s Guide English Edition After the User Input Field has been defined it can then be assigned to the host name in this example You can click on any recorded URL in the main menu which contains the correct host name that is the host name which you want to make variable Then click on the assign icon Y in the HTTP request header http 127 0 0 1 7990 Proxy Sniffer URL Details Var Handler Mozilla Firefox Les Proxy Sniffer Cweb Admin URL Details Var Handler Help Project Search Generate Sav Navigator q e Overall Load Test Session TAE amp amp Refresh Close 1 2 Host 192 16 4 5 3 User Agent Mozilla 5 0 Windows U Windows NT 4 5 GET Y format css HTTP 1 1 Accept text css q 0 1 Accept Language en us gt Server Apache Last Modified Thu 04 May 2006 22 46 06 GMT ETag 1 2c 1212 2b66e780 m 1 HTTP 1 1 200 OK 2 Date Fri 02 Jun 2006 17 51 32 GMT 3 4 5 es os 1 body p a ulol li table trtd form input select option font size 12px font family Verdana Arial Helvetica sans serit l 2 header font size 1 Opt font family Verdana Arial Helvetica sans serif color 000066 3 header2red font size 1 Opt font family Verdana Ari
186. ificate authentication on the load test program If you choose the option use common active PKCS 12 certificate from Personal Settings menu see chapter 3 1 2 3 the same client certificate is used for all simulated users and this certificate will be automatically transferred into the source code of the load test program If you choose the option apply individual PKCS 12 certificate per user from input file each simulated user uses its own certificate in which case you must manually create two Input Files pkcs12auth txt a text file which contains a line for the PKCS 12 filename and the password of the PKCS 12 file for each simulated user These two elements on each line must be separated by semicolons pkcs12certs zip a zip file containing in one archive all PKCS 12 client certificate files which are referenced in pkcs12auth txt Both Input Files must be located in the same directory as the generated load test program After compiling the load test program inside the Project Navigator you must first ZIP the compiled class of the load test program together with the pkcs12auth txt file and pkcs12certs zip files Then you must execute the zipped archive itself as load test program e Program Description optional arbitrary text description of the load test program The description will be transferred to the generated Java code Hint instead of clicking on the Continue button you can also just press the enter key The following dial
187. ile Scope new line per user e ee Line Order sequential User Input Fields Adel Input Field Comment Tag Var Delimiter semi colon ene Trim Extracted Values EOF Action reopen file vt Load Test Plug ins reopen fie i required use a simple file name with no path nana Recommended file extension txt ar dat The input file must be located in the same directory where the generated load test program resides Add Please note that the name of the input file should have the file extension txt recommended or dat 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 42 171 Apica ProxySniffer V5 0 User s Guide English Edition The following options are available when defining an Input File File Scope Defines the scope of the variables which will be later extracted from the Input File global one line this scope is usually not useful for Input Files because only one line will be read during the entire load test at the start of the test new line per user a new line will be read for each simulated user during the load test This is the proper scope for reading user account data username password The line remains the same for all executed loops of the same user new line per loop a new line will be read each time an simulated user executes a loop The new lines are distributed over all users and loops NL per inner loop a new line will
188. ing tools We also provide support and think time which will be applied during the load test This is the time anoe eunsundiho deona dieen na ieaie which a human user needs to study the content of the web page before clicking on the subsequent page The percentage value near AAnnouncemente Prepare a load test script for breakfast the time is the randomized range of the think time which will be 3 calculated new every time for each user and page call during the test Entries 3 Prepare a load test script for breakfast This means that concurrent users will not use the same think time internet Click on the Stop Recording icon in Web Admin after you have finished recording all web pages 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 11 171 Apica ProxySniffer V5 0 User s Guide English Edition 3 Further Hints for Recording Web Surfing Sessions 3 1 1 Support of Technical Client Programs and Web Services SOAP and XML Data Communication over HTTP S A web browser is only required in order to use the Web Admin GUI This means that you can also record web surfing sessions of non web browser based technical client programs which exchange ASCII SOAP or XML data with the web server by using the HT TP S protocol Please note that you have to configure the proxy settings of the technical client program to record web surfing sessions proxy host 127 0 0 1 proxy port 7999 for HTTP and proxy port
189. into a load test program Normally you should only have to enter the name of the load test program with an annotation or description of what the program does without having to choose or modify any other options gt http 127 0 0 1 7990 PRX Main Menu Mozilla Firefox File Edit View History Bookmarks Tools Help E 7 o FAS t See Proxy Sniffer Main Menu Sa i f Web Page Personal Project Web Admin Professional Edition Help Tools Scanner Settings Navigator lb a f se Refresh ests Display Recorded Items 40 Recording State STOPPED Search Overa x 0 0 HO Page 1 Start Page user s think time O seconds 0 Item Test Offset Position Content Size Time HTTP Request HTTP Response x 1 1 0 00sec EN 31 263 bytes 422 ms I GET http 11192 16 4 5 200 OK CUNA 4 4 PRX Generate HTTP S Load Test Program Mozilla Firefox 127 0 0 1 7990 dfischer webadmininterface PopupCreateLoadtestWeblet Apica ProxySniffer Generate HTTP S Load Test Program Help Project Navigator Refresh Close Start Stop Reset Recording Recording Recording PRX Generate HTTP S Load Test Program Mozilla Firefox 127 0 0 1 7990 dfischer webadmininterface PopupCreateLoadtestWeblet action createPrograme amp classname TEST_O1 amp codeStructureModel 2 amp selectConte ABP y Apica e e x ProxySniffer Generate HTTP S Load Test Program Help roject Back Refresh Close ject Navigator A Apica Ja
190. ion Resp Throughput Chart per Call as part of the Additional sampling rate by URL call when starting the load test Furthermore you should configure a maximum acceptable response time in order that Proxy Sniffer can calculate and compare the necessary network throughput English Edition This feature is especially useful for Web pages that contain videos and allows to detect if jerky video playback occurs during viewing of a video respectively to diagnose if enough network bandwidth is available for all simulated users so that the video can be viewed by each user without interruption However this feature can also commonly used as a reference for the optimization of any response data The corresponding charts are showing in different colors the times elapsed for receiving fragments of user data in red color and the times elapsed for receiving the overhead data of the chunked protocol in blue color PRX Execute Load Test Mozilla Firefox gr EEx f http 127 0 0 1 7990 dfischer fwebadmininterface PopupDirectoryMNavigatorStartLoadTestWeblet filePathB64 0zpcRG9rdWAlbnRITHHYuZCBFawWSzdGYsbHWL wo BH we x Help Jobs Refresh Close PRX Result Detail Mozilla Firefox SEE z http 127 0 0 1 7990 dfischer webadmininterface PopupAnalyseLoadtestDetails Weblet dataTickOutputFormat 1 amp action percentile amp subAction displayDataTicks amp key c15561b28b9aae1cl CUSSU Tests yuuruveve YO mpi corr TO II DO users
191. ion of the test run before you Page 2 login maske m eae eee generate the PDF report However all retroactively entered real Page 4 link zahlungen time comments are not permanently stored inside the result data Page 5 link inland chf Page 6 1 eingabemaske Page 7 2 eingabemaske Page 8 link konto Page 9 link auftragsuebersicht Page 10 auftragsuebersicht suchen Page 11 lupe einzelzahlungen Page 12 lupe pendent Page 13 loeschen Page 14 logout Data Source Test Result File Test01_c1_285ep09_1 42434_1600u prsres lt TT gt Done Ka 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 110 171 Apica ProxySniffer V5 0 User s Guide 10 1 2 Diagram Response Time per Page Displays bar chart diagrams about the average response times and the 90 percentile value of the web pages Displays also diagrams about the response time progression of the web pages x x Average 90 Percentile Detail per Page seconds Response Time per Page Average Page 1 6 Load Test Test 1 27 Nov 2668 23 19 49 User 26 Hint Click inside the diagram on the Il bars to display details 2012 2013 Ingenieurb ro David Fischer AG Switzerland Start Page Special Features Download Customers amp Parners Buy Support FAQs Save image to disk Average 90 Percentile Detail per Page Page 1 starts
192. ion sessionTimePerPageDiagram 6 9 6 See Proxy Sniffer 1 icti 3 E X Web Admin Load Test Result Detail Statistics and Diagrams PRX Analyse Load Tests Select and Compare Results Mozilla Firefox I Report Export Close LF http 1127 0 0 1 7990Jdicherwebacmiinterface PopupAnalyseLoadtestWebletloadFleB64 02pcR Gai BORRIINVUZCEF aWS2dGVsbHVUZZVUXGI 146 eee A o Advanced Test Parameter Measured Results per Single User per Loop Overall Test Results Startup Delay per User 200 ms AV Session Time per Loop 12 58 seciloo Web Transaction Rate 8 ul calls sec Proxy Sniffer d p P p p p gt 3 Web pt dmin Analyse Load Tests Selecteat Compare Results Request Timeout per URL 60 sec AY Response Time per Pagi 0 12 secipage Session Failure Rate 15 0 A A Statistic Sampling Interval 15 sec Network Throughput pergfSer 25 2 kBytes sec Total Network Throughput 1 95 MBit sec Total Transmitted 60 MB N Upload File Test Scenario Results per URL Call Overview DD Results per URL Call Details Diagram Response Time Percentiles gt Diagram Top Time Consuming URLs gt Diagram Concurrent Users Diagram Session Time Diagram Web Transaction Rate Diagram Users Waiting for Response gt Diagram Completed Loops Diagram TCP Socket Connect Time Diagram Network Throughput Diagram HTTP Keep Alive Efficiency gt Diagram SSL Cache Efficiency Diagram Sessi
193. ional Sampling Rate per Page Call 100 Additional Sampling Rate per URL Call H Add recommended y Debug Options none recommended y Additional Options NS fe vzwarm s y Annotation CT We recommend that you also read the Application Reference Manual E cua aiid adele 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 171 171
194. is called The reason for this is that the local proxy server cannot not recognize when a web page starts and when it finishes The local proxy server only sees singe URL calls such as requests for HTML data or image files Adding a page break manually here is necessary in order to record the session properly Use the following strategy during the recording of a web surfing session over several web pages 1 First plan which URL or hyperlink you will call and record next but don t click on it just yet 2 Then insert a page break comment into the Web Admin GUI Enter a comment describing the expected result of the next recorded web page 3 Now call the desired URL by clicking on a hyperlink or submitting a form Repeat this strategy for each web page that you call during recording Remember that you must insert the page break before you click on the next hyperlink or submit the next form First Web Browser Window Web Admin GUI Second Web Browser Window Web Application PRX Main Menu Windows Internet Explorer e eT F 127 0 0 1 p Welcome safearea com au Windows Internet Explorer P Favoriten B PRX Main Menu Welcome ad H 7s Proxy Sniffer Main Menu Web Admin Professional Edition Google Sitemap Generator l l l Welcome Safe Area Pty Ltd The time in seconds near the page break comment is the user s We re the australian distributor of Proxy Sniffer one of the leading web load test
195. ish Edition 12 2 2 Message Headlines The Message Headlines for all Alert Notifications can be configured and support placeholders The values of the placeholders are calculated at runtime and are replaced within the message headlines Generic Placeholders which can be used in every type of alert notification are e timestampk The current date and time when the alert notification was created Example 01 Jun 2010 13 45 38 ECT e generator The name of the Exec Agent load generator which releases the alert notification e jobld The job ID of the Exec Agent job e programName The program name of the Exec Agent job Specific Placeholders e During the Execution of a Job Information at Every Interval and at the End of a Job Information o sessionFailureRate The measured session failure rate in percent o avResponseTimePerPage The measured average response time per page in seconds e During the Execution of a Job and at the End of a Job if the Session Failure Rate is greater than 96 o sessionFailureRate The measured session failure rate in percent o sessionFailureRateLimit The configured threshold for the session failure rate in percent e During the Execution of a Job and at the End of a Job if the Average Response Time per Page is greater than seconds o avResponseTimePerPage The measured average response time per page in seconds o avResponseTimePerPageLimit The configured threshold for the average respon
196. ition O O All without Resp Content additionally collects the request headers the request content and the response headers of the URL calls All full URL Snapshots additionally collects all data of the URL calls Warning capturing additional URL data takes much memory and uses also much CPU Therefore the test duration should not exceed 10 minutes if you use one of these add options in combination with 100 sampling rate per URL call Reducing the sampling rate to 10 may allow a load test duration up to 30 minutes Hint these additional URL data can be displayed and or exported in the form of an HTML table when the test run has been completed see Chapter 10 1 5 Debug Options these options allow you to debug the inner workings of the load test program The result is written to the job_ out file which is usually only used to analyze internal errors in the load test program O none recommended recommended default value Note that all measured performance data and all error snapshots are already stored inside the result file prxres therefore special debug options are not necessary in order to analyze the load test result debug loops including var handler writes the log data of all executed web surfing sessions loops to standard output including information about dynamically extracted session parameters and Input Files debug headers amp loops includes the above option debug loops and in addition
197. j servlet prt portal prteventname Navigate prtroot pcd 3aportal content Zzfevery user TOO0000 200 OK TEXT HTML 116599 bytes 7892 ms TOOO000 lt lt lt windowId WID1149857318747 TOO0000 156 GET http com 50100 irj servlet prt portal prtroot com sap portal ui uiservice treepreload images irj porta TOO0000 200 OK TEXT HTML 1146 bytes 60 ms TOO0000 157 GET http com 50100 irj servlet prt portal prtroot pcd 3aportal content 2fcom sap portal migrated 2fep 5 0 2 TOO0000 200 OK TEXT HTML 46244 bytes 6249 ms TOOO0000 158 GET http com 50100 irj servlet prt portal prtroot pcd 3aportal content 2fcom sap portal migrated 2fep_5 0 2 TOOOOOO 200 OK TEXT HTML 7410 bytes 221 ms TOO00000 lt lt lt htmlbdoc_idl htmlb 4461 T0O00000 lt lt lt htmlbevt_frml htmlb 4481 htmlb 3124 TODOOOO 159 GET http com 50100 irj servlet prt portal prtroot pcd 3aportal content Zfcom sap portal migrated zfep 5 0 2 TOOOOOO 200 OK TEXT HTML 7409 bytes 70 ms TOO0000 java lang NullPointerException TODODOO at MssEssMgrl ll fischer executePage 4 MssEssMgrl ll fischer java 6877 TODODOD at MssEssMgrl ll fischer execute MssEssMgrl_1l_ fischer java 420 TOOOOOO at MssEssMgrl ll fischer run MssEssMgrl 1l fischer java 14466 TODODOD at java lang Thread run Unknown Source TOO0000 error unable to extract var htmlbdoc_id Z from html form parameter Finally the activity of all users at the
198. k connection has been opened to the web server Av Wait average time per URL call waiting for the first byte of the HTTP response header from the web server measured after the HTTP request data have been transmitted Av Header Rcv average time per URL call receiving the HTTP response header from the web server measured after the first byte has been received Av Content Rcv average time per URL call receiving the HTTP response content from the web server HTML data images etc measured after the HTTP response header has been received Min Time smallest ever measured time of the URL call request response Av Time average time of the URL call request response Max Time highest ever measured time of the URL call request response AV Size average size of transmitted and received data per URL call or per web page Av Throughput average network throughput per URL call request response 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 117 171 Apica ProxySniffer V5 0 User s Guide 10 1 5 Diagram Response Time Percentiles English Edition This screen contains per web page and per URL the response time percentile diagram These diagrams display a cumulative statistical distribution of response times but are only available if an Additional Sampling Rate per Page Call and or an Additional Sampling Rate per URL Call option was set when starting the test run Chapter 9 and a
