Thread: [OpenSTA-users] OpenSTA making n number of connections
Brought to you by:
dansut
Menu
▾
▴
opensta-users
|
From: Dev G. <dev...@gm...> - 2007-02-19 14:56:45
|
Hi All, I need your help, Here is a situation... I am testing a web application which is using Oracle database where we have setup connection pool with the limit of 75 connections, We are running a test using Load Runner with the settings of 150 VUs ( Duration of test-: 1/2 hr ) and then we monitor and find that only 50 connections are bsing used from the connection pool. But when we run the same test using OpenSTA with the settings of 150 VUs ( Duration of test-: 1/2 hr ), we find all 75 database connections are being used and if we try to open application , we get error message saying "increase the connections in connection pool". However i know OpenSTA sends HTTP requests to webserver and receive response sent by webserver. More info about script--> Script is parameterized. Doing modular testing. (one script contains login information, second script contains navigation in application) If any other infomation is required about the script, please let me know. If anyone has ever faced such type of problem, please suggest me the solution as soon as possible. Thanks Dev |
|
From: Dan D. <ddo...@me...> - 2007-02-19 17:06:33
|
Dev, Seems you may be driving the app harder with STA...hence exhausting all 75db pool connections. Are you sure the think times in the STA script are the same as the think times in the LR script? ..Dan Dan Downing www.mentora.com -----Original Message----- From: ope...@li... [mailto:ope...@li...] On Behalf Of Dev Grover Sent: Monday, February 19, 2007 9:57 AM To: ope...@li... Subject: [OpenSTA-users] OpenSTA making n number of connections Hi All, I need your help, Here is a situation... I am testing a web application which is using Oracle database where we have setup connection pool with the limit of 75 connections, We are running a test using Load Runner with the settings of 150 VUs ( Duration of test-: 1/2 hr ) and then we monitor and find that only 50 connections are bsing used from the connection pool. But when we run the same test using OpenSTA with the settings of 150 VUs ( Duration of test-: 1/2 hr ), we find all 75 database connections are being used and if we try to open application , we get error message saying "increase the connections in connection pool". However i know OpenSTA sends HTTP requests to webserver and receive response sent by webserver. More info about script--> Script is parameterized. Doing modular testing. (one script contains login information, second script contains navigation in application) If any other infomation is required about the script, please let me know. If anyone has ever faced such type of problem, please suggest me the solution as soon as possible. Thanks Dev ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV -- OpenSTA-users mailing list Ope...@li... Subscribe/Unsubscribe/Options: http://lists.sf.net/lists/listinfo/opensta-users Posting Guidelines: http://portal.opensta.org/faq.php?topic=UserMailingList |
|
From: Bernie V. <Ber...@iP...> - 2007-02-19 17:38:49
|
> Dev, > > Seems you may be driving the app harder with STA...hence exhausting all > 75db pool connections. Are you sure the think times in the STA script are > the same as the think times in the LR script? > Dan makes a good point. Start by comparing the throughput of the two methods. You can calculate this by dividing the count of some significant timer(s) by total test runtime. If you don't have a timer or set of timers that would allow a direct comparison of throughput under LR and OpenSTA, then just count the number of script iterations completed by all users divided by the elapsed time of the test. If throughput is roughly the same then I'd check to be sure that you haven't inadvertently left some connections open (i.e. you might no longer be executing disconnect commands because of the loop control logic you have added to the script. ) If it is script pacing that is the culprit, it may be due to something simple like needing to tweak wait times so the two scripts are more equivalent. You may also need to look at manually inserting synchronize commands to make OpenSTA playback more realistic. -Bernie www.iPerformax.com |
|
From: Dev G. <dev...@gm...> - 2007-02-21 05:05:42
|
