none
FIM 2010 R2 SP1: stopped-server on FIM MA RRS feed

  • Question

  • Upgraded FIM 2010 to FIM 2010 R2 recently i my dev environment recently.

    Decided to clear all the objects out of the FIM Portal, but ran into a problem when I discovered [debug][DeleteOrphanedRulesByType] is gone in FIM 2010 R2 SP1.

    Decided to use  [debug].[PurgeObjectsOfType] to get rid of ALL EREs (it's a dev environment after all), but now I get a "stopped-server" error on import.

    I note  in 4.0.3627.2 for FIM 2010 that extra logging detail was added to the Application Log, but I don't see an equivalent R2 hotfix. Certainly, nothing is showing up in the Event log except:

    The management agent "FIMMA" failed on run profile "Full Import" because the server encountered errors.

    Which is not much help

    Order of steps to produce this issue was:

    - Clear out users with Powershell (not deleting my admin users
    - Clear out groups with Powershell
    - Full Import on FIM MA. Stop when I realise EREs are coming in
    - Run PurgeObjectsOfType
    - Perform Full Import -> receive "stopped-server" message

    Have tried:

    - Restarting the development server this runs on
    - Reinstalling FIM Service and Portal
    - Confirming that I am using the correct FIM MA account
    - Checked request log on FIM Portal - nothing shows.
    - Checked Application Log and the three FIM "Applications and Services Logs" in Event Viewer.
    - Turned debug tracing on in FIM Service. Nothing shows

    Anyone have any suggestions?

    - Ross Currie


    FIMSpecialist.com | MCTS: FIM 2010 | Now Offering ECMA1->ECMA2 Upgrade Services

    Wednesday, March 20, 2013 2:25 PM

All replies

  • Followed these steps to enable FIM MA tracing:

    http://social.technet.microsoft.com/wiki/contents/articles/10227.how-to-enable-fim-service-management-agent-tracing.aspx

    Did not know you could enable tracing only on the FIM MA!

    Unfortunately, nothing here that I didn't already see in the FIMSync log.

    The only thing I have been seeing is some warnings with the following:

    The Synchronization State Machine is now in this state: InitialState.ready  Thread #4
    The Synchronization State Machine is now in this state: InitializedState.initialized  Thread #4
    The Synchronization State Machine is now in this state: OpenForFullImportState.open for full import  Thread #6

    I assume the import starts around here:

    The Synchronization State Machine is now in this state: FullImportState.full import  Thread #6
    Sync: Sync.FullImport.ExecuteReader.Enter
    Sync: Sync.FullImport.ExecuteReader.Exit
    Completed method ReadObjectsFromDataSourceFull.  Time: 11:39:24 AM.  Thread #6
    Started method CloseDataSourceForFullImport.  Time: 11:39:24 AM.  Thread #6
    Import Session: Type FullImportSession, Watermark: 1
    Completed method UpdateWatermark.  Time: 11:39:24 AM.  Thread #6
    Completed method QueryCapabilities.  Time: 11:39:24 AM.  Thread #6
    Started method QueryDnConstructionXml.  Time: 11:39:24 AM.  Thread #6
    Completed method QueryDnConstructionXml.  Time: 11:39:24 AM.  Thread #6
    The Synchronization State Machine is now in this state: InitializedState.initialized  Thread #6
    Completed method CloseDataSourceForFullImport.  Time: 11:39:24 AM.  Thread #6
    The AppDomain's parent process is exiting.
    The Synchronization State Machine is now in this state: StoppingState.stopping  Thread #2

    No errors are thrown in the trace logs here. Also, none present in the FIMService trace log.

    Have also tried re-creating the FIM MA. Same issue presents.

    Worth noting that a delta import completes successfully.

    - Ross Currie


    FIMSpecialist.com | MCTS: FIM 2010 | Now Offering ECMA1->ECMA2 Upgrade Services

    Thursday, March 21, 2013 3:47 AM
  • Gave up and reinstalled FIM Portal with a fresh database.

    And this is why we don't use the debug Stored Procedures on the FIMService database...

    - Ross Currie


    FIMSpecialist.com | MCTS: FIM 2010 | Now Offering ECMA1->ECMA2 Upgrade Services

    Monday, March 25, 2013 5:01 AM
  • Why just dont delete ERE's with powershell as well?
    Tuesday, April 9, 2013 11:46 AM
  • Because it takes too long. First you have to get PowerShell to read all the objects, then you have to wait for it to delete them one by one. When you have thousands of Orphan EREs, this can take hours.

    The [debug][DeleteOrphanedRulesByType] method would delete them in a matter of seconds. I was hoping to find something that worked as simply.

    - Ross Currie


    FIMSpecialist.com | MCTS: FIM 2010 | Now Offering ECMA1->ECMA2 Upgrade Services

    Tuesday, April 9, 2013 3:53 PM