199. k condition but inclusive incrementing the inner loop iteration counter All Rights Reserved Page 37 171 Apica ProxySniffer V5 0 User s Guide English Edition In addition it is also Supported to report a red fatal error after all iterations of an inner loop have been executed no break was made in an iteration before the last iteration If such a red fatal error is reported the simulated user will abort the current Outer Loop and will start the next Outer Loop PRX Inner Loop Conditions Mozilla Firefox EB http 127 0 0 1 7990 dFischer webadmininterFace PopupManageInnerLoopConditionsWeblet gos S 26 e Ad Conditions for Inner Loop loop 1 Help Search Refresh Close Text Input Fields List of All Conditions for Inner Loop loop 1 No Condition of inner loop loop 1 exegit Ok New Condition added Add New Condition Break Inner Loop ifthe HTTP Response Code ofltem 1 v is 200 OK MR v Check Cgsfffitiopgfrier extracting variables Break v Inner Loop ifthe Response Content ofltem 1 v contains the Text Fragment i OR ivl Check nditig after extracting variables v Break Inner Loop before execution ofltem 1 v ifthe Value of the Variable vinnerLoopCount is equal to A YA neck Condition immediately Break v Inner Loop after execution of tem IE ifthe Value of the Variable vinnerLoopCount v is equal to
200. key Name loop var amp Input File m 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 63 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 7 HTTP File Uploads If a recorded web surfing sessions contain HTTP file uploads you can also use a variable for each file upload which allows to select the uploaded file dynamically during the load test Such a variable is often extracted from an input file whose lines contain different file names without file paths 1 POST jprxtool sendetWebMainvenu HTTP 4 1 Q plaadFile tE plus Thalldomidevirin C plus Tr Assign Var to Multipart Form 2 Accept image gif imadgehbexbitmap image jped ima Download Content i aeg d P JEREJ 3 Assign from Var 3 Referer http www d fischer com 8 8D pndaol servieb filename CAScratchisniffer test bd Pa AcceptLanguage de ch HTTP Disposition 3 E dnamicFileMarme Assign idynamicFileMame user var En amp Input File km fm E iz e e IE A El Content Type TERT HTML GET Jpntoal servietA ebMainMenu FileMameaList tt 3 Expires O HIDDEN BbeurentDir 1 H 4 File scope new line per user Pa Cache Cantral no cache mustrevalidate Note before you start the load test you have to place all files which should be uploaded into the same project navigator direc
201. kie Injector Sets a Cookie before or during the execution of a Load Test Get Cookie Value Extracts the value of a Cookie into a GUI Variable The extracted value can be later assigned to a CGI parameter of a succeeding HI TP S Request among other targets Defer Load Test Start Delays the start of a Load Test Program for a configured time expressed in minutes Delay Full Load Limits the load respectively the number of the simulated users for a configurable time After this time is elapsed the load is increased to the originally number of planned users DNS Round Robin Load Balancing Supports web servers which are using DNS Round Robin for load balancing dynaTrace Integration Creates additional data during a Load Test for analysis using dynaTrace Diagnostics The dynaTrace Integration Handbook which is available on our website contain further information about how to integrate Proxy Sniffer with dynaTrace During a Load Test writes the values of up to 6 GUI Variables line by line to a text file The file scope is freely configurable lines can be written per virtual test user per loop execution or per URL call Generic Output File Reads from a meta file a list of input files and assigns each simulated user a own input file The simulated users are reading a new line from their input file each time before they are executing a new loop Input File List Large Input File Heads data from a large input file
202. king on an error counter the detailed results of the corresponding test run is shown PRX Analyse Load Tests Load Curves Mozilla Firefox E http 127 0 0 1 7990 dFischerJwebadmininterfacejAnalyseLoadtestResultDiagr amwWeblet eee go Proxy Sniffer web Admin LASKE Test01 User 2 Start 27 Nov 2008 22 58 55 Duration 2 21 min Session Failure Rate 0 00 Annotation ESTA Test01 User 10 Start 27 Nov 2008 23 03 55 Duration 2 29 min Session Failure Rate 0 00 Annotation ESA Test 1 User 20 Start 27 Nov 2008 23 19 49 Duration 3 08 min Session Failure Rate 35 19 Annotation veo Test01 User 50 Start 27 Nov 2008 23 12 18 Duration 4 30 min Session Failure Rate 100 00 Annotation Test 1 Test 2 passed 13 passed 40 Page 2 Special Features passed 13 passed 40 Page 3 Download passed 13 passed 40 Page 4 Customers amp Parners passed 13 passed 40 Page 5 Buy passed 13 passed 40 Page 6 Support FAQs passed 13 passed 40 on the icons to display details Analyse Load Tests Load Curves Session Failures Page 1 Start Page passed 48 falled 6 passed 25 fall passed 44 failed 4 passed 1 failed 6 passed 42 failed 2 passed 39 falled 3 passed 36 falled 3 passed 35 falled 1 Test 1 Test 1 failed loops o 5 A time O No Errors 0 00 00 0 00 30 0 01 00 0 01 30 0 02 00 0 02 30 Test 1 27 Nov 2008 22 58 55 User 2 Test 1 27 Nov 2008 22 58 55 User 2 Error S
203. lIndexesOfPage yll pageSize pageSize urlDataRecord getAverageSize pageTime pageTime urlDataRecord getAverageTime cumulatedPageSize cumulatedPageSize urlDataRecord getTotalSize System out println pageName size pageSize bytes time float pageTime 1000 0f seconds total transmitted bytes over all calls cumulatedPageSize System out println loop over all measured url calls and page breaks for int x 0 x performanceDataRecord length x switch performanceDataRecord x getDataType case PerformanceDataRecord TYPE PERFORMANCE DATA long urlSize performanceDataRecord x getAverageSize long urlTime performanceDataRecord x getAverageTime long cumulatedUrlSize performanceDataRecord x getTotalSize System out println performanceDataRecord x getInfoText size urlSize bytes time float urlTime 1000 0f seconds total transmitted bytes cumulatedUrlSize break case PerformanceDataRecord TYPE PAGE BREAK System out println performanceDataRecord x getInfoText break default break catch Exception e e printStackTrace 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 168 171 Apica ProxySniffer V5 0 User s Guide 16 2 Example 2 Extracting Error Snapshots English Edition The following programming example extra
204. le name ombinedTest xml Label of template set md itus u dimi M gumen gee rent Us t d Test E max LUOPS pel Annotation Job Status configured Test Test01 Test Arguments uU 4 d 60 t 60 sdelay 200 maxloops 0 sampling 15 percpage 100 percurl 20 maxerrsnap 40 nolog dl Concurrent Users 4 t Duration 1 00 min gt gt Start Li XML Load Test Template Attributes Absolute file path to compiled load test program class or load test program ZIP archive downlinkBandwidth apinkBandwidth requestTimeout maxErrorSnapshots Limits the number of error snapshots taken during load test execution 0 unlimited Negative value 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 102 171 Apica ProxySniffer V5 0 User s Guide English Edition maximum memory in megabytes used to store all error snapshots counted overall Exec Agents recommended Positive value maximum number of error snapshots per URL per Exec Agent not recommended statisticSamplingInterval Statistic sampling interval in seconds percentilePageSamplingPercent Additional sampling rate per Web page in percent 0 100 percentileUrlSamplingPercent Additional sampling rate per URL call in percent 0 100 percentileUrlSamplingPercentAddOption Additional URL sampling options per executed URL call numeric value 0 no options 1 all URL performance details network con
205. le to export the individual measurements in the form of an HTML table Page 118 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 6 Diagram Top Time Consuming URLs Shows a compilation of the slowest URLs average and 90 percentile response time values and the response time distribution of the slowest URL per page for each page and the response time per media type text html image gif Average 90 Percentile O Most Time Consuming URL per Page Average 90 Percentile Most Time Consuming URL per Page m 16023 ms Page 1 Start Page contains 17 URLs 1 17 URL 431 1 589 topshareware_8831 gif 15 1 318 ms Most Time Consuming URL of Page GET http www proxy sniffer com 80 bullet ok red gif URL 191 1 578 features html URL C15 1 318 bul let_ok_red gif URL 12 1 311 000000 gif URL 33 1 220 SessionCutterPreview gif URL 26 1 208 ExecAgentCluster gif URL 16 1 183 4C9C9C9 14u5 aif URL 40 1 178 download html URL C45 1 156 customersAndPartners html URL 49 1 135 remsa gif URL C52 1 138 cnlab gif URL 5 1 123 wolke2 jpg URL 36 1 120 PercentilePreview gif URL C37 1 115 FinalReportPreview gif seconds Top Time Consuming URLs Average URL Response Time 166 Sampling Rate URL Response Time Percentile 1887 Sampling Rate E151 bullet ok red gif 15 bullet_ok_red gif milliseconds milliseconds 1 6 1 4
206. lect the SSL protocol version to be used to communicate with HTTPS servers encrypted connections e Annotation here you should enter a short comment about the scan You can abort a running scan by clicking on the Abort Scan button PRX Page Scanner Mozilla Firefox N See Proxy Sniffer baf amp web Admin Page Scanner Help Navigator Refresh Close Starting Web Page http www proxy sniffer com ii Scan Status scanning Scan started at 06 Dec 2008 20 31 25 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 154 171 Apica ProxySniffer V5 0 User s Guide English Edition When a scan has completed you should save the scan result to a file The file will be saved in the selected Project Navigator directory and will always have the file extension prxscn PRX Page Scanner Mozilla Firefox SEF LE http 127 0 0 1 7990 dfischer webadmininterFace PopupPageScannerWeblet e ES eee 0 Proxy Sniffer LA z oa 2 9 http 127 0 0 1 7990 Proxy Sniffer Save Page Scanner Result Windows Internet Explorer web Admin Page Scanner Tai Lu ramen a Proxy Sniffer Web Admin Project Navigator Save Page Scanner Result Help Refresh Close MyTests fischer Save Page Scanner Result MyTests Filename proxy sniffer ch 80 08May07 202850 Annotation Starting Web Page http www proxy sniffer com Scan St
207. let key c15561b28b9aae1cF2750c8199684eF5 amp EargetIndex 40 amp data 7 x SEE oc gt S ex Help Refresh Close 1p v24 Iscache3 c youtube com 80 videoplayback sparams id 2Cexpire 2Cip 2Cipbits 2 itags e2Calgorithm 2 burstie2 factor 2Coc 36 3 LIORP SFFR VIF SKNOO V8 8 1IH amp fexp 905 Maximum Acceptable Response Time in Milizeconds 180000 Q uncontfiqured accept any Apply Number of Loaded Charts 10 Average Insufficient Throughput Rate 0 0 Maximum Acceptable Response Time Exceeded 0 0 Response Content Throughput Sample Ho 1 C46 GET http v24 lscache3 c0 youtube com Savi decp layback Fsparamns idkeCexpi rex2eCipseCipbitse 2Citagk eCalgorithme2cy total bwtesz 50 000 000 45 000 000 40 000 000 55 000 000 50 000 000 5 000 000 20 000 000 15 000 000 10 000 000 5 000 000 2 gn0 4 0 0nn bo ooo 30 000 100 000 Ee time to receive the response content 41 mz 45 Load Test Response Content Throughput Sample Ho yoytubedz Request Start Time Hax 12 0 000 elspsed time before start of Feceiwing the response content 2 Max Acceptable Response Time 186 8686 ms 161 6557 ms 41 mz Muser data Mchunk protocol gt BPrequired throughput time milliseconds i4 n is g u 150 000 00 000 Insufficient Throughput Rate 0 Ba 18 Apr 2811 19 36 02 Acceptable Response
208. load curves you must select from inside the Analyse Load Tests menu several test runs which have been made with the same load Test program but with a different number of users Then choose the diagram type Load Curve and click on the Compare button http 127 0 0 1 7390 Proxy Sniffer Analyse Load Tests Select and Compare Results Mozilla Firefox i E nj x oe H GE n uu Analyse Load Tests Select and Compare Results gt E a x 4 Project Directory MyTestsiTrash 5 Upload File Browse File Extension prxres Project Navigator Use Project Navigator to load result files ll A Einkauflnregil 29 Jul 2005 02 55 05 7 19 min 8 18 trisec 1 02 MBit sec A GenerischerTest 09 Jun 2006 17 25 43 an 13 05 min 3 20 tr sec i 0 79 WBitsec Ei 4 MssEssMgri 11 fischer 12 Jun 2006 13 40 01 1 1 38min 1 58 trsec 00 00 0 22 MBitsec A ProwSniffer 1 06 May 2006 19 45 08 10 215min 26 15 trsec 00 3 64 MBitisec Erster Test A ProwSniferd1 OB May 2006 19 49 46 20 2 17 min 51 50 trisec 0 00 T 16 MBit zec Zweiter Test A ProxySnifferm1 DE May 2006 19 53 27 40 223 min 98 66 trsec 0 00 13 72 MBitsec Dritter Test A ProxSniffer 1 06 May 2006 20 04 41 100 2 24 min 224 55 trisec 0 00 31 22 MBitisec Vierter Test xx 123455709 Axx 123455788 A ProwSniferd1 OB May 2006 20 08 13 200 2 41 min 161 55 trisec 0 00 22 15 MBitisec F nfterTest A ProwSniffero1 06 May 2006 20 12 02 400 3 33 min
209. ly used by web applications but contains a long text string Setting the accept header field to reduces the data transfer and makes the load test program smaller Load Test over HTTP S Proxy this option allows the execution of a load test through an outgoing proxy server by applying the next proxy configuration from the menu Personal Settings You should use this option only if you have no direct TCP IP connection between the load test program and the web server Basic Authentication this option enables user specific individual basic authentication against the web server Please note that Proxy Sniffer already automatically supports common basic authentication If all simulated users use the same username and password for basic authentication this option must not be enabled If this option has been enabled you must manually create an Input File named basicauth txt which contains a line for the username and the password for each simulated user These two elements on each line must be separated by semicolons The Input File must be located in the same directory as the generated load test program After compiling the load test program inside the Project Navigator you must first ZIP the compiled class of the load test program together with the basicauth txt file and then execute the zipped archive itself as the load test program Digest Authentication Digest Authentication this option enables digest authentication against the web ser
210. mbedded content types should also be included in the scan Page Scanner uses the file extensions of the URL paths to determine the content type if available because this can be done before the hyperlink of the embedded content itself is processed This saves execution time but it might have the effect that a few URLs for excluded content types flow into the result from scanning because the MIME type of the received HTTP response headers is only used in detecting web pages You can remove these unwanted URLs after the scan has been 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 152 171 Apica ProxySniffer V5 0 User s Guide English Edition completed by using the remove URL form in the Display Result window Content Type Sets Corresponding File Extensions Images Flash CSS JS img bmp gif pct pict png jpg jpeg tif tiff tga ico swf stream css stylesheet js avascript PDF Documents pdf Office Documents doc ppt pps xls mdb wmf rtf wri vsd rtf rtx ASCII Text Files txt text log asc ascii cvs Music and Movies mp2 mp3 mpg avi wav avi mov wm rm mpeg Binary Files exe msi dll bat com pif dat bin vcd sav e Include Options allows you to select or to de select specific file extensions by using the add or remove keyword Example remove gif add mp2 e Max Scan Time limits the maximum scan time in minu
211. mbinations are possible providing maximum flexibility Variables can be extracted by using the Web Admin GUI from the following sources from Input Files whose data are read at run time during the load test chapter 7 2 from HTML form parameters for example hidden form fields chapter 7 8 from values of received XML and SOAP data chapter 7 6 1 from CGI parameters contained in hyperlinks form actions or HTTP redirects chapter 7 8 from any text fragments of received HTML and XML data chapter 7 5 2 from User Input Fields which are arbitrary configurable load test input parameters chapter 7 3 from HTTP response header fields from output parameters of Load Test Plug Ins chapter 7 4 Additionally it is also possible to define stand alone variables which have constant or dynamic initial values chapter 7 9 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 39 171 Apica ProxySniffer V5 0 User s Guide English Edition A variable can be assigned as follows irrespective of how it was extracted to the value of an HTML form field chapter 7 8 to the value of a CGI Parameter of a URL call chapter 7 8 to values of XML and SOAP data of a URL call chapter 7 6 1 to a text fragment of a URL call within the HTTP request header or the HTTP request content chapter 7 6 to the protocol http https the host name or the TCP IP port of one or all URL calls chapter 7 8
212. me variable on a per user basis or on a per loop basis for each user o To control the number of inner loop iterations o To enter user specific data into forms such as an article number during a purchase transaction It is also possible to define several Input Files for the same load test program 7 3 User Input Fields User Input Fields are arbitrary global variables whose values are requested each time a load test is started The following example uses a User Input Field to make the host name of the URL calls variable in order that the same load test program can be executed against a development system and a test system without the need to record two web surfing sessions Var Handler Ga Ub Ld XE Var Handler da Gb E 3G no vars defined m Create new User Input Field m Var Name hostname Input Files add File File Var Label Text Host Name nana Default value wes d fischer com Var Scope global var User Input Fields Add Field N none Create Input Fields Var Name arbitrary new variable name he name can only contain the characters A Z a Z 0 9 and Spaces are not permitted The name must not start with an underline character _ Var Label Text denotes the label description which is displayed on the GUI when starting the load test Default Value the default value of the variable which is also displayed on the GUI when starting the load test 2012 2013 Ingen
