none
Wierd behaviour: WCF SQL send port execution time increases for same operation RRS feed

  • Question

  • I am using WCF SQL adapter (BTS 2009) (request-response send port : UseAmbientTransaction is False, using PassThru pipeline) to receive typed messages by executing SP. The issue I am facing is, when I execute same message multiple times, I see an consistent increase in SQL operation time that too only send operation. When I run 1st time, SQL send operation (start and end time of Send Shape in Orch debugger) takes 1sec, 2nd run takes 4sec, 3rd run takes 13 sec, 4th run takes 25sec, this leads to increase in total time for the process.

    Also, one thing I noticed is when I use scope (Transaction Type: None) then first SQL send takes more time and rest of the send operation within same scope takes less than 1 sec time.

    Have anyone seen such behaviour or any idea what could be wrong? Any suggestion to improve performance for WCF SQL adapter?


    Regards, Ajeet Kumar MCTS Biztalk Server

    Thursday, June 28, 2012 7:25 AM

Answers

  • there could be memory leaks add private bytes counter in perfmon and see if its increasing, we had similar issue with biztalk 2009 after upgrading to 2010 issue got resolved.

    also have you installed cumulative updates for adapter pack & biztalk?


    KiranMP

    Thursday, June 28, 2012 10:27 AM
  • Are you doing two actions on particular table that can be blocking the other then it might causing time-out issue. Try to investigate like sp_who using Management studio.


    Thanks, Raja MCTS BizTalk Server 2010, MCC If this answers your question please mark it accordingly


    Monday, July 2, 2012 8:00 AM

All replies

  • The first run usually takes a while and post that the subsequent runs are quicker as most properties are cached thereby improving the run time.


    Regards,
    Bali
    MCTS: BizTalk Server 2010,BizTalk Server 2006 and WCF
    My Blog:dpsbali-biztalkweblog
    -----------------------------------------------------
    Mark As Answer or Vote As Helpful if this helps.

    Thursday, June 28, 2012 7:58 AM
  • Yes, I agree with you. But here I see completly different behaviour. Also, I noticed, this trend gets reset if I restart Host Instance i.e. the 1st run after Host Instance restart takes around 1sec, 2nd run 2.5 sec, 3rd run 3 sec and keep on increasing with subsequent run. Also each run has multiple SQL operation and the time for these SQL operation (within the same run) remains same.

    Regards, Ajeet Kumar MCTS Biztalk Server

    Thursday, June 28, 2012 8:34 AM
  • there could be memory leaks add private bytes counter in perfmon and see if its increasing, we had similar issue with biztalk 2009 after upgrading to 2010 issue got resolved.

    also have you installed cumulative updates for adapter pack & biztalk?


    KiranMP

    Thursday, June 28, 2012 10:27 AM
  • Are you doing two actions on particular table that can be blocking the other then it might causing time-out issue. Try to investigate like sp_who using Management studio.


    Thanks, Raja MCTS BizTalk Server 2010, MCC If this answers your question please mark it accordingly


    Monday, July 2, 2012 8:00 AM
  • Hi, did you solve the issue?

    I have the same problem in a test environment. I updated BT2009 to cumulative update 6.

    I created a simple orchestration that receives a file, calls a stored procedures via SQL Adapter and sends the result to a folder via FILE Adapter.

    The last step takes very long time, I see the send port service (class Messaging) running in BizTalk Admin Console and if I restart the host instance the file is immediately sent to folder. This is systematic.

    I created an host and relative host instance dedicated for send, but It hasn't solved the issue.

    Thks and regards

    Massimo


    Massimo

    Wednesday, December 12, 2012 4:01 PM
  • Is the Tracking enabled on Send Port, which might be causing the delay in committing the transaction to Tracking DB.
    Wednesday, February 6, 2013 2:55 PM