DX Unified Infrastructure Management

  • 1.  nexec probe timeout ignored

    Posted Feb 19, 2018 05:30 AM

    Hello all,

     


    I am developing a new inventory system that makes use of the nexec probe.

     

    When making a callback via the Ctrl-P utility, I can set the timeout to whatever value I like.

     

    However, no matter what the setting, the timeout seems to be fixed to 16 seconds.

     

    In my test environment, it works just fine. Out in our live estate it does not.

     

    I have tried to eliminate firewall rules and tunnels by making the callback between two servers that sit next to each other on the same subnet. That didn’t make a difference.

     

    In test the version is 7.80 for robot and hub.

     

    The versions in live are 7.92HF11 for the robot and 7.92HF10 for the hub.

     

    The nexec probe is version 1.36 in all cases. If I write a custom probe which uses the nimNamedRequest function for making callbacks, I see the issue there too.

     

    Does anyone know why timeout seems to be overridden somewhere?

     

    Thanks

    Paul



  • 2.  Re: nexec probe timeout ignored

    Broadcom Employee
    Posted Feb 21, 2018 03:42 AM

    Thanks for reaching out to the community, Paul.

    As we are working on a related support case I will mark this thread as answered.

    When the issue is resolved we will post here the resolution notes. 



  • 3.  Re: nexec probe timeout ignored

    Posted Feb 21, 2018 04:33 AM

    But the question is not answered?

     

    Please can we keep the question open as it will give someone else who might have had the problem, the chance to answer.



  • 4.  Re: nexec probe timeout ignored

    Broadcom Employee
    Posted Feb 21, 2018 06:13 AM

    Thanks for the update, Paul. On the support case you have also logged additionally to this thread, we are pending to know if you have the same issue when running the nexec probe on a robot 7.93.



  • 5.  Re: nexec probe timeout ignored

    Broadcom Employee
    Posted Mar 06, 2018 10:07 AM

    Hello Paul, is this still a question for you ? I saw in the referenced case that you got steps that should help to resolve this.

     

    Kind regards,

    Britta Hoffner

    CA Support



  • 6.  Re: nexec probe timeout ignored

    Broadcom Employee
    Posted Mar 06, 2018 11:47 AM

    I converted this thread in an open discussion. The root cause is not identified, but this is the progress done:

     

    Seems like the execution fails "socket disconnected " error sockCloseErr: 0000000000A5F610. regardless the timeout configured.

     

    Failing system:

     

    Feb 27 09:42:17:610 [18052] nexec: (r_p) - profile ping, command ping 127.0.0.1 -n 100, args
    Feb 27 09:42:17:610 [18052] nexec: (r_p) - profile ping, commandline ping 127.0.0.1 -n 100
    [...]
    Feb 27 09:42:47:690 [18052] nexec: SREPLY: status = 0(OK) ->12.34.56.78/62505
    Feb 27 09:42:47:815 [24668] nexec: sockCloseErr: 0000000000A5F612.34.56.78/62505 socket disconnected
    Feb 27 09:42:47:815 [24668] nexec: sockClose:0000000000A5F610:12.34.56.780/48013

     

     

     

    Below is a test from my lab with timeout 2 secs (for reference of working scenario)

     

    Feb 27 11:27:37:526 [324104] nexec: (r_p) - profile ping, command ping 127.0.0.1 -n 100, args
    Feb 27 11:27:37:526 [324104] nexec: (r_p) - profile ping, commandline ping 127.0.0.1 -n 100
    [..]
    Feb 27 11:27:40:634 [324104] nexec: SREPLY: status = 0(OK) ->12.34.56.78/49809
    Feb 27 11:27:41:525 [321764] nexec: RREQUEST: _close <-12.34.56.78/49809 h=344 d=0
    Feb 27 11:27:41:525 [321764] nexec: sockClose:000000000076A42012.34.56.78/48054

     

    Suggestions provided:

    Check if any Antivirus or Malware protection running on this 7.93 windows 2012 robot 
    Check if nexec on an other robot on the same UIM instance fails the same way
    Check what user is running the Nimsoft Service controller? And compare with a working scenario