213. me HITP aquest 1 0 00sec MD 43779bytes 170ms n GET htt om 200 OK TEXT HTML 2 0 17 see E 4517 bytes m oxy sniffer com formatcss 200 OK TEXTICSS 3019sec _ Y 3 43 bytes i T http y sniffer comRX gi 200 OK IMAGE GIF 4021sec 1 J 234 bytes Y ET http www proxy sniffer com flagGerman gif 200 OK IMAGE GIF 5022sec ME 1 212 bytes ET hitp ffwww proxy sniffer com flagEngla Lal 200 010 IMAGE GIF amp 046sec T 1 9791 bytes ms JB GET http www prony sniffer comiimages en SessionRecorderP gif 200 OK IMAGE GIF 7 0 48 sec 1 9 708 bytes i ET hitp fiveww proxy sniffer com images en VarHandlerP gif 200 OK IMAGE GIF 8050sc T 4 7141 bytes i ET hitp ffwww proxy sniffer com images en ExecAgentClusterP gif 200 OK IMAGE GIF 89052sec L1 J 10904bytes ET hitp vww proxy sniffer com images e i ResultP gif 200 OK IMAGE GIF 10053sc TI 35 bytes ET hitp lifveww proxy sniffer com 000000 gif 200 0 09 IMAGEIGIF 11055sec 12 320 bytes ms i GET http ilveww proxy sniffer comlimages en PointOfCollapse gif 200 OK IMAGE GIF 12 0 57 see T 8270 bytes T http www prowy sniffer cominoProgra al 200 OK IMAGE GIF 13 058see ll Il 88 bytes T ET hitp lwww proxy sniffer com arrow red 12x9 qif 200 OK IMAGE GIF 14060sec Sd 67 bytes hitp bullet ok red g tp fM v prox e D Total 0 62 sec 108 107 bytes 14 A 175 78 rare 02NOV
214. me protocol and the same host name and the same TCP IP port It may be necessary to assign the host name again to https requests if you have recorded a session which uses both the http and https protocols within the same web surfing session Page 48 171 Apica ProxySniffer V5 0 User s Guide The User Input Field will be displayed when the load test program is started A maximum of 12 User Input Fields can be defined http 127 0 0 1 7990 Proxy Sniffer Project Navigator Execute Load Test Mozilla Firefox 1O x tell NON P M Project Navigator Execute Load Test So u x Execute Load Test Job Test01 Execute Test from Host Local Exec Agent y Host Name 11921645 Number of Concurrent Users 1 M Load Test Duration 1 min Max Loops per User unlimited Startup Delay per User 200 milliseconds Max Network Bandwidth per User unlimited Downlink unlimited y Uplink Request Timeout per URL 60 v Seconds Max Error Snapshots per URL 30 m Statistic Sampling Interval 15 v Seconds Percentile Sampling Rate 100 per Page BH per URL Debug Options none recommended y SSL v2Nv3 TLS y Additional Options Annotation gt gt Continue recommended will be displayed as hint in Project Navigator Done 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved English Edition Page 49 171 Apica ProxySniffer V5 0 User
215. n be successfully done against a web server or against an outbound proxy server a so named Kerberos Ticket first must be issued by a KDC by an Active Directory Server However acquiring such a ticket requires that always the full DNS host name of the web server is transferred to the KDC even if only an IP address or a short DNS name alias is used for the HTTP S requests during recording of a web surfing session Therefore you have to add all full DNS host names of all web servers for which Kerberos authentication is required to the Kerberos configuration Example www domainname net is a full DNS host name in contrast to server23 which is only an alias e Additional IP Addresses and Aliases Optional Allows to configure a list of IP addresses and or aliases for the corresponding full DNS host name The entries in the list must be separated by commas or by semicolons The configured IP addresses and or aliases can be used when recording web surfing sessions as well as when executing load tests 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 16 171 Apica ProxySniffer V5 0 User s Guide English Edition e Kerberos Configuration krb5 conf you must configure inside this file the name of the Kerberos Domain and the DNS name or the IP address of the KDC server Active Directory Server Modifications in this file are immediately applied at runtime which means that no restart of Proxy Sniffer is r
216. n contain only URL calls to the web server s you want to test 2 Has the automatically applied content test check for the recorded web pages been correctly configured 4 2 1 Reviewing the Stressed Web Servers Some of the recorded web pages may contain embedded in them images with a size of 1x1 pixels originating from an external web session tracking server Or if you have recorded an encrypted HTTPS session Microsoft Internet Explorer may have first made a call to a server at Microsoft corporation to check the validity of the root certificates instead of directly calling the web server whose web session is being recorded In order to not stress external tracking servers and or Microsoft servers we recommend that you remove these URL from the recorded web surfing session You should also review the host names or the IP addresses of all recorded URLs If you find some unnecessary or unwanted hosts you should remove such URLs by clicking on the red cross near the item number at the left side of the URL Alternatively you can use the host field of the URL filter to suppress any unwanted URL You can enter also an exclamation mark in front of an unwanted hostname to achieve that all URLs from this host are suppressed Additionally several host names can be entered separated by commas 0 HO Page 1 Start Page user s thinktime 0 seconds 0 Position Content Size Time HTTP Request HTTP Response a d io 31 a 2 m
217. napshots Top Error Types Session Failures Session Failures Test 2 Test 2 Error Snapshots Top Error Types failed loops o x j N time O tNo Errors 0 00 00 0 00 30 0 01 00 0 01 30 0 02 00 0 02 30 Test 1 27 Nov 2008 23 63 55 User 10 Test 1 27 Nov 2008 23 83 55 User 10 Test 3 Test 3 failed loops e o c Bl Request Timeout expired 19 Errors 0 00 00 0 00 40 0 01 20 0 02 00 0 02 2012 2013 Ingenieurb ro David Fischer AG Switzerland Error Snapshots Top Error Types Session Failures Hint Click inside the diagrams 6 6 All Rights Reserved English Edition Li 54 U b4ms 2126 ms 305 bytes GET hip www proxy snifer com sU flagberman gil 81 54 0 T40ms 1822 ms 10 481 bytes GET httpvwww proxy sniffer com 80 images_en SessionRecorderP gif 191 54 0 738 ms 2128 ms 10 391 bytes GET httpviwww proxy sniffer com 80 images_enfVarHandlerP gif O 7832 bytes mail 51 2 2150 ms 11 597 bytes GET httpvwww proxy sniffer com 80 images_en measurementResultP gif nal 51 0 2 852 ms 701 bytes GET httpwww proxy sniffer com 80 000000 gif 13 48 3 1034 ms 3 185 ms 13 010 bytes GET httpvwww proxy sniffer com 80 images_en PointOfCollapse gif 14 48 W 1 054 ms 2 823 ms 8 946 bytes GET http hvwww proxy sniffer com 80moProgramming dif na 48 W 1318ms 3 232 ms 740 bytes GET http vwww proxy sniffer com 80 bullet ok red gif 16 48 W 1183ms 3
218. nd English Edition This menu option displays a summary about the performance data of the test runs The following measured values are shown in the Performance Overview Table Passed Failed Loops total number of passed failed loops of the test run Average Session Time per Loop average time of a loop calculated over all simulated users and loops Average Response Time per Page average response time per web page calculated over all web pages Web Transaction Rate Hits per Second number of successfully executed URL calls per second Average Outstanding Requests average of outstanding HTTP S Requests executed at exactly the same point in time Total HTTP S Calls sum of all by the test run executed HTTP S calls HTTP Keep Alive Efficiency percentage of re used network connections SSL Session Cache Efficiency percentage of abbreviated SSL handshakes Average TCP Socket Connect Time average time per URL to open a new network connection to the web server Average Network Throughput average network traffic released by the test run Total Transmitted Bytes total data volume which was transferred during the test run All Rights Reserved Page 140 171 Apica ProxySniffer V5 0 User s Guide English Edition 11 Distributed Load Tests Architecture and Configuration Load tests can also be transmitted and started on remote computers Similarly a single load test can be divided
219. nderline character _ no vars defined Add File ser ccounts tt b am amp File Scope new line per user 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 44 171 Apica ProxySniffer V5 0 User s Guide English Edition The following example shows the definition of an Input File first without the assignment of variables Var Handler OR The read bars with the title texts password and username are the names of the extracted variables The variable scope is shown in brackets next to the title text password user var m Input File The blue left arrow indicates that the value of the variable has been extracted More details about how the variable was extracted can be displayed by clicking on the corresponding magnifier icon username user var A Input File A variable or the Input File definition itself can be deleted by clicking on the red bar Input Files mem The Input File definition can be displayed and modified by clicking on the corresponding magnifier icon yser ccounts td dea a File scope new line per user To finish this example it is now necessary for the username and password to be assigned to the URL call which performs the login All URL calls can be reviewed in the main menu Click on the corresponding URL to display the URL s details menu in which the assignment can be done Alternatively if you do n
220. ne offset in the field Extract Var on 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 58 171 Apica ProxySniffer V5 0 User s Guide Var Handler Ge Gh Ed X Extract Var from Text Line Pattern Text Line contains text Search Text ima id ctlReportingcarc Extract var an located line Loken Delimiters ExMact Token Mr 1 T Test Extract HBIEDETOBIE AILE Extracted on Line 232 Map to Var Name 2012 2013 Ingenieurb ro David Fischer AG Switzerland English Edition 5 Now try clicking on Test Extract in the Var Handler input form and check to see if the value of Extracted on Line has the same line number within the HTTP response content as the line number where the variable text fragment the result should be extracted If the line numbers are not identical your unique text pattern is not unique and you will have to find another text pattern Inspect the HTTP response content for the preceding and succeeding characters which bracket the variable text fragment In this example these characters are r and anmausenver ShewLawverrcdivSubtypeLaver15104 7 block Enter these characters into the field Token Delimiters After this click again on Test Extract and then click on the blue question mark At this point a pop up window is displayed which shows a list of text fragments tokens Enter the number of th
221. nect time request transmit time request header request content form data request header amp request content response header response header amp response content all but without response content 8 all full URL snapshot debugOptions Debug options string value dl debug loops including var handler dh debug headers amp loops dc debug content loops dC debug cookies loops dK debug keep alive loops dssl debug SSL handshake amp loops additionalOptions Additional options string sslOptions SSL HTTPS options string value all automatic SSL protocol detection TLS preferred tls SSL protocol fixed to TLS v3 SSL protocol fixed to v3 v2 SSL protocol fixed to V2 testRunAnnotation Annotation for this test run string userlnputFields Label variable name and default value of User Input Fields 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 103 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 6 Project Navigator The Project Navigator Menu or Project Navigator offers additional useful functions aside from starting and managing load test programs These additional functions are briefly described in this chapter http 127 0 0 1 7990 PRX Project Navigator Windows Internet Explorer First it is recommended that a simple directory structure be defined
222. ng the Var Finder or by extracting values using the Var Handler However it is also possible to define variables directly for special purpose use Depending on the scope directly defined variables can have special initial values which are set M S amp w X during the load test by the load test program itself Supported combinations of scope and initial values are Help Navigator Overall Load Test Session Refresh Close Var Handler Define New Stand Alone Var da Var Mame vEvalue Var Scope user var loop var Initial value a 112556 Ey mul Q current user counter D 1 2 loop counter 0 1 2 Q system time millizeconds Q load source IP host name Q load source IP address inner loop counter system time milliseconds load source IP host name load source IP address G global var U user var L loop var IL inner loop var outer loop counter overall users outer loop counter of the user nclusive multi homing support chapter 12 u 1 2 3 Initial Values constant value the variable is initialized with an arbitrary constant value null the value of the variable is not valid undefined at initialization time current user counter the variable is initialized with the sequence number of the simulated user 0 1 2 loop counter global var scope the variable is initialized with the outer loop counter 0 1 2 counted over all sim
223. nk time 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 30 0 20 00 0 22 30 0 25 00 Load Test Test 1 28 Sep 2669 14 24 34 User 1666 10 1 9 Diagram Web Transaction Rate Shows the number of successfully executed URL calls per second hits per second during the test run measured over all simulated users The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Heb Transaction Rate Sampling Interval 15 Seconds number of successful completed url calls per second hits per second 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 30 0 20 00 0 22 30 Load Test Test 1 28 Sep 2669 14 24 34 User 1600 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 121 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 10 Diagram Users Waiting for Response Shows the number of users which are waiting for response from the web server measured over all simulated users The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Users Waiting for Response Sampling Interval 15 Seconds humber of users waiting for response from the web server time 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 30 0 20 00 0 22 30 0 25 00 Load Test Test 1 28 Sep 2669 14 24 34 User 1600 10 1 11 Diagram Completed Loops Shows the number of successfully completed
224. number before the server name shows the number of times the hyperlink was ignored by Page Scanner 19 Non Processed Web Servers 4x http suppart micrasaft corm a 4x hip madssl org 8 4x http www topshareware cam B8 3x http www remsa de 8U 2x http e docs bea cam 8 2x http entwickler cam 8 2x http technetz microsoft cam 8 2x httpvwww adnavurm ch a 2x http www apica se 8 2x httpvwww cnlab ch 8 2x httpwww d fischer com 83 2x httpvwww infawarld corm 8 2x httipuwww lvllard da 8 2x httpwww planet it ch 80 2x httpwww postfinance ch 8 2x httpwww safearea com au B0 2x httpsibwww vellawrnet ch 443 1x http downlaad cam carm 8 1x http safotex cam a 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 158 171 Apica ProxySniffer V5 0 User s Guide English Edition Scan Result per Web Page displays all scanned web pages The embedded content of a web page such as images is always displayed in a Web Browser Cached View For example this can mean that a particular unique image is only shown once inside the web page in which it has been referenced for the first time All subsequent web pages will not show the same embedded content This behavior is more or less equal to what a web browser does it caches duplicate references over all web pages within a web surfing session More details about a specific URL call can be shown by clicking on the corresponding URL hyperlink
225. ogram Description Continue recommended wil be displayed as hint in Project Navigator apply heuristic methods from recorded session means that the automatically applied content test algorithms will be used including for modifications which have been done manually see section 4 2 2 Additionally the received HTTP status code 200 302 and the MIME type text html image gif of each URL call will also be verified This is the only option which ensures that the received web pages are correctly verified compare all URL calls with recorded size 5 means that only the size of the received content is compared with the recorded size The automatically applied test algorithms will not be applied during the load test however the HTTP status code and the MIME type will be verified The allowed tolerance range of the received size is implicitly set to 5 for all URL calls This option is not recommended because you may get misleading errors if a dynamically generated HTML page changes in size or you may not detect some errors which are embedded within a HTML page which is of the correct size none content test disabled means that only the HTTP status code and the MIME type will be verified during the load test The results of such tests are often invalid because errors embedded within an HTML page will not be detected 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page
226. ogue will then be displayed 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 72 171 Apica ProxySniffer V5 0 User s Guide ajaja 127 0 0 1 7990 dfischer webadmininterface PopupCreateLoadtest Weblet action createProgram amp classname TEST_O1 amp codeStructureModel 2 amp selectCont Y o aa X Generate HTTP S Load Test Program gt T PRX Generate HTTP S Load Test Program Mozilla Firefox Apica ProxySniffer A Apica Excellence in Cloud Performance Dr Back Refresh Close E Display Load Test Program TEST _01 java C4 MyTests Response Verification Summary a 1 Et A a clear improvement of the test quality is achieved URL Execution 3 5 5 Q size 5 6 1740 bytes no failure action Serial Executed 18 02Feh2011 6 5 amp size 5 6 12200 bytes no failure action Parallel Executed 82 14 14 A size 15 680 bytes no failure action Threads SITES 6 02Mar2011 19 19 Q size 5 770890 bytes no failure action l 03Feb2011 22 22 A size 15 177150 bytes no failure action 03Mar2011 04Mar2011 040ct2011 050ct2011 068ep2011 07Mar2011 088ep2011 098ep2011 I 105ug2011 11Aug2011 26 26 amp Download Proxy Sniffer Installation Kits 27 27 A size 15 56680 bytes no failure action 30 30 Q Fixed license fee per year for local load generators 34 34 The number of virtual users is counted locally on the E 36 36 a size 15 2058