Thanks Dna and Bernie, As suggested i checked Think Time, In LoadRunner and OpenSTA we are using same Think Time, In STA script few connections were not closed so i used "DISCONNECT" command to close the connections, I ran one more test with just 50 VUs but still its using 75 connections but the improvement is that earlier almost 40 call were in que but now only 20 - 30 calls are in que. Please let me know if anything else i can try. Thanks Dev On 2/19/07, Bernie Velivis <Ber...@ip...> wrote: > > > > Dev, > > > > Seems you may be driving the app harder with STA...hence exhausting all > > 75db pool connections. Are you sure the think times in the STA script are > > the same as the think times in the LR script? > > > > Dan makes a good point. Start by comparing the throughput of the two > methods. You can calculate this by dividing the count of some significant > timer(s) by total test runtime. If you don't have a timer or set of timers > that would allow a direct comparison of throughput under LR and OpenSTA, > then just count the number of script iterations completed by all users > divided by the elapsed time of the test. If throughput is roughly the same > then I'd check to be sure that you haven't inadvertently left some > connections open (i.e. you might no longer be executing disconnect commands > because of the loop control logic you have added to the script. ) > > If it is script pacing that is the culprit, it may be due to something > simple like needing to tweak wait times so the two scripts are more > equivalent. You may also need to look at manually inserting synchronize > commands to make OpenSTA playback more realistic. > > -Bernie > www.iPerformax.com > > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > -- > OpenSTA-users mailing list Ope...@li... > Subscribe/Unsubscribe/Options: http://lists.sf.net/lists/listinfo/opensta-users > Posting Guidelines: http://portal.opensta.org/faq.php?topic=UserMailingList > |
|
From: Dan D. <ddo...@me...> - 2007-02-21 18:30:38
|
Dev wrote: >As suggested i checked Think Time, In LoadRunner and OpenSTA we are >using same Think Time, >In STA script few connections were not closed so i used "DISCONNECT" >command to close the connections, I ran one more test with just 50 VUs but still its >using 75 connections but the improvement is that earlier almost 40 call were in >que but now only 20 - 30 calls are in que. Did you test using SYNCRHONIZE after each group of PRIMARY GET/POST plus subsequent GET URIs, as per Bernie's earlier suggestion? Suggest you read about this instruction in the Help ..Dan www.mentora.com |
|
From: Dev G. <dev...@gm...> - 2007-02-22 04:54:14
|
Hi Dan, I have implemented Bernie's suggestion but havn't been able to not run the test because development team is doing some enhancements in App Server, As soon as i'll run the test, i'll check the result and will share my experience. Thanks Dev On 2/22/07, Dan Downing <ddo...@me...> wrote: > Dev wrote: > >As suggested i checked Think Time, In LoadRunner and OpenSTA we are > >using same Think Time, > >In STA script few connections were not closed so i used "DISCONNECT" > >command to close the connections, I ran one more test with just 50 VUs but > still its >using 75 connections but the improvement is that earlier almost > 40 call were in > >que but now only 20 - 30 calls are in que. > > Did you test using SYNCRHONIZE after each group of PRIMARY GET/POST plus > subsequent GET URIs, as per Bernie's earlier suggestion? Suggest you read > about this instruction in the Help > > > ..Dan > www.mentora.com > > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > -- > OpenSTA-users mailing list Ope...@li... > Subscribe/Unsubscribe/Options: http://lists.sf.net/lists/listinfo/opensta-users > Posting Guidelines: http://portal.opensta.org/faq.php?topic=UserMailingList > |
|
From: Daniel S. <da...@Op...> - 2007-02-23 03:30:49
|
Bernie Velivis wrote: > You may also need to look at manually inserting synchronize > commands to make OpenSTA playback more realistic. I think this statement needs a little explanation. Inserting SYNCHRONIZE commands after the primary GET can make replay more realistic as load increases because it allows the script to react to the extended wait and load times of this GET before starting its secondary GETs. However, doing this means that your script will no longer start downloading the secondary items in a page until the whole of the primary (HTML) is downloaded - and this is *not* how a browser behaves. The default recorded scripts do not react to extended download times and therefore as load increases the secondaries may end up being requested before the primary download has even got to the point where it should know about the secondary - this is definitely unrealistic and why using the SYNCHRONIZEs is seen as "more" realistic. Bernie knows the above and understands the comprimises involved - you should also attempt to do this before you make your script do something which isn't really what you want. ;) As far as diagnosing your "extra" DB connections - I would be looking at tracing what your Web app is doing under load ... after all, it is this that is making the DB connections and not OpenSTA. Cheers /dan -- Daniel Sutcliffe <Da...@Op...> OpenSTA part-time caretaker - http://OpenSTA.org/ |
|
From: Osoata, C. <Chr...@at...> - 2007-02-20 09:27:41
|