227. ollow CGI Parameters this by default disabled option acts as protection against receiving almost identical URLs many times if they differ only in their CGI parameters If disabled only the first similar URL will be processed Example the first URL http www lt domain gt showDoc context 12 will be processed but subsequent similar URLs such as http www lt domain gt showDoc context 10 and hitp www lt domain gt showDoc context 13 will not be processed e Authentication allows you to scan protected web sites or web pages The following authentication methods are supported Authentication Method Note Apply HTTP Basic Authorization Base64 encoded username password send within all HTTP request headers Base You should also enter a username and a password into the corresponding input fields Apply NTLM authentication for all URL calls if requested by the Web server The NTLM configuration of the NTLM Personal Settings menu Chapter 0 will be used P Apply a HTTPS SSL client certificate for authentication The active PKCS 12 client certificate of the Personal Peete GUION ica oettings menu Chapter 3 1 2 3 will be used e Browser Language used when scanning multilingual web sites to tell the web server which default language should be preferred e Use Proxy this option allows you to scan through an outgoing proxy server by applying the next proxy configuration of the Personal Settings menu e SSL Version allows you to se
228. on Failures Diagram Error Types Diagram Number of Errors per Page gt Diagram Number of Errors per URL gt Diagram External Measured Data Project Navigator Average 90 Percentile Detail per Page seconds Response Time per Page Average Page 1 Start Page Download Page 3 Support Page 4 References Page 5 About Us e e Hint execute the same load test program several times with a different number of concurrent users and compare the measured results Ci Hint if a permanent error occurs at the same URL you should call the Var Finder menu see chapter 7 5 and verify if the handling for dynamically exchanged session variables must be applied Eee A eene AA Hint Click inside the diagram on the ll bars to display details Save image to disk Done o O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 29 171 Apica ProxySniffer V5 0 User s Guide 5 Session Cutter PRX Main Menu Windows Internet Explorer ee Es IC xl do m le http 127 0 0 1 7990 sessionCutter false 11 X Live Search Pr gt deh Gees A A eee lt lt N s 5 proxy sniffer Main Menu Toa o ih e b Admi Hel Web Personal Project Generate analysa Refresh Web Admin Professional Edition P Tools Settings vigator Load Test Load Tests Display English Edition The Session Cutter Menu allows to combine one or more web surfing sessions to form a new
229. or snapshots can be displayed on the Analyse Load Test menu after the load test result data have been 41ms 27 201 bytes 5R acquired ES amp Page 4 References 28ms 22 529 bytes EFN LB amp Page 5 About Us 21ms 13691 bytes 0 Auto Refresh FTU sive aen Refresh Page 1 Start Page 0 sec Test AV Time AV Size Passed Failed 1 308 ms 31 911 bytes 202 24 GET Http 192 16 4 5 80 2 12ms 3 592 bytes 202 GET http 1 82 16 45 80 format css 3 ms 663 bytes 202 GET http 192 16 4 5 80 4 X XX XX gif 4 7 ms 798 bytes 202 GET http 1 92 16 45 80 flaqGerman qif 5 ms 1 847 bytes 202 GET http 1 92 15 45 8 flagEnaland dgif 6 7 ms 716 bytes 202 GET http 1 92 15 4 5 80 arrove red 12x9 qif 7 ms 917 bytes 202 GET http 1 92 16 4 5 80 pdf icon 16x16 qif 8 ms 8542 bytes 202 GET http 4192 16 4 5 80 screenshots 1 p gif 8 ms 8236 bytes 202 GET http 192 16 4 5 30 screenshots 3 p gif 10 10ms 33 580 bytes 202 GET http 1 92 15 45 G0 responsetime git raat mT TIAA A lee dew O AMT I 7T AD ATI nn nm ne nn UT ou E oO 2 O O O Q OOOO 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 28 171 Apica ProxySniffer V5 0 User s Guide English Edition 2 PRX Job 1332 Acquire Load Test Result Mozilla Firefox EZ X PRX Project Navigator Mozilla Firefox ex E http 11127 0 D 1 7990 dfischer webadmininterface PopupDirectoryMNavigatorAcquireLoadTe
230. ot know on which URL the login was made you can search for a specific text in the entire recorded session In this example you should use as the search string the password which was entered during recording Click on the Search Overall icon and enter the password as the search string N 1 0 x http 127 0 0 1 7990 Proxy Sniffer Search Overall Visible Items Mozilla Firefox E B xj p gt QQ oO 8 ex Se Project Search Generate Save Refresh Close t ooo Search Overall Visible ltems Help x Navigator Overall Load Test Session Help Search ASCII Text alex Match Case M Include URL Encoded values Inside M HTTP Request Header HTTP Request Content Matinside M Referer li Var Handler Y DOS M HTTP Response Header M HTTP Response Content M Cookies Search A zu Bu N 15 POST http www d fischer com B 8tlprztooliservietVVebMainMenu Afterwards click on the red right arrow inside the search result to see Found in Request Content Line 1 Posiion38 herspassword alex the URL details of the login Total 1 Result Note a red right arrow inside the search result means that the search string has been sent by a URL call to the web server Blue left arrows inside the search result mean that the search text was found in a response to a URL call which was received from the web server javascript Popupw
231. pee NEN TM poe DES Pod o lt Person xsi types lava AE ererson gt e Taken Delimiters NT 4 ContentType textixml charset utt 8 Branchencode xsi nil true gt Pa lt ErfassungsDatum gt 20071026 lt ErfassungsPatum 4 Extract Taken Mr E r 5 SOAPACtOn lt IstInteressent gt true lt IstInteressent gt e B MPowered By Serlet2 4 JSPIZ0 om v X Powered By Serviati2 4 JSPI2 0 E lt IstInteressentYor gt 20071026 lt IstInteressentYon gt Trim whitespaces yes lt IstInteressentBis xsi nil true gt Ped lt IstGesperrt false lt IstGesperrt gt IstGesperrtVon xsi nil true gt S Test Extract lt IstGesperrtBis xsi nil true gt lt Korrespondenzanrede gt S A cern n lt Korrespondenzanrede gt Extracted Value respondenzanredelstFreite gt false lt KorrespondenzanredelIstFreitext gt nNr gt 10001 rn 1100000000000856383 4 E soapenviEnvelope xmlns soapenv httpschemas xrmli Typ i nil true gt lt Key gt 11 0000000000856 3 lt Key gt lt Person0ID gt lt Werbesperre gt false lt Werbesperre gt lt Sprache gt 0 lt Sprache gt ii AA E ee Iv assign var automatically to all HTTP requests lt LifeCycleStatusDatum gt 20071026 lt LifeCycleStatusDatum which contain the same text pattern extracted lt Kurzbezeichnuno gt Thalmann Andy lt Kurzbezeichnuna gt valu E Extract Geburi loop var A Input File 1 HTTP Request Content Pattern ap to Var Name
232. per Page 4 43 sec page NetWork Throughput per User 2 3 kBytes sec Total Transmitted 2 790 MB Test Scenario Baram Response Time per Page gt Results per URL Call Overview gt Results per URL Call Details gt Diagram Response Time Percentiles gt Didgram Top Time Consuming URLs gt Diagram Concurrent Users gt Diagram Session Time gt Diagram Web Transaction Rate Diagram Users Waiting for Response gt Diagram Completed Loops Diagram TCP Socket Connect Time Diagram Network Throughput Diagram HTTP Keep Alive Efficiency gt Diagram SSL Cache Efficiency Diagram Session Failures Diagram Error Types Diagram Number of Errors per Page gt Diagram Number of Errors per URL gt Diagram External Measured Data Results per URL Call Overview 0 Page 1 startseite user s think time 15 0 seconds E Passed Failed AVTime lt 90 AV Size URL 4 660 405 ms 768ms 13 355 bytes GET httpsef testix post ch 443 efisecure html login amp resetlogin amp p spr cd 1 E 4 660 30 ms 27 ms 56 607 bytes GET https et testix post ch 443 efipublicicc jsiiquery 1 2 6 min js m 4 660 15 ms 14ms 2 514 bytes GET https ef testix post ch 443 efipublicicc is jquery bgiframe pack js nu 4 660 0 17 ms 13ms 42289 bytes GET https eftestixpost ch 443 ef publiciccijsitabbed js g 4 660 13ms 12ms 1 041 bytes i icfeciistaria j al 4660 14ms 14ms 6 326 bytes Jfet testi icicc js tooltip2 j B
233. porne 777 byes 1520 D GET nous def let post ceci EAT AMES OCDE BOME LE POE HDE IER BH Ce NENAS 5777 bytes 1 Request 24 72 bytes ec Inner Loop 2 Executed by 75 of all Users 6 2 Break and Continue Conditions in Inner Loops n ee Abort inner loop after current iteration Enable Pacing fixed to B seconds Min loop duration per user O variable vinnerLooptounti seconds amp Apply Inner Loops Started from Page 1 Start Page EndPage Iterations Conditions If Dur Exc Pacing UG X 1 x1 vlnnerLoopCount 4 abort Desciption Joop 1 Done 2012 2013 Ingenieurb ro David Fischer AG Switzerland Input File Settings File Scope new line per user Line Order randomized EOF Action reopen file After you have defined an inner loop you can also define additional conditions which allow you to control the run time behavior inside of an inner loop If such an additional condition applies becomes true the corresponding action can be break or continue Break means jump out of the inner loop After a break the simulated user will call the next URL Call subsequent to the end of the inner loop Continue means jump back at the start of the inner loop without calling the subsequent URL Calls of the current iteration inside the inner loop However such a jump is not executed during the last iteration of an inner loop In such a case the inner loop is immediately finished similar to the brea
234. r el x Gr Dr BO DF nnzr oo 1 7990 Project Navigator Execute Load Test m Web Admin J g Help Jobs Close GEN Main Menu T 8 4 la e 3 Web Admin Professional Edition Help Tek Per Mai Loue en sad Yes Bu Test01 Test Completed Local Exec Agent Job 93 File Size Modified m ob 93 err 0 15 Jun 2006 22 21 41 u job_93 in 413 15 Jun 2006 22 21 41 job_93 out 258254 15 Jun 2006 22 23 05 http 127 0 0 1 7990 Proxy Sniffer Analyse Load Tests Select and Compare Results Mozilla Firefox job_93 status 3 15 Jun 2006 22 23 05 gee Nie X e e d Ni er LLU A EA REA eee Proxy Sniffer ee u Test01 class 334 73 15 Jun 2006 22 21 39 Web an Analyse Load Tests Select and Compare Results Help Clear Close v Test01_15Jun06_222141_20u prxres 84 562 15 Jun 2006 22 23 04 4 Project Directory MyTestst 5 Upload File Browse File Extagsion pres EA userAccounts bd Navigator Use Proflict Navigator to load result files ll Acquire Selected Files VW oad prxres File on Analyse Load Test Menu IV Close window after acquire Sad Users estDuraon Web Trans Sess Faures Net Throughput Remote Directory 9 CADOKUME 1imutongiLOKALE amnpiPrExecAgentlobsijb 93 127 0 0 1 u o LJ Woompare Local Directory CaProgrammeiProxySniffenMyTests Project mageetor Hint execute the same load test program several times with a diff
235. r than jsessionid because the name itself E El can be configured inside the web application server You must enter the actual parameter name in the Enable dynamic URL rewriting field Rewrite Parameter Var extractors for dynamic URL rewriting successfully created It is also possible that the value of the URL rewriting parameter can change during the web surfing session for example after logging in to the web application or after logging out In this case you will see corresponding var assigners for HTTP two or more extractors for the URL rewriting parameter inside the Var Handler requests are automatically applied jsessionid loop var 15 amp URL Rewrite Context 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 68 171 Apica ProxySniffer V5 0 User s Guide English Edition 8 Generating Load Test Programs Note that only URLs which are visible in the main menu are used by the load test program This means that you can use the URL filter to exclude certain types of URLs from being executed by the load test program d si y C OO Fier Filter Input Fields e No Binary Data Images suppresses all URLs which are received along with a 200 ok HTTP status code but with non ASCII content data This will strip away all images and other kinds of binary data such as flash animations e No CSS JS Only HTML suppresses all successfully received 200 ok HTTP status code ASCII text da
236. res de html 31 ms 21 731 bytes Proxy Sniffer Web Load and Stress Testin Proxy Sniffer Produkt Features http www proxy sniffer chifeatures de html 156 ms 130 823 bytes 78 ms 63 349 bytes 47 ms 38 466 bytes fimages en SessionRecorderL dif images en SessionRecorder qif fimages en varHandler jpd fimages en Levels dif GET http www proxy sniffer com 80 HTTP 1 1 Accept Accept Encoding gzip deflate User Agent Mozilla 4 0 compatible MSIE 7 0 Windows NT 5 1 NET CLR 1 1 4322 NET CLR 2 0 50727 AcceptLanguage en Host www proxy sniffer com Connection Keep Alive Proxy Connection Keep Alive n Demas fable m vitm mr Mania Tenia lain 1 HTTP 1 0 200 OK 2 Date Sat 06 Dec 2008 19 19 07 GMT 3 Server Apache 4 Content Location index html en 5 Vary negotiate accept language B TCN choice 7 Content Type text html B Content Language en sIDOCTYPE html PUBLIC M3C DTD HTML 4 01 Transitional EN HTML HEAD META HTTP EQUIV content type CONTENT texthtml charset 150 8859 1 gt META HTTP EQUIV content language CONTENT en gt lt TITLE gt Web Load and Stress Testing Tool Proxy Sniffers TITLE gt lt META NAME description CONTENT Professional web load and stress testing tool Tests and analyzes the response time and the stabilit lt META NAME DC Language content en gt lt LINK REL stylesheet TYPE texticss HREF format css gt
237. ret JavaScript code and does not submit forms Only hyperlinks are considered Cookies are automatically supported Note 2 Page Scanner keeps the entire scanned web site in its transient memory RAM in compressed from This means that large web sites can be scanned but it also means that transient memory is not unlimited If you encounter in the Proxy Sniffer console the error message java lang OutOfMemoryError you should stop Proxy Sniffer and start it again with 1024 MB memory fay ProxySniffer k t Application Reference Manual k ES Compile and Run Proxy Sniffer Load Tests RE Proxy Sniffer GUI Please note that the Page Scanner tool may return no result or may return an fan gt dfe Proxy Sniffer GUI Server without Exec Agent incomplete result because some web sites or web pages contain malformed fan gt d Proxy Sniffer Server HTML code or because old unusual HTML options have been used within the an oy pose ue scanned web pages Although this tool has been intensively tested we are not an A e ately s able to provide any warranty of error free behavior Possible web site or web m de Proxy Sniffer Server 1024 ME page related errors may be impossible to fix because of divergent requirements m TA Proxy Sniffer User s Guide or because of complexity The functionality and behavior of this tool is similar to 2 IES other search engines which have also similar restrictions 2 2012 2013 Ingenieurb ro David Fischer
238. rface PopupDirectoryNavigatorStartloadTestWeblet flePathB64 02ncRG9rdWw1IlbnRlIHvuZCBFaWwSzdGYsbHvuz2 YY this template can be used to rerun repeat the same load e x test ad o s Project Navigator Execute Load Test a etae Execute Test Form denotes from which computer or load Execute Load Test Job Testo releasing cluster the load test will be executed If you did Load Test Input Parameter C v save not define additional remote Exec Agents or Exec Agent Execute Test fram Host Local E Agent Clusters chapter 11 only the option Host Local Exec Number of Concurrent Users 0 v Agent is available indicating that the load test program is Load Test Duration 20min w executed by your local system Max Loops per User unlimited e Startup Delay per User 200 Milliseconds Number of Concurrent Users number of users which are Max Network Bandwidth per User unlimited Downlink unlimited Uplink simulated during the load test Request Timeout per URL 60 Seconds Max Error Snapshots ZOMB memory Load Test Duration planned test duration After the test Ea 5 Bl Seconds duration has elapsed each user will terminate the current Additional Sampling Rate per Page Call 10096 loop repetition of the web surfing session before the test Additional Sampling Rate per URL Call 20 Add run completes thus the duration of the test run will be a Debug Options none recommended little bit longer than the planned test duration given
239. rloop Note when using this option the resolved IP addresses and therefore the stressed Web servers may alter inside the executed loop of a simulated user at any time suddenly from one URL call to the next one o dnsfixttl seconds Enable DNS TTL by using a fixed TTL value of seconds for all DNS resolves This option cannot be used in combination with the option dnsperloop Note when using this option the resolved IP addresses and therefore the stressed Web servers may alter inside the executed loop of a simulated user at any time suddenly from one URL call to the next one o dnsperloop Perform new DNS resolves for each executed loop All resolves are stable within the same loop no consideration of DNS TTL within a loop This option cannot be used in combination with the options dnsenattl or dnsfixttl Note consider when using this option that the default or the configured DNS servers are stressed more than usual because each executed loop of each simulated user will trigger one or more DNS queries o dnsstatistic Effects that statistical data about DNS resolutions are measured and displayed in the load test result by using an own DNS stack on the load generators Note there is no need to use this option if any other more specific DNS option is enabled because all other DNS options also effect implicitly that statistical data about DNS resolutions are measured If you use this option without any other DNS option the own D
240. rnnnnrnnnnncnnrnnnninnnns 31 6 INNEL TRENT AAA ii 34 6 1 Conditional Execution of Parts of the Web Surfing Session ccccccccscccseeeceeecceeeceueeceeecsucecsueeseuecsueecsueesueessueeseuessusesueesseecsueessuessusesueeseeesaeeess 36 62 Breakand Continue Conditions Indnrier LOO DSc coe ato uii oca t RE noa hace Mean 37 1 Dynamic Session Fal AlCl Ol T Eo oo oom 39 7 Nanable Handler Var Handler ida iii 41 TE MEME ep RE S E T TU PUE 42 7 2 1 More FING OF USING HADUL FICS destro nu ende 47 AS o RU M 47 7 3 1 More PIMS ION USING ic RM oU CENE tect nei tie as EE 50 7 3 1 1 Example Adjustable Users TAMK Dean 50 y 1039 mE 52 7 5 Dynamically Excnhanged Session PararmielelS i oai eda adeps ee e A ii liinda 54 7 9 1 Automated Handling of Dynamically Exchanged Session Parameters Var Finder esses nenn nenne nenn nano nnns 55 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 2 171 Apica ProxySniffer V5 0 User s Guide English Edition 75 2 Manual Extraction of Dynamically Exchanged Session Parameters cccceccceecceeeeceeeeeeeeseeeeseeeeseeseueeseeessueesueeseeseueesseesseeeseueeseeesneeneees 57 Amb RE NaCN ke Xt F Atle MIS iussit orenrdidb POPE PI nck Sane meen Nini dic Music a DRE Aqu NOEL deans NUBE 62 7 6 1 Extracting and Assigning Values of XML and SOAP Data oen nia a a ts 63 1
241. rors o a B EH BEE ED a time 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 50 0 15 00 0 17 30 0 20 00 0 22 50 0 25 00 Load Test Test l 28 Sep 2669 14 24 54 User 16606 Top Error Types All failed URL Calls E User Specific Test Failed 57 Errors M Content Test Failure String Not Found 16 Errors O ok 266 Wrong HTTP Status Received 6 Errors Load Test Test l 25 Sep 26659 14 24 34 User 1664 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 126 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 18 Diagram Number of Errors per Page Shows a compilation of the web pages which experienced the most errors Number of Errors per Page 4 Errors Page 13 Bestellung abschicken 4 Errors Page 11 An welche Adresse weiter 2 Errors Page 3 Bad 2 Errors Page 1 Start Page Load Test EinkaufUnregif 2 29 Jul 2665 62 55 65 User 36 10 1 19 Diagram Number of Errors per URL Shows a compilation of the URLs which experienced the most errors Number of Errors per URL Errors Page 13 Bestellung abschicken Errors Page 11 An welche Adresse weiter URL 7 Error Page 3 Bad URL 8 Error Page 3 Bad 4 3 2 Errors Page 1 Start Page 1 1 URL C26 1 Error Page 411 An welche Adresse weiter Load Test EinkaufUnregif 2 29 Jul 2865 62 55 65 User 36 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Right
242. roxy Sniffer Additional Services 9 Pro ys nif fer Ab gt E gl Ra PE PT O AAA A u Fertig 4 Lokales Intranet 100 4 Individual web pages can be selected by clicking on the name or the number of the Web Page The selected web page s can be moved or copied by using the move here or copy here button After the splicing of the new web surfing sessions is complete the Session Cutter Menu can be closed by clicking on the Close button or by clicking again on the Session Cutter icon O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 30 171 Apica ProxySniffer V5 0 User s Guide 5 1 Importing Web Surfing Sessions from External Definition Files English Edition The Session Cutter allows additionally to import web surfing sessions from external definition files PRX Main Menu Mozilla Firefox File Edit wiew History Bookmarks Tools Help em Y a A E http 127 0 0 1 7990 sessionCutter true PRX Main Menu eee goo Proxy Sniffer Main Menu gt Web Admin Professional Edition Help T E o Web Personal Project Load Test Jobs P Generate Ana Tools Scanner Settings Naviga 0 lyse Load Test Load Tests Display PRX Import Session Mozilla Firefox B http 127 0 0 1 7990 dFischer webadmininterFace ImportSessionWeblet sessionIndex 0 Refresh e i zm Ps ES nee J Scratch Area 1 Session load import save gt delete selected
243. roxy Sniffer URL Details Var Handler Mozilla Firefox gee Proxy Sniffer Web Admin URL Details Var Handler we Help Project Search Generale Navigator Overall Load T T zinl xl a we Pide Refresh Close Y zel re TEN ie dl HTTP Request Header gt d fischer com 8080 Y GET prxtool serdetWebMainMenu HTTP 1 1 r 2 Host www d fischer com 8080 serAgent Mozilla 5 0 Windows U Windows NT ficcinn Var HTTP Response Header HTTP 1 0 200 OK Form o 2 Content Type TEXT HTML 3 Expires 0 ipntool servletWvebMainMenu PASSWORD password username Qb LoginFlag 1 Cache Control no cache must revalidate TEXT 5 Pragma no cache HIDDEN m m H 1 lt HTML gt 2 lt HEAD gt lt META HTTP EQUIVZ CONTENT TYPE CONTENT textihtml charsetziso 88 lt TITLE gt Proxy Sniffer Project Master Logins TITLE STYLE TYPE texticss gt 1 zl hady n a illi tahla tr tA form innit calact antion f fant ciza Ant fant famihr Arial Halatica 1 a HTTP Response Content Unique Hyperlinks Extract xj All Rights Reserved Fvtract Mar no vars defined none Add Field none Page 41 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 2 Input Files Input Files can be used to extract variables from a text file such as a username and a password per
244. s 5093 ms 894 bytes GET http 32 15 4 33 808 0 prtaol setup gif 2950ms 3484ms T8B bytes feet http i192 15 4 33 8080 pmtool Back uif m z me Page 3 Login Man Ment AV Time 16740 ms 4 078 ms 7907 ms TT ms 5174 ms TEM I 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved English Edition Columns Test consecutively numbered Clicking on a number displays the URL detail menu Passed total number of successful calls Failed total number of failed calls If this value is greater than Zero you can click on it to display the corresponding error snapshots Chapter 10 2 AV Time average response time per URL call or per web page lt 90 slowest response time within the fastest 90 of all measured values 9096 percentile value This result is only available if the response time has been collected at least 5 times depending on the percentile sampling rate which was selected when the test run was started Clicking on this value displays the corresponding response time percentile diagram AV Size average size of transmitted and received data per URL call or per web page URL the URL called Page 112 171 Apica ProxySniffer V5 0 User s Guide 10 1 3 1 Response Content Throughput In Depth Measurement of HTTP S Response Streams The in depth measurement of HT TP S response streams is only available if you have to enable the additional opt
245. s The number of data points depends on the Statistic Sampling Interval which was set when the test run was started Overall Network Throughput Sampling Interval 15 Seconds overall network throughput in megabits per second 0 00 00 0 02 30 0 05 00 0 07 30 0 10 00 0 12 30 0 15 00 0 17 50 0 20 00 0 22 50 Load Test Test 1 28 Sep 2669 14 24 34 User 1600 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 123 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 1 14 Diagram HTTP Keep Alive Efficiency Shows the efficiency of the HTTP keep alive protocol option percentage of reused network connections measured over all users and URL calls HTTP Keep Alive Efficiency Re Used Network Connections Bl Heu Created Network Connections 19 45 491 Bl Successful Re Used Network Connections 80 55 28234 Load Test GenerischerTest 59 Jun 2666 17 25 45 User 5H 10 1 15 Diagram SSL Cache Efficiency Shows the efficiency of the client side SSL session cache which depends on the web server SSL configuration In other words this shows the percentage of abbreviated SSL handshakes measured over all users This diagram is only available when each user has executed at least 5 successful loops and when the encrypted HT TPS protocol has been used SSL Cache Efficiency Re Used SSL Sessions M New Created SSL Sessions 8 672 1 M Successful Re Used SSL Sessions 99 33 149 Load Test HT
246. s Reserved Page 127 171 Apica ProxySniffer V5 0 User s Guide English Edition 10 2 Error Snapshots If errors occurred during a load test a frozen snapshot of the entire error situation is taken for each error as long as the number of maximum allowed error snapshots not exceeded The maximum number of allowed error snapshots is set when the test run is started test input parameter Max Error Snapshots An error snapshot contains the following data e The date and time the error occurred e The defective URL including a reference to the web page e The error type and the HTTP status code e The internal execution step of the failed URL call at the point in time when the error has occurred for example open network connection or receive content e All data about the failed URL call HTTP request header HTTP request content only if transmitted HTTP response header only if received HTTP response content only if received e The Error Log The session log of the simulated user This includes also actual information about the values of variables which have been defined by using the Var Handler e A Thread Statistic at Error Time a system snapshot of the activity of all other concurrent users The upper part of the window contains a list of all error snapshots The content of this list depends on the context from which the menu was invoked error snapshots of the entire test run per web page or per URL The list c
247. s inside the Var Handler are First Extract li First Assign V Recorded Value automatically created a 1 CGlParam type 163283 19ColParam N Bart first C 2 gt 19 91 Param levid levid 94153 On the other hand if you receive an error message you must manually extract the value of the dynamically exchanged session parameter see the next subchapter O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 55 171 Apica ProxySniffer V5 0 User s Guide English Edition Automated dynamical handling of parameter id not possible manual handling required help First Extract First Assign Var Name Parameter Name A 1 CGlParam type A 19 CGlParam E act 2 18 CGlParam levid E levid 4 amp 30 CGlParam No agenda A 33 CGI Param y id A 41 CGlParam y id In this example the parameter VIEWSTATE could be handled automatically but the parameter id must be extracted manually Since this parameter is listed twice the same name with different values the extraction must also be done twice once for each distinct value Hint you can use this menu as a checklist of parameters which are already dynamically handled irrespective of whether the extraction done automatically or manually The handling is already done if the line contains a blue extract arrow and a red assign arrow 2012 2013 Ingenieurb ro David Fischer AG Switzerland Hecorded Value
248. sdelay 200 maxloops O sampling 15 percpage 100 percurl 20 maxerrmem 20 nolog annotation First Test Run Test started at 28 Oct 2009 23 19 10 ECT Planngd e st duration 4 00 min 4 14 min elapsed Test started at 28 Oct 2009 23 19 10 ECT Planned test duration 4 00 min 4 29 min elapsed Lid ul LS Total Passed URL Calls 1 Total Passed Loops 182 Users Waiting For Response 1 of 10 Total Passed URL Calls 6 962 Total Passed Loops 192 Users Waiting For Response D of D Total Failed URL Calls Total Failed Loops 34 TCP Socket Connect Time ms 2 Total Failed URL Calls 34 Total Failed Loops 34 TCP Socket Connect Time ms 2 HTTP Keep Alive Efficiency 5 User s Think Time per Loop sec 12 0 AV Network Throughput Mbit s 2 10 bp Alive Efficiency 97 1 User s Think Time per Loop sec 12 0 AW Network Throughput Mbit s 1 95 Av Web Trans Rate URL calls sec 1 Session Time per Loop sec 18 4 Total Transmitted Bytes 63154804 Rate URL calls sec y 26 3 Session Time per Loop sec 18 4 Total Transmitted Bytes 63 426 888 Abort Job 1332 rores 0 co Disable Detailed Statistic TT TT MTM User Page AV Time AV Page Size Passed Failed No Real Time Data Job Completed 446 ms 190018 bytes 24 x Note the real time statistic of this window is directly extracted from the running load test job and no longer available when the job has been completed sims 64382 bytes l The collected job statistics and the err
249. se time per page in seconds e During the Execution of a Job and at the End of a Job if the Average Response Time of the Slowest Page is greater than seconds o slowestPageName The name of the measured slowest page o avResponseTimeOfSlowestPage The measured response time of the slowest page in seconds o avResponseTimeOfSlowestPageLimit The configured threshold for the response time of the slowest page in seconds 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 150 171 Apica ProxySniffer V5 0 User s Guide 13 Page Scanner English Edition Page Scanner browses and explores web pages of a web server automatically in a recursive way similar to a Web Spider or a Web Crawler e Primary Purpose the scan result can be turned into a normal web surfing session and from this a load test program can be generated This provides a simplified way to create a web surfing session instead of recording single web pages manually However Page Scanner can only be used to acquire web surfing sessions which do not require HTML form based authentication This tool is not a replacement for recording web surfing sessions of real web applications e Other Purposes Page Scanner allows the detection of broken links inside a web site and provides statistical data about the largest and slowest web pages It also supports searching for text fragments over all scanned web pages Note 1 Page Scanner does not interp
250. sec Web Transaction Rate Hits per Second 5 callsisec 16 callsisec 14 calls sec Average Outstanding Requests 0 4 Total HTTP s calls 754 2 320 HTTP Keep Alive Efficiency 0 00 96 0 00 96 SSL Session Cache Efficiency Average Network Connect Time 1 millisec 1 millisec 4 millisec Average Network Throughput 1 MBit sec 2 MBit sec 2 MBit sec Total Transmitted Bytes 14 MB 42 MB 46 MB ice See Average Session Time ITA Heb Transaction Rate seconds url calls per second E A 5 TIS o E E 0 00 00 0 00 37 0 01 14 0 01 51 0 00 00 0 00 37 0 01 14 0 01 51 Test 1 27 Nov 2008 22 58 55 User 2 Test 1 27 Nov 2668 22 58 55 User 2 LAIM 8 Average Session Time ILS 8 Web Transaction Rate seconds url calls per second i 0 time 0 00 00 0 00 37 0 01 14 0 01 51 0 00 00 0 00 37 0 01 14 0 01 51 0 02 25 Test 1 27 Nov 2008 23 63 55 User 18 Test 1 27 Nov 2008 23 63 55 User 18 ITA ESA Average Session Time ITA ESA Heb Transaction Rate seconds url calls per second E at Done 10 4 3 Session Failures ice See Network Connect Time milliseconds ti 0 00 00 0 00 57 0 01 14 0 01 51 0 02 25 Test 1 27 Nov 2668 22 58 55 User 2 UFAM 9 88 Network Connect Time milliseconds time 0 00 00 0 00 37 0 01 14 0 01 51 0 02 25 Test 1 27 Nov 2008 23 63 55 User 10 IA ESA Network Connect Time milliseconds This menu option displays the same data as described in chapter 10 3 4 O 2012 2013 Ingenieurb ro David Fischer AG Switzerla
251. sed humber of successful SS Lipa url calls per second hits per second ner number of successful SORT ed loops per ute sessions per minute Lurrent Ime e Users Waiting For Response The number of users which are waiting for response from the web server TCP Socket Connect Time 15 Sec Sampling Interval Remote Exec Agent Job 188 Network Throughput 15 Sec Sampling Interval Remote Exec Agent Job 188 Test 1 started at 13 Oct 2009 20 18 48 11 46 min elapsed Te stel started at 13 Oct 2009 20 18 48 11 48 min elapsed time to establish a tcp network connection to the web server in milliseconds Cone Ti overall network throughput in megabits per second u ent ne e Session Failures The Amy number of failed sessions TRITT Taerar vec TEE TI ET ESOC STS EA TVS wT E which is the same as the number of fatal errors e Session Time per User per Loop The session time for one loop per simulated user This value is the sum of the response time of all URLs and all user s think times per successful completed loop e Web Transaction Rate The number of successful completed URL calls per second measured overall simulated users e Completed Loops per Minute The number of successful completed loops sessions per minute measured overall simulated users e CP Socket Connect Time The time in milliseconds per URL call to open a new network connection to the web server e Network Throughput The total n
252. session similar to splicing motion picture film together to create a complete movie This process can only be performed using raw web surfing sessions that is recorded sessions which have not yet been enhanced using the Var Finder described in chapter 7 5 1 or using the Var Handler described in chapter 7 1 If a enhanced web surfing session is loaded into the Session Cutter Menu a warning message will be displayed If the warning is ignored all enhancements will be deleted that is after using the Session Cutter the Var Finder and or Var Handler enhancements will have to be done over again PRX Main Menu Windows Internet Explorer sy Y http 127 0 0 1 7990 vix ve Search VE d ge PRX Main Menu E deh y Gp Extras y eee N 3 e proxy sniffer Main Menu T amp OQ ih b Adral u Hel Web Personal Project Generate Analyse Refresh Web Admin Professional Edition P Tools Settings Navigator Load Test Load Tests Display Session Scratch Area 1 Scratch Area 2 gt delete selected gt a B a b gt b 1 Proxy Sniffer Web Load and S Fess Testing Tool no pages HP Sniffer Download gt 2 Pri ry S iffer icensin g o jtions and Prii Ces gt 3 Proxy Sniffer Load Test Tool Special Product F gt ie Proxy Sni jan gt roxy Sniffer Customers angfPariners b b 7 Proxy Sniffer Supporti FAQS gt 8 P
253. session into the transient memory area of the main menu or into a scratch area of the Session Cutter 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 162 171 Apica ProxySniffer V5 0 User s Guide fa English Edition a http 127 0 0 1 7990 action loadProxyDataDump amp cselectDir QzpcUHJv23JhbSBGaWxlc1xQrm94eVNuaWZm2 Windows Internet Explorer pM x a E eee gt SS qe Proxy Sniffer Project Navigator lt gt e dd 2l Ih 36 Web Admin Help Setup Network Jobs Analyse Refresh Close After the web surfing session has been stored it will be automatically loaded into the Main Menu if the Load session into checkbox was selected After this you can generate the load test Program see chapter 8 a a DOCE TAD ho EE C Program Files ProxySniffer MyTests fischer MyTests LL ul L2 iiil alil roThinkTime Daze LEJ Cluster EM e p a PRX Main Menu Mozilla Firefox File Edit Yiew History Bookmarks Tools Help GB Q 2 amp F netp f127 0 0 1 7990 1x z coo PI DM Proxy Sniffer Main Menu Y i I e ll 1 A m Pa Personal Proj Generate Analyse Refresh Web Admin Professional Edition Help Todes Scamer Settings Navigator Load Test load Te Le ests Display ME ERBEN EST sert m m m m im Gi Im m Gi im im Im m m MA Ue OT TI TO DO DO ODO TOTO Offset Position Content Size Ti
254. set Recording Recording Recording Recording State STOPPED Segreh Session LAVA Cj MyTests Save Recorded Session s Filename Test01 ScriptExamples Comment first test session without login t iod Project autho O ie l Save onds 35 EM e et required Enter a simple filename with no path and no file extension a The file extension is always prxdat Example SurfSession_01 E 5 agg rE E E 4 recommended will be displayed as hint in Project Navigator http 127 0 0 1 7990 Proxy Sniffer Project Navigator Mozilla Firefox n ni xi eee gt web Admin l g Help Setup Network Jobs Analyse Refresh Close Saved e Sessions C PProgramme ProxySnifferMyTests Hoag or HE scriptexamples Mm Trash BUE 4 1 EE ta ed sessions can be restored by clicking on the Load Session icon Ed inside the Project Navigator After saving the session the Project Navigator menu appears You can later restore the web surfing session by clicking on the Bone y ES icon of a saved session inside the Project Navigator 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 18 171 Apica ProxySniffer V5 0 User s Guide English Edition 4 2 Reviewing the Recorded Web Surfing Session After you have recorded a web surfing session you should review the results by checking the following 1 Does the recorded sessio