Hi All, I am in the process of using OpenSTA for performance testing, however I don't understand the graphs that are displayed for the result. Can anyone please inform me the best graph to determine the performance of a website application. There are currently several graphs but I'm not sure what the results are showing, I want to be able to determine the performance for different number of virtual users. Many thanks Chris This email and any attached files are confidential and copyright protected.= If you are not the addressee, any dissemination of this communication is s= trictly prohibited. Unless otherwise expressly agreed in writing, nothing s= tated in this communication shall be legally binding. The ultimate parent company of the Atkins Group is WS Atkins plc. Register= ed in England No. 1885586. Registered Office Woodcote Grove, Ashley Road, = Epsom, Surrey KT18 5BW. Consider the environment. Please don't print this e-mail unless you really = need to.=20 |
|
From: Dan D. <ddo...@me...> - 2007-02-20 14:26:28
|
Chris, If there is one key graph that captures the essence of stress testing it is "timer values vs active users" -- as this illustrates the "scalability" of the business process and "pages" of the app, i.e., "how does response time vary as load is increased". However, to assess "reliability" of the test, you need to look at the error log and the http data list (filtered for >399 and <599 errors). Too many timeout, script, or http errors could invalidate your results, and you should strive to at least explain if not eliminate most errors. Also, a very key graph is Custom NT Performance, created by defining a Collector that captures Bytes Sent and Bytes Received from your load injector. Graphed against Active Users, you can gauge whether there is a bandwidth throughput bottleneck somewhere in your system. My opinion: The opensta graphs are rather primitive and limited, and I much prefer exporting Timers and graphing in Excel (using Pivot Tables to summarize values into 1 or 2 minute intervals. You can view examples on our website). ..Dan www.mentora.com -----Original Message----- From: ope...@li... [mailto:ope...@li...] On Behalf Of Osoata, Christabel Sent: Tuesday, February 20, 2007 4:27 AM To: OpenSTA users discussion and support Subject: [OpenSTA-users] Reading the results Hi All, I am in the process of using OpenSTA for performance testing, however I don't understand the graphs that are displayed for the result. Can anyone please inform me the best graph to determine the performance of a website application. There are currently several graphs but I'm not sure what the results are showing, I want to be able to determine the performance for different number of virtual users. Many thanks Chris This email and any attached files are confidential and copyright protected. If you are not the addressee, any dissemination of this communication is strictly prohibited. Unless otherwise expressly agreed in writing, nothing stated in this communication shall be legally binding. The ultimate parent company of the Atkins Group is WS Atkins plc. Registered in England No. 1885586. Registered Office Woodcote Grove, Ashley Road, Epsom, Surrey KT18 5BW. Consider the environment. Please don't print this e-mail unless you really need to. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV -- OpenSTA-users mailing list Ope...@li... Subscribe/Unsubscribe/Options: http://lists.sf.net/lists/listinfo/opensta-users Posting Guidelines: http://portal.opensta.org/faq.php?topic=UserMailingList |
|
From: Danny R. F. <fa...@te...> - 2007-02-20 14:51:50
|
Dan Downing wrote: > My opinion: The opensta graphs are rather primitive and limited, and I much > prefer exporting Timers and graphing in Excel (using Pivot Tables to > summarize values into 1 or 2 minute intervals. You can view examples on our > website). Thanks, Dan - you have a nice sample report. I'm not an Excel power user, and I haven't found any useful documentation explaining how pivot tables work. Can you recommend a reference? I want to invite you and any other testing service providers here to submit a listing to the testingfaqs.org Contractors and Consultants List if you're not already listed there. The form to add a listing is http://testingfaqs.org/newcon.html. If you have submitted a listing and don't see it posted yet, polite but persistent encouragement is helpful. :-) Disclosure - the listing is free, and the free listings are the primary focus of the site, but I may give you a low-key pitch to become an advertiser. -- Danny R. Faught Tejas Software Consulting http://tejasconsulting.com/ |
|
From: Daniel S. <da...@Op...> - 2007-02-23 04:10:49
|
Dan Downing wrote: > My opinion: The opensta graphs are rather primitive and limited, > and I much prefer exporting Timers and graphing in Excel It might be your opinion, but it is also a fact ;) The graphs in the OpenSTA Commander Results were intended to be very basic - allowing the production of a commercial Results analysis and report generation module (which was never made) and to encourage the sale of more consulting on top of OpenSTA. This wasn't really a secret and is still a revenue generation model that could work for OpenSTA - someone has to fund its development ... Cheers /dan -- Daniel Sutcliffe <Da...@Op...> OpenSTA part-time caretaker - http://OpenSTA.org/ |
Thanks for helping keep SourceForge clean.
X