255. simulated user which can be assigned to a login form However the functionality of input files is generic which means that variables for any purposes can be extracted Click on the Add File button inside the Var Handler to define of a new Input File and enter a simple file name without a directory path Please note that this action creates only the definition of the input file but that it does not create the input file itself on disk This means that the input file must also exist on disk and that it must be placed inside the same Project Navigator directory where the load test program is stored You can create the input file on disk before or during or after the definition is made or you can also copy an existing file to the corresponding Project Navigator directory auia EU Create a new Input File on disk inside the current Li Project Navigator directory gt aA 9 amp eX Project Search Generate Save Navigator Owerall Load Test Session Petresh Close Help i o a 6 amp eX Praje Search Generate ave Havigator Overall Load Test Session Petresh Close Name of the Input File Definition A var Handler Ga db gt E Var Handler Eo dO 32 S Eg oelect an already on disk existing Input File which is E Add Input File 3 i l S S located in the current Project Navigator directory J File Hame user Accounts txt do me E un iA new file name none F
256. sion requiring such a certificate can be recorded Note normal HTTPS sessions do not require client certificates The PKCS 12 file must first be loaded by using the Personal Settings menu Also ensure that the certificate is active by clicking inside the red bar on the certificate The red bar will change to a green check mark when the certificate is properly active LIPS Client Certimfirate BUT id thentication Note to execute a load test which uses client certificates you must also enable the option HTTPS Client ile Browse Certificates in the Generate HTTP S Load Test Program menu see chapter 8 The allocation of individual client Nr Neme mm certificates per simulated user is supported when generating load test programs 1 amp X v 000000001 912 3 1 2 4 HTTPS Client Certificate Authentication PKCS 11 Device Proxy Recorder Allows to you to use in Proxy Recorder X509 SSL TLS client certificates which are embedded in PKCS 11 Security Devices support for HSMs and smart cards Note normal HTTPS sessions do not require client certificates 4 e i ES E Ny A 2 pe TF E Because the proxy recorder operates as a man in the middle between the web browser and the web server the client certificate must be loaded and activated before a web surfing session requiring such a certificate can be recorded Please read the separate documentation Using PKCS 11 Security Devices for
257. sponseContentType content type Allows the setting of the expected response content type If this option is not used and if the command defaultResponseContentType has not been previously used the response content type will not be verified responseContentCheck text fragment gt Checks to see if the response content contains a specified text fragment The response content will not be verified if this option is not set responseContentSize content size deviation Checks the size of the response content The size of the response content will not be verified if this option is not set Argument 1 contains the size in bytes and argument 2 contains the maximum allowed deviation of the size in percent 0 100 Hint the URL option requestContentFile can for example be used to POST XML data Example defaultURL http www d fischer com defaultRequestContentDirectory D XmlData POST putDataDo action addAddress 200 requestContentFile requestData xml requestContentType text xml 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 33 171 Apica ProxySniffer V5 0 User s Guide English Edition 6 Inner Loops It is possible to define inner loops which include only some web pages of a recorded web surfing session As an example inner loops can be used during a load test after the point where the users did login to repeat the web pages between login and logout several x sus see
258. ssions is supported by using any web browser such as Internet Explorer or Safari You can use also Firefox without installing the Firefox Recording Extension You must reconfigure your web browser before you will be able to record a web surfing session B da gie LE les sien fog emis bel vin as described in the Installation and Configuration Guide chapter 2 proxy host 127 0 0 1 proxy port i 127 0 0 1 7999 for HTTP and proxy port 7997 for HTTPS do not use Proxy for 127 0 0 1 Use semicolons 1 to separate entries 2 1 Recording the First Web Page Start a second web browser window Clear the web browser cache and all cookies Click on the Start Recording icon in the Web Admin GUI in the first web browser window Po E x Enter the desired start page of the web server or web application in the second web browser window The first web page should now be recorded Click on the Refresh Display icon in the right upper corner inside the Web Admin GUI to see if the recording of the web page was successful If no data was recorded you should check the proxy configuration of the web browser Please note that you must first clear the web browser cache and all cookies every time you start recording a new web surfing session Chapter 3 in the Installation and Configuration Guide contains some illustrations about how to clear the web browser cache and all cookies 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved P
259. st 1 28 Sep 2009 14 02 27 1200 18 57 min 243 3 callsiser 0 0 96 0 0 17 11 MBit sec E a ae a CE ES noa O E ie JA Hint execuN the same load test program several times with 3 differnt numpar oF concurrent Lua re results Click on the A icons to display details Load Curves Diagrams And comparisons between several test runs DDD URL Error Rate percentage of failed URL calls OOOO E A E Net Throughput average network throughput during the test run Annotation comment which was entered when starting the test run Done 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 108 171 Apica ProxySniffer V5 0 User s Guide 10 1 Detail Results Many different detail results can be displayed about a single test run 3 PRX Result Detail Mozilla Firefox EISE P http 127 0 0 1 7990 dfischerfwebadmininterface PopupAnalyseloadtestDetailsweblet key c47ce4cff198bb8eade2d309aa5beeD1 M i Load Test Result Detail Statistics and Diagrams PDF Report Export Close Advanced Test Parameter Overall Test Result Test Result Startup Delay per User 300 ms Web Transaction Rate 279 2 URL calls sec prev Request Timeout per URL 60 sec Session Failure Rate 1 70 96 Statistic Sampling Interval 15 sec Total Network Throughput 19 66 MBit sec Measured Results per Single User per Loop AV Session Time per Loop 271 99 sec loop AV Mesponse Time
260. stWeblet filePathB64 02pcRG9rdW1IbnRIIHYuZCEF Y PO http 11127 0 0 1 7990jdfischer webadmininterface PopupDirectoryNavigator Weblet selectDir QepcRG9rdW IbnRITHYUZCBFaWSzdGVsbHVUZ2VuXG1 1dG9u2 1 xQcm94eVNuaWZmZXICTXIUZXNOCG 7 eee eee rox iffer H S 3 rox niffer lt gt 4 E i ah Project Navigator Acquire Load Test Result So A x ee Project Navigator gt s ED a RJ N id x Test01 Test Completed Local Exec Agent Job 1332 ie size Modified C1 iok 1332 00 0 280ct2009 23 19 10 C1 iob 1332 in 621 28 Oct 2009 23 19 10 O iob_1332 0ut 35 811 28 Oct 2009 23 23 40 Reportlemplates CJ ioe 1332 status 3 28 Oct 2009 23 23 40 sib eas ata D Test01 class 40 600 28 Oct 2009 23 19 06 E Amen eccoccceseseee ooo 39 Y 33861 28 Oct 2009 23 23 30 o C Dokumente und Einstellungen mutong ProxySniffer MyTests i Pel eu be Plugins ecc eee gt gt Acquire Selagted Files 9 5 pnires File on Analyse Load Test Menu Close window after acquire M Remote Directory CADOKLIME 1 wnutongiLOKALE 1 iTemp PrxExecAgentJobsyjob_1332 Local Exec Agent Project HavigatomDirectory C Dokumente und Einstellungen mutongProxySnifferWyTests PRX Result Detail Mozilla Firefox SEE 6 DEN p A o 6 PO http 127 0 0 1 7990 dfischerfwebadmininterface PopupAnalyseLoadtestDetailsWeblet key 9635c7a963d6c1e07d3190b119f779518 act
261. starting item number to Item ending item number Note included Page Breaks will also be deleted English Edition 2 Delete all URLS with radio button o Content Type AND HTTP Status Code deletes all URLs which are selected by a particular content type MIME type and are additionally selected by a particular HTTP status code Only URLs which match to both selection criteria are deleted any content type means that any URLs which have a valid content type are selected irrespective of the content type value but not URLs which have no content type none content type means that only URLs with have no content type are selected absence of content type any HTTP status code means that the HTTP status code ofthe URLs is not considered all HTTP status codes selected o Host AND HTTP Port deletes all URLs which are selected by a particular hd t included Page Breaks will also be deleted The following chapters contain a step by step guide to using the Proxy Sniffer product Brief Instructions if you are in a hurry The easiest way to use Proxy Sniffer is to use a Firefox web browser and to download and install the Firefox Recording Extension from https www proxy sniffer com download PrxRecExti xpi enter this URL into Firefox After that you should start the Proxy Sniffer Console Then click on the Lil icon inside the Firefox Recording Extension and follow the instructions File Control Help O A ro est Response Hon
262. startup settings file prxsniff dat to change these values permanently For other Unix like systems set the program arguments tz and dgs to the corresponding values see Application Reference Manual 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 17 171 Apica ProxySniffer V5 0 User s Guide English Edition 4 Next Steps after Recording a Web Surfing Session 4 1 Saving the Recorded Web Surfing Session Proxy Sniffer keeps the entire recorded web surfing session in its transient memory cache For this reason you should save the recorded web surfing session to disk by using the Save Session icon inside the Web Admin GUI All data from the web surfing session are saved including all HTTP request and response headers all recorded HTTP content data and all page break definitions Any special session enhancements made by using the Variable Handler chapter 7 1 or by using the content test configuration menu chapter 4 2 2 are also saved We recommend that you also enter a small comment describing the recorded session mek http 127 0 0 1 7990 Proxy Sniffer Save Session Mozilla Firefox m xi e o 6 eee SS lt See Proxy Sniffer 1 i i 26 Y i GA T a A d Bb init gt Ben Project Navigator Save Session Help Refresh Close Help Tools Scanner Settings a edd Jobs Load Test Load Tests Display o Recorded Items 27 4 JU Start Stop Re
263. sts Jobs Job Templates Every time when a load test is started an additional job definition template file is stored in the actual Project Navigator directory in XML format Such a job definition template file contain all configuration date which are needed to rerun the same load test job again If you click on the corresponding icon of a job definition template file in Project Navigator the load test job inclusive all of its input parameter is automatically transferred to the Exec Agent or to the Exec Agent Cluster and immediately ready to run otart load test as usual with configuring the job input parameters Rerun the same load test 3 ENA ES http 127 0 0 1 7990 7filePathBB54 QzpcUHTv23JhbsBGaM lci kQcm4eYNuaWzmezX Tc TXIUZZN CTEG Wert d Project Navigator Execute Load Test Execute Load Test Job Testo Tem Y Execute Testfrom wer HT yu do IE Humber of Concurrent Users alil E nad Tact Miir atin 1 min e LA 1 Edit the template Rename or copy the template 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 101 171 Apica ProxySniffer V5 0 User s Guide English Edition Additionally if you wish to trigger several load test jobs at the same time to be ready to run by using only one mouse click you can zip several templates to one zip archive After this click on the corresponding icon of the zip archive FirstTest te set XML fi
264. t least 5 individual measurements have been collected during the test run Page 2 Login Workplace J URL E Page Response Time Percentile 188 Sampling Rate 121 Page 2 Login Workplace milliseconds Back to URL Overview Display Sampling Details A Time min sec is y i 2 E z Output wa HTML Table Load Test lastfau28 22 Sep 2006 16 14 41 User 58 Apply Page Response Time 188 Sampling Rate C12 Page 2 Login Workplace time 0 00 00 0 03 20 0 06 40 0 10 00 0 13 20 0 16 40 0 20 00 0 23 20 0 26 40 0 30 00 0 33 20 Load Test lastfax26 22 Sep 2006 16 14 41 User 58 Sample Measured on i offset min sec millis Response Time millis 22 Sep 2006 16 14 47 245 0 05 610 658 22 Sep 2006 16 14 47 967 0 05 332 799 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved By using the option lists you can select the web page and within the page the URL The option for a URL means that the percentile diagram for the web page is displayed instead of a specific URL of the web page Cumulative statistical distribution means that only the slowest URL call within a percentage of all fastest URL calls is flowing inside the curve For example 95 means that 95 of all URL calls have a response time faster than or equal to the shown value The collected individual measurements can be displayed by clicking on the Apply button It is also possib
265. t locations and can also use different protocols to communicate with the Web Admin GUI or rather with the local cluster job controller Local Web Application Exec Agent Server Exec Agent Cluster Load Test plain TCP IP plain TCP IP S e Load Test j UH te Load Test Proxy Sniffer GUI 8 HTTP HTTPS Cluster Job Controller Tunneling HTTP HTTPS Tunneling Remote Exec Agent LAN LAN DMZ Internet Exec Agent Cluster External Hosting Center 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 141 171 Apica ProxySniffer V5 0 User s Guide English Edition 11 1 Configuring Additional Load Releasing Systems Exec Agents Additional load releasing systems can be added by using the Network menu which can be invoked from the Project Navigator http 127 0 0 1 7990 Proxy Sniffer Project Navigator Mozilla Firefox E lO xl e l Se ia Pe Project Navigator E an a da I Close C Programme ProxySniffer MyTestsiTrash E A MyTests E Projekt A PA ScrintExamnles http 127 0 0 1 7990 Proxy Sniffer Project Navigator Exec Agent Network Configuration Mozilla Firefox 4 ol xj n th eu pper left p art of th e W In dow a ist of curre ntly P ira e aig c Project Navigator Exec Agent Network Configuration ei x defined Exec Agents is shown The Exec Agent
266. t must not occur inside the received content lt int gt lt search text gt The search text must occur a minimum of lt int1 gt times but not more than lt int2 gt times inside the received content 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 21 171 Apica ProxySniffer V5 0 User s Guide English Edition Search Text Examples The search text hello must occur at least once inside the received content IORA 01652 The search text ORA 01652 must not occur inside the received content 1 Dear Mr The search text Dear Mr must occur exactly one time inside the received content 1 2 Order Number The search text Order Number must occur a minimum of one time and a maximum of two times inside the received content 3 new order The search text new order must occur a minimum of three times inside the received content Note one or more variable text patterns in the form of lt variable name gt are supported as a part of the search text for example Welcome sex name More information about variables can be found in chapter 7 1 Failure Action The Failure Action determines what happens in case the URL call fails e Abort Loop gt gt Next Loop means that the current loop repetition of web surfing session of the simulated user is aborted and that the simulated user executes subsequent to that the next loop If more loops per user are planned or if the duration of the loa
267. ta which are not in HTML format This will strip away style sheets CSS and JavaScript files e No Cached Data 304 suppresses all browser side cached URLs received with a 304 found HTTP status code from the web server recommended option e No Errors suppresses all URLs with an incomplete response from the web server and also suppresses all error responses from the web server HTTP status codes equal to or greater than 400 If you do not activate this option the load test will check that error is still there that is an error SUCCESS e Host suppresses all URLs which are not received from a given hostname You may use this option to strip away foreign content such as advertisements from a banner server Additionally the usage of an exclamation mark in front of the hostname is also supported which means that items from this host are suppressed Several host names can be entered separated by commas with or without an exclamation mark Click on the Generate Load Test icon in the main menu or in the URL Details Var Handler menu to generate the load test program http 127 0 0 1 7990 dfischerwebadmininterface htdocs dataRecordDetails html displayIndex 21 Li N o N o y en v1 2 kb iz o q amp eX en A age pariona OR Lonc Tart iene ed ie neum Proxy Sniffer URL Details Var Handler Project Sesh Genera Save ools canner Settings avigator obs oad Test ad Tests Display gt Web Admin Help Navigator Overall Load Test
268. te Load Test ac eee a x Execute Testtom Cluster Cluster HostName 192 16 4 5 Cluster Cluster 1 Host Local Exec Agent Load Test Duration Host TestPC Il Host Sun Fire W240 unlimited Y Number of Concurrent Users Max Loops per User Startup Delay per User 200 v Milliseconds Max Network Bandwidth per User unlimited y Downlink unlimited y Uplink Request Timeout per URL 60 y Seconds Max Error Snapshots per URL 30 Statistic Sampling Interval 15 Seconds Percentile Sampling Rate 100 y per Page NH per URL Debug Options none recommended y Additional Options Poo sse vawart s y Annotation gt gt Continue recommended will be displayed as hint in Project Navigator Done A 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 144 171 Apica ProxySniffer V5 0 User s Guide English Edition 12 Using Multiple Client IP Addresses per Load Releasing System Optionally you may want an Exec Agent to use multiple client IP addresses during the load test in order to simulate users from different network locations In the case where a load balancer is placed in front of a web server cluster or web server farm the load balancer will often route all HT TP S requests of one client IP address to only one member of the web server cluster This is because web applications use session cookies whose context information is only stored in
269. tec chitheckuplindex html http wvww avantec chi kundenirsa Iogo dif http ew avantec chikundenirsa git http www avantec chiTabor index html http fer avantec chilabar http www avantec chiworkshop http sfr avantec chAvorkshoplindex html 2012 2013 Ingenieurb ro David Fischer AG Switzerland Largest Web Pages WW ProHy snitfer ua PPORY Soi ffer Wh prowy snif fer WWW ProHy snitfer Wil PPORY Sniffer Wh Prowwo snif fer WW ProHy snitfer ua ProHy snitfer u ProHy snitfer Wh prowy sniffer WWW ProHy snitfer fil ProHy snitfer Wh Prowyo snif fer WW proxy sniffer WW ProHy snitfer 5 3 2 17 14 41 page nr http u proxy sniffer ch Bi May 2661 21 58 83 ch fesatures_de html chrfeatures html ch support_de html cheen EnFree ScreenShothlebAdmin html che cheen javadoc36 indes all html cheen javadocsbe df ischer ut i ls HttpTe chens javadocsbe dFf ischer zutils Perfor cheseryjces_de htrl ch customersAndPartners_de html chraboutus_de html che index_en html chef reeEdition html ch support html ch en javadac3B df ischer zutils Perfor All Rights Reserved English Edition Page 160 171 Apica ProxySniffer V5 0 User s Guide e Slowest Web Pages displays a list of the slowest web pages Seconds 40 2012 2013 Ingenieurb ro David Fischer AG Switzerland Slowest Heb Pages proxy sniffer
270. tes The scan will be stopped if this time is exceeded e Max Web Pages limits the maximum number of scanned web pages The scan will be stopped if the maximum number of web pages is exceeded e Max Received Bytes limits the maximum size of the received data in megabytes measured over the entire scan The scan will be stopped if the maximum size of the received data is exceeded e Max URL Calls limits the maximum number of executed URL calls measured over the entire scan The scan will be stopped if the maximum number of executed URL calls is exceeded e URL Timeout defines the response timeout in seconds per single URL call If this timeout expires the URL call will be reported as failed no response from web server e Max Path Depth limits the maximum URL path depth of scanned web pages Example http www lt domain gt docs content about html has a path depth of 3 e Follow Redirections limits the total number of followed HTTP redirects during the scan e Follow Path Repetitions limits the number of path repetitions which can occur within a single URL path This parameter acts as protection against endless loops in scanning and should usually be set to 1 default or to 2 Example http www lt domain gt docs images images images x gif has a path repetition value of 3 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 153 171 Apica ProxySniffer V5 0 User s Guide English Edition e F
271. that each user uses during the load test its own client IP address If fewer IP addresses are available than concurrent users are running the IP addresses are averaged across the users 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 77 171 Apica ProxySniffer V5 0 User s Guide English Edition o dnshosts lt file name gt Effects that the load test job uses an own DNS hosts file to resolve host names rather than using the hosts file of the underlying operating system Note that you have to ZIP the hosts file together with the compiled class of the load test program To automate the ZIP it s recommended to declare the hosts file as an external resource w o adding it to the CLASSPATH o dnssrv lt IP name server 1 gt lt IP name server N gt Effects that the load test job uses specific own DNS server s to resolve host names rather than using the DNS library of the underlying operating system When using this option at least one IP address of a DNS server must be specified Multiple DNS servers can be configured separated by commas If a resolved DNS host name contains multiple IP addresses the stressed Web servers are called in a round robin order user 1 uses resolved IP Address no 1 user 2 uses resolved IP Address no 2 etc o dnsenattl Enable consideration of DNS TTL by using the received TTL values from the DNS server s This option cannot be used in combination with the option dnspe
272. the display of the log vector by using the optional program argument dl debug loops or by selecting the debug option debug loops including var handler when starting the test run from the Web Admin GUI 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 166 171 Apica ProxySniffer V5 0 User s Guide 16 Direct Access to Measured Data The Proxy Sniffer Java API also contains classes and methods which allow direct access to all measured values stored within a statistics result file of a load test run prxres file This enables you to create your own extracts and or compilations from the measured data The main entry point to access these data is the method PerformanceData readObjectFromFile lt result file name gt 16 1 Example 1 Extracting Performance Data The following programming example extracts the most important performance data of the web pages and the URL calls import java io import dfischer utils PerformanceData import dfischer utils PerformanceDataRecord public class AnalyzeResult public static void main String args try read result file from disk PerformanceData performanceData new PerformanceData performanceData readObjectFromFile args 0 PerformanceDataRecord performanceDataRecord performanceData getPerformanceDataRecord display System System System System System System System System out out out out o
273. ther web page oe ne ct BM ab op oe Clicking on a bar inside the lower diagram displays the detailed results for the c PERDE I eee stators orale corresponding test run WTest 21 Testa sers E Test 3 Testd1 sers 20 Response Time Comparison per URL Average Page 1 Start Page WTest 11 Test 1 Users rt 27 Nov 2008 22 58 55 Du WTest 2 Test 1 Users 18 ats rt 27 Nov 2008 23 03 55 Du WTest 31 Testa aera art 27 Nov 2008 23 19 49 Du Hint Click inside the diagrams on the la bars to display details Done 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 139 171 Apica ProxySniffer V5 0 User s Guide 10 4 2 Performance Overview PRX Analyse Load Tests Test Result Comparison Mozilla Firefox E http 127 0 0 1 7990 dfischer webadmininterface AnalyseLoadtestResultDiffDiagramWheblet oe 0 Proxy Sniffer ab Admin Analyse LOgd Tests Test Result Comparison Test01 User 2 Start27 Nov 2008 22 58 55 Duration 2 21 min Session Failure Rate 0 00 Annotation ESTO Test01 User 10 Start 27 Nov 2008 23 03 55 Duration 2 29 min Session Failure Rate 0 00 Annotation Test01 User 20 Start 27 Nov 2008 23 19 49 Duration 3 08 min Session Failure Rate 35 19 Annotation Performance Overview Test 1 Test 2 Test 3 Passed Failed Loops 13 0 40 0 35 19 Average Session Time per Loop 20 36 sec 33 17 sec 70 60 sec Average Response Time per Page 0 89 sec 3 03 sec 9 27
274. tion 1 83 min Fai lures Er WTest 21 Test 1 Users 64 Start 16 Jun 2006 15 27 04 Duration 1 04 min Failures 100 00 WTest 23 Test 1 Users 93 Start 16 Jun 2006 15 27 05 Duration 1 03 min Failures 100 00 seconds Response Time Comparison per URL Page 1 Start Page url page 1 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 97 171 Apica ProxySniffer V5 0 User s Guide English Edition 9 3 Jobs Menu gt http 127 0 0 1 7990 PRX Main Menu Mozilla Firefox File Edit View History Bookmarks Tools Help P i o TAGS E g See Proxy Sniffer Main Menu SO 3 Ih qe b A dmi a Hel Web Page Personal Project Load Test Generate Analyse Refresh web Admin Professional Edition elp Tools Scanner Settings Navigator Jobs cad ad Load ens DENS Recorded Items 40 Q El I 00 QR PRX Jobs Mozilla Firefox All load test programs which are started from the Project http 127 0 0 1 7990 dfischer fwebadmininterface PopupDirectoryMavigatorDisplayJobswWeblet Lf N avi g ato r are always exec u ted as batc h j O bs by a n exte rn a m Sy Exec Agent process or by an Exec Agent Cluster This means a RT Exec Agent Jobs Local Exec Agent en x that it is not required to wait for the completion of a load test Display Exec Agent Jobs of Local Exec Agent Display Cluster Jobs program on the Execute Load Test window you can close the Ex
275. tory as the generated load test program 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 71 171 Apica ProxySniffer V5 0 User s Guide English Edition After compiling the load test program inside the Project Navigator you must first ZIP the compiled class of the load test program together with the ntlmauth txt file and then execute the zipped archive itself as load test program e Kerberos Authentication this option enables Kerberos authentication against web servers and or for outbound proxy servers If you choose the option use common Kerberos account from Personal Settings menu the same username and password is used for all simulated users By choosing the option apply individual Kerberos account per user from input file each simulated user uses its own username and password In such a case you must manually create a file named kerberosauth txt which contains on each line the username and the password of a user account These two line elements must be separated by semicolons The file must be located in the same directory where the generated load test program is stored After compiling the load test program inside the Project Navigator you are requested to ZIP the compiled class of the load test program together with all Kerberos configuration files After that the load test can be started by clicking on the corresponding ZIP file e HTTPS Client Certificates this option enables HTTPS X509 client cert
276. tory where the compiled load test program resides Then before you start the load test you have to zip the compiled class of the load test program together with all files which should be uploaded and also together with all used input files After this execute the zipped archive itself as load test program tsplaad ey Ex wernnuunnnnunnnnnnnnnnnnnen 5 sup Ww Wr p pd pa pa a pw DDR O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 64 171 Apica ProxySniffer V5 0 User s Guide 7 8 Overview of most commonly used Extract and Assign Options The following illustration is not exhaustive http 127 0 0 1 7990 Proxy Sniffer URL Details Var Handler Mozilla Firefox a a X Save Refresh Close 2 q go i Proxy Sniffer Hel Project Search Generate P M ata vers 030 a ci Web Admin URL Details Assign variable to the Assign variable to a protocol host name or CGI Parameter TCP IP port of an URL Replace a text pattern of a form parameter with a variable amp _ Assign variable to a ar J Status1 ins_step1 type rbLeben brameter Status1 ins_step1 subtype rbRate form parameter Content Paramete Status1 ins_step1 imgNext x 11 a Replace a text pattern of an Status1 ins_step1 imgNexty 8 v E rece esr TES moet URL call with a vari
277. tp 127 0 0 1 7990 Proxy Sniffer Search Overall Visible Items Mozilla Firefox of the parameter was found in a response received mr 3e a ee Search Overall Visible Items va from the web server HTTP response header or HTTP WebAdmin Help Close response content M Search ASCH Text 451047 IY Match Case M Include URL Encoded Values Hed arrows pointing to the right indicate that the Inside M HTTP Request Header Iv HTTP Request Content Motlnside M Referer value of the parameter was found in a request which M HTTP Response Header WM HTTP Response Content M Cookies Search was sent to the web server HTTP request header or HTTP request content em 19 GET https f control test atilevel accountlist aspx act first amp levid 94 153 Found in Response Content Line 232 Position 136 vSubtypelayer 51047 block You NOW need to extract the value from a response Found in Response Content Line 232 Position 193 viubtypelayeric1047 none s before it is sent the first time back to the web server In Foundin Response Content Line 233 Position 39 vSubtypelayerd51047 style posi this example this must be done on item 19 URL 19 Ilem 33 GET https ifxcontrol test atlevel agenda_list aspx id 45 1047 Found in Request Header Line 1 Position 32 list aspxtid 451047 Important also iS whether the value must be extracted Found in Response Content Line 10 Position 64 list aspx id 451047 id Forml from the HTTP response
278. tress Test your Web Applic User s think time O seconds 35 Max acceptable response time ms lt Mi a mi ae ee Gn E anm 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 23 171 Apica ProxySniffer V5 0 User s Guide i Note that you should invoke the Auto Configuration after all definitions of variables have already been made just before generating the load test program For that reason you can invoke the Auto Configuration also from the Generate HTTP S Load Test Program menu DER o Y PRX Serial Parallel Runtime Execution Mozilla Firefox 127 0 0 1 7990 dFischer webadmininterFace PapuplManageconcurrencywWweblek dataRecordId 1340398027875 Am Apica ProxySniffer Item 35 Configure Serial Parallel Runtime Execution e x Refresh Close URL 35 Runtime Execution Behavior Serial Execution Parallel Execution URL 35 Var Handler Overview Assigned Vars nana Extracted Vars nana C Disallow Auto Configuration for this URL protect from automatic modification Apply for all URLs with HTTP status code 200 OK AND response cantenttype image gif wt Apply Session Wide Settings Reset Configure Serial Runtime Execution for all URLs and allow Auto Configuration Reset All Serial Auto Configure Parallel Execution Session Wide Statistic Total Serial Executed URLS 119 Apply Auto Configuration
279. tup dat located inside the Proxy Sniffer installation directory and add the entry javaVirtuallpAddresses Enter here all IP addresses on one line separated by comma characters After these two steps have been completed you can start the load test by using the additional option multihomed which initializes the Exec Agent to use multiple local IP addresses when executing a load test This option is also supported by Exec Agent clusters load injector clusters in which case each load releasing cluster member Exec Agent uses its own configuration of client IP addresses Warning please contact your network administrator to get additional free IP addresses An incorrect configuration of additional IP addresses without consulting the network administrator may have an impact on several other computers of the same LAN such that these other computers could lose their network connection due to IP address conflicts 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 145 171 Apica ProxySniffer V5 0 User s Guide English Edition 12 1 1 Step1 Configuring Multiple IP Addresses at the Operating System Level 12 1 1 1 Windows Local Area Connection Properties Internet Protocol TCP IP Properties Advanced TCP IP Settings General Authentication Advanced General IP Settings DNS WINS Options Connect using You can get IP settings assigned automatically if pour network supports IP addresses BI Intel R
280. ues Elapsed Time 11 10 min Total Passed URL Calls Total Failed URL Calls Average Size Req Resp URL 1 Response Time GET http 192 16 4 5 80 166 Sampling Rate Remote Exec Agent Job 1883 Average Ws lues 13 Oct 2009 20 29 59 11 18 min elapsed URL 1 Response Time GET http 192 16 4 5 58 response time in milliseconds Max Response Time Min Response Time Av TOP Socket Connect Time 100 Av Request Transmit Time Av Response Header Wait Time Av Response Header Receive Time 0 05 00 0 04 30 0 04 00 0 03 30 0 03 00 0 02 30 0 02 00 0 01 30 0 01 00 0 00 30 0 00 00 Av Response Content Receive Time URL 1 Errors From Start of Test upto Current Time Real Time Profile of Error Types for URL 1 Remote Exec Agent Job 188 11 GET http 192 16 4 5 286 245 Errors number of errors per error type y Current Time B rnetuork Connection failed 243 Errors time 0 00 00 0 01 20 0 02 40 0 04 00 0 05 20 0 06 40 0 05 00 0 09 20 0 10 40 0 12 00 0 15 20 Load Test Test 13 Oct 2869 28 29 69 Elapsed Time 11 18 min User 266 DER EB http 1127 0 0 1 7990 dfFischer webadmininterface PopupDirectoryNawigatorDisplayLiveResponseTimeOFUrlWeblet jobld 188 execAgentld 12403061524068filePathB64 0zpcRG9rdw tIbnRIIHVuz CBFaWwSzdavsbHwvuz 5 7 So Y Help Refresh Close Measured Values Calculated Overall URL 1 Calls 6182 243 31 797 bytes 21 566 me 4 me 55 me
281. uest Content Annotation Request Headers amp Content Response Headers A i 0 00 00 0 01 20 0 02 40 0 04 00 0 05 20 0 06 40 0 08 00 0 09 20 0 10 40 0 12 00 0 15 20 Response Headers amp Content Load Test youtube 2 18 Apr 2611 18 33 66 User 16 All But wo Response Content All Full URL Snapshots trecommended will be display Detailed Measurement Summary of HTTP S Protocol Steps E URL gt Av Net Con gt Av Req Trm Av Header Rev AvContenlEcv MinTime AvTime MaxTime AvSize AvThroughput 40 100 2 ms 162037 ms 161 613 ms 1627081 ms 163369 ms 20 778 067 bytes 128 2 kbytes sec No Sample Measured on Offset ms 1 Offset min sec ms Response Time ms 1 Net Con Time de Req Trm Time m Wait Time m Header Rcv Time ms Content RE 1l18 Apr 2011 18 36 02 046 181296 3 01 296 161698 2 18 Apr 2011 18 36 43 140 222390 3 42 380 161833 Er 18 a 2011 18 36 58 703 237953 zaresala 57 953 161636 n j 3 learn VO ara ena ze v A AMATAS a pee LAr mat A ATO X 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 113 171 Apica ProxySniffer V5 0 User s Guide PRX Response Content Throughput Charts Mozilla Firefox FM It fe Proxy Sniffer Web Admin 40 GET ht Response Content Throughput Charts http 12 7 0 0 1 7990 dfFischerwebadmininterFace Poapup nalyseLaadtestDisplavResponseThraughputaraphsweb
282. uestContentType lt content type gt Allows the setting of a new default value for the request content type for all URL calls which contain request content data This overrides the default value used when this command is not applied of application x www form urlencoded e defaultRequestHeaderField request header field gt Allows the setting of an additional HTTP request header field to be applied for all URL calls This command can be called several times allowing the definition of several additional header fields Example defaultRequestHeaderField Accept Language en us e defaultResponseContentType content type gt Allows the setting of a default expected response content type such as text html The use of this command is only appropriate if all defined URLs return the same response content type By default the response content type of the URL calls will not be verified e autoPageBreak number of URLs think time random deviation Allows the automatic insertion of Page Breaks to be inserted after every specified number of URL definitions are processed The second 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 32 171 Apica ProxySniffer V5 0 User s Guide English Edition parameter the user s think time must be set in seconds and the third parameter the random deviation of the think time must be set in percent 0 100 addPageBreak lt comment gt lt think time gt
283. ulated user user var scope the variable is initialized with the outer loop counter of the actual simulated user 0 1 2 inner loop counter the variable is initialized with the iteration counter of the inner loop 0 1 2 of the actual simulated user system time milliseconds the variable is initialized with the current operating system time in milliseconds since 1970 load source IP host name the variable is initialized with the Exec Agent host name load source IP address the variable is initialized with the Exec Agent IP address 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 66 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 10 J2EE URL Rewriting A Java J2EE application server can be configured by the developers of the web application such that a procedure called URL rewriting is used to build the session context instead of using session cookies In this case the server will assign at runtime a special dynamic session parameter to every returned hyperlink and to every form which contains the session context An example of a hyperlink with applied URL rewriting is as follows A HREF http www d fischer com 8080 prxtool serviet WebMainMenu jsessionid bu3fy0bbj1 currentDir 344 gt weiter lt A gt The URL rewriting parameter is appended to the URL file path separated by a semicolon and appears before the normal CGI parameters which start with a question m
284. ults This menu allows to modify the response verification algorithm Session Cutter allows one or more recorded web surfing sessions to be combined into a single new web surfing session using a process analogous to the splicing of motion picture film Additionally this menu allows to import web surfing sessions from external definition files from which load test programs can be created Execute Load Test displays the most important statistics during the execution of a load test Errors can be displayed and analyzed in real time as they occur Load Curve Diagrams displays the performance curve of a web server or web application under load showing how response time throughput and stability behave under various load conditions The maximum performance capacity of a web server or web application can be determined using this menu Comparison Diagrams provides a graphical comparison of the response times of the same load test program which was executed at different times for example before and after server tuning activities allowing the effect of the tuning on response times to be determined Detail Statistics amp Diagrams displays in detail all collected measurements related to a single load test Over 21 different statistics and diagrams are available Error Details shows the details of all errors occurring during a load test error snapshots This menu can be invoked during the load test as well as after the completion of a load test
285. umber nf Concurrent Users E rT Diagram Web Transaction Rate Diagram Completed Loops Diagran Diagram HTTP Keep Alive Efficiency Diagram SSL Cache Efficiency Diagra Diagram Number of Errors per Page Diagram Number of Errors per URL Diagra Test Scenario Objectives Test Start Date 17 Feb 2008 00 52 26 Load Test Program Test01 class Load Source Host fischer 110 5 0 1 Load Source OS Load Source OS Windows AP Target Hosts www prosas niffer ch B0 ee carm at Applied HTTP Version Test Input Parameter User Input Fields Concurrent Users E User s Think Time 10 Planned Test Duration 1 00 min O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 51 171 Apica ProxySniffer V5 0 User s Guide English Edition 7 4 Load Test Plug Ins Proxy Sniffer Load Test Plug Ins are Extension Modules to the Proxy Sniffer product Plug Ins are configured using the GUI and are executed during a Load Test The following Plug Ins are already predefined and delivered as part of the Proxy Sniffer installation Plug In Designation in the GUI Plug In Functionality Abort Failed Test Aborts a running Load Test if too many errors occur within a configured time interval Assign File Data to Request Content Read the data of a file from disk and assign it to the request content of an URL call only useful for HT TP S POST requests and some WebDAV methods Coo
286. ut out out out English Edition overall data println users performanceData getParallelUsers println test duration performanceData getTestDurationMillis 1000 seconds println hits per second performanceData getWebTransactionRate println passed loops performanceData getPassedLoops println failed loops performanceData getFailedLoops println average response time per page float performanceData getAveragePageTime 1000 0f seconds println average network connect time per URL call performanceData getAverageNetworkEstablishTime milliseconds println display page data int pageBreakIndex performanceData getPageBreakIndexes for int x 0 x lt pageBreakIndex length x String pageName performanceDataRecord pageBreakIndex x getInfoText long pageResponseTime performanceData getPageTime pageBreakIndex x get all url calls per page int urlIndexesOfPage performanceData getValidUrlIndexesOfPage pageBreakIndex x calculate average size of page long pageSize 0 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 167 171 Apica ProxySniffer V5 0 User s Guide English Edition long pageTime 0 long cumulatedPageSize 0 for int y 0 y lt urlIndexesOfPage length y PerformanceDataRecord urlDataRecord performanceDataRecord ur
287. va Classname gt EST_01 Java Code Model large v Excellence in Cloud Se mede Content Test Algorithm apply heuristic methods from recorded session to check received conter Character Encoding ISO 8859 1 v Generate External Files for XML and SOAP Request Data gt 4096 Bytes w URL Execution 3 required enter a simple classname for the load test program with no path and no file extensign Serial Executed 19 Parallel Executed 82 mu Threads p User B HTTP Protocol Options E HTTP Protocol Version 11 1 Allow Keep Alge Strip Referer Header Field Strip Accept Header rieg ui Load Test over HTTP S Proxy Apply next proxy configuration fropfersonal settings Basic Authentication O Apply individual Besit Authentication per user from input file basicauth bd Digest Authentication O O Apply indgaual Digest Authentication per user from input file digestauth bd E use gofimon Username Password NTLM Authentication O e common NTLM account from Personal Settings menu Kerberos Authentication use common Kerberos account from Personal Settings menu v HTTPS Client Certificates a O apply individual PKCS 1 2 certificate per user from input file pkes1 2auth bd v bn Descripgt gt recommended will be displayed as hint in Project Navigator 2012 2013 Ingenieurb ro David Fischer AG Switzerland A Apica Cy MyTests E Display Load Test Program TEST
288. ver If you choose the option use common Username Password the same username and password is used for all simulated users By choosing the option Apply individual Digest Authentication per user from input file each simulated user uses its own username and password In such a case you must manually create an input file named digestauth txt which contains on each line the username and the password per simulated user These two line elements must be separated by semicolons The input file must be located in the same directory where the generated load test program is stored After compiling the load test program inside the Project Navigator you must ZIP the compiled class of the load test program together with the digestauth txt file and then you must execute the zipped archive itself as load test program NTLM Authentication this option enables NTLM Windows authentication If you choose the option use common NTLM account from Personal Settings menu see chapter 0 the same NTLM username and password is used for all concurrent users By choosing the option apply individual NTLM account per user from input file each simulated user uses its own username and password in which case you must manually create an Input File named ntlmauth txt which contains a line for the domain the username and the password for each simulated user These three elements on each line must be separated by semicolons The Input File must be located in the same direc
289. w the Remote Exec Agent 192 16 4 20 33 33 133 Concurrent User Input File data would be distributed across the cluster AE PEAD 2293 133 Ceme tieni wea members If you do not use the split functionality each cluster member would receive an entire copy of the Input File File Name Line Comment Tag Split Fig The distribution of users across the cluster members can also voor eh OLE be modified manually however this is useful only if a cluster member is currently not available marked with light red ee a AAA background in which case the cluster job can not be started schedule Cluster Job 206 for Day Time HourMinute 15 e 21 iw Mote select also Split Input Files options In this Case you can assign the users of the unavailable cluster member to other cluster members and then try to start the cluster job again This redistribution may take a few seconds to complete Display Real Time Statistic Alternatively the load test program can also scheduled to be executed at a predefined time However the local Job Controller process must be available running at the predefined time because the scheduling entry for the cluster job is stored inside the Job Controller working directory which is monitored by the Job Controller itself Especially if you have started the Job Controller implicitly by using the Proxy Sniffer Console you must keep the Proxy Sniffer Console Window open in order that the cluster job will be started 1
290. which coordinates all cluster members Exec Agents The cluster job controller creates a cluster job and allocates a cluster job number The cluster job is now in the state configured ready to run but not yet started PRX Start Cluster Job 206 Mozilla Firefox alada The number of concurrent users will be automatically EB http HH127 0 0 1 7990 dfischer webadmininterFace PopupclusterStarkConfigLIserswWwebleE jobId 2068 concurrentLIsers 400 5f distributed across the cluster members depending on the 5 capability of the individual computer systems called load H gt Proxy Sniffer al Hi web Admin Help Jobs Refresh Close factor In cases where the load test program uses Input Files you are Cluster Job State configured a Test Test01 asked for each Input File if you wish to split the content of Test Arguments 3 U 400 d 1200 t 60 sdelay 200 maxloops DO sampling 15 percpage 100 percurl 20 maxerrmem 20 nalog the Input File This can be useful for example If the Input File Concurrent Users 400 Saran ae eM TET RS contains user accounts usernames passwords but the web Max Loops per User unlimited application does not allow duplicate logins In this case each Cluster Member Host Load Factor Users Default Modify Load Distgibution 3 cluster member must use different user accounts By clicking LocalExecagent 127 0 01 33 33 134 Concurrent User on the corresponding magnifier icon you can view ho
291. which has an unlimited size 1 GB Large Response Content Allows to receive response content data of a large size up to 2 GB for one or several URLs Note that all response data are read as usual during load test execution but that only a part of them are stored internally Limit the receiving of response content data to a specified size Further reading of data from the web server during load test execution is aborted skipped for the configured URL when the maximum size is reached PKCS 11 Security Device Support for Smart Cards PKCS 11 Security Devices which contain a SSL Client Certificate used for authentication against web servers Remove Cookie Removes a cookie from the cookie store of a simulated user 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 52 171 Limit Response Content Apica ProxySniffer V5 0 User s Guide English Edition User Synchronization Point 1 Retains all active users at a configurable synchronization point until all of the users have reached this point After that the users are rereleased by applying a configurable deblock delay which is multiplied with the no of the actual user 0 1 2 The configuration of a Plug In respectively adding a Plug In to a recorded web surfing session can be done in the Var Handler N A N e N 2 a 3 amp X amp 90 amp eX a d amp ex i ct at
292. wn in the Var Finder Activate the checkbox Assign var to all request parameter with same recorded value and let the checkbox Try URL Encoding remain activated Then click on Extract Var Handler Go db ES X MIEWSTATE 1 loop var m 48 HTML Farm Parameter 51 amp HTTP Request Content Paramet WIEW STATE 2 loop var 52 2 HTML Form Parameter 54 HTTP Request Content Paramet VIEWS TATE 3 loop var 55 amp HTML Form Parameter 57 HTTP Request Content Paramet id 1 loop var A Text Line Pattern amp HTTP Request URI amp HTTP Request URI 2 HTTP Request URI lewid loop var 2 amp HTML Hyperlink 19 2 HTTP Request URI All Rights Reserved The configuration inside the Var Handler now shows that the value of the parameter is extracted from URL 19 and assigned to the URLs 33 35 and 38 This matches exactly with the first estimate which was made by clicking on the magnifier icon inside the Var Finder Hint in this example you would have to repeat the same steps to handle the second value of the parameter id It is recommended that you save the recorded web surfing session periodically after making changes inside the Var Handler Page 60 171 Apica ProxySniffer V5 0 User s Guide English Edition Further Hints Dynamically Exchanged Session Parameters can also be extracted in an easy way from redirections from forms and from hyperlinks You should always use th
293. work Influence Average Network Connect Time Average Network Connect Time per created TCP IP Connection milliseconds 150 200 concurrent user successful completed loops per minute sessions per minute Overall Network Throughput total network throughput that is network load O 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved You can click within the diagrams on the red rhombuses to display the detailed results of the corresponding test run 9 different diagrams are displayed Average Session Time per User per Loop cumulative time for a loop per user that is response time behavior of the server Web Transaction Rate Hits per Second number of successfully executed URL calls per second hits per second that is server throughput Session Failure Rate percentage of failed loops that is server stability Average TCP Socket Connect Time average time per URL call to open a network connection that is network performance in combination with the TCP IP stack performance of the server Users Waiting for Response average of the number of users which are waiting for response from the server URL Error Rate percentage of failed URL calls HTTP Keep Alive Efficiency percentage of reused network connections SSL Session Cache Efficiency percentage of abbreviated SSL handshakes Completed Loops per Minute the number of Page 135 171 Apica ProxySniffer V5 0 Us
294. x29 29 045sec 1M 55668byes 31ms MMGET http 192 16 45imap earth aif You can define an inner loop by clicking on the item index at the left side of a page break Inner Loop Configuration e Inner Loop Description description of the inner loop mandatory eese tom 4 32 e Inner Loop End Page the end page of the inner loop including all URL C Ba calls on the end page itself Inner Loop Description opt 0 e Loop Iterations number of iterations This can be a fixed value or a Inner Loop Start Fage Fage 3 Login Main Menu variable value which can be extracted for example from an Input File or Inner Loop End Page Page 3 Login Main Menu from a User Input Field see Chapters 7 2 and 7 3 eee ieee comede miles e Action if planned duration of Load Test exceeded the option Abort C variable 8 gt loops amp current loop after current iteration means that at the end of the load test EPUM act inner op aner curent ration T when the maximum duration of the test has elapsed the inner loop is ML E i aborted after the end of the current iteration and remaining iterations are Enable Pacing fedta 60 v seconds not executed The option Continue with iterations means that the end of en variable f seconds the load test will be postponed until all iterations have been completed m um Apply e Enable Pacing enabling this option sets a minimum
295. ystem parameters of the operating system on which the Exec Agent runs 11 The network connection to an outbound SSL proxy server has failed 10 Unknown host DNS problem or wrong hostname 9 Unable to open the network connection to the web server connection refused 8 The web server has first accepted but later closed aborted the network connection before all response data have been received connection reset by peer 2012 2013 Ingenieurb ro David Fischer AG Switzerland All Rights Reserved Page 130 171 Apica ProxySniffer V5 0 User s Guide English Edition The web server response violates the HTTP protocol invalid protocol data have been received 2 Request timeout expired no response from web server The URL call was aborted by the load test program 1 Generic request error If the HTTP response content was received in HTML format the content of the defective web page can be displayed in the web browser without images by clicking on Display Response in Web Browser This web page is taken directly from the data of the captured error snapshot therefore the defective web page can also be displayed even if the web server is no longer reachable HTTP Request Header gt 1 POST prxtool servietVVebMainMenu HTTP 1 1 Host 192 16 4 33 8080 User Agent Mozilla 5 0 Windows U Windows NT 5 1 en US rvc1 8 0 4 Gecko 20060508 Firefox 5 0 4 Accept mE m Accept Language en us Accept Encoding gzip de

Proxy Sniffer User`s Guide

Contents

Download Pdf Manuals

Related Search

Related Contents