Stopped-Ma error during import RRS feed

  • Question

  • Hi,

    I'm seeing an odd error when I do an import on my MA. 

    I have written an ECMA 2.0 extension, and I use the portal to create MPR/Workflow/Sync Rules to use it. Recently, I made a change to the schema for this ECMA extension and refreshed the MA.

    After that, exports work ok but when I do an Import (either delta or full), I get a "stopped-ma" Status. I have refreshed the schema on the MA and it found some updates, I have then done a sync as well but no luck, I keep getting that error during imports.

    I even created a new MA and attached it to my ECMA extension dll but I still get the same error when I initiate a full import. My code has exception handling everywhere, and I also create log files during the run, but I can't see any errors being generated from within the code.

    Any idea what this "stopped-ma" status is, and how can I potentially fix this?


    • Edited by kmittal82 Wednesday, July 3, 2013 11:36 AM
    Wednesday, July 3, 2013 11:19 AM

All replies

  • Found it, for anyone else who gets this error, check the PageSize you have set in your ECMA extension DLL. I followed the MS example and set the import size to 12, and I was importing 16 objects, hence the error (shown in the server logs)
    • Proposed as answer by Dan_Walters Friday, November 29, 2013 4:52 AM
    Wednesday, July 3, 2013 12:59 PM
  • I'm suffering the same issue. If I read your message you are implying that if the import size is set to 12 that you can never have more than 12 objects? if I'm not mistaken that is no the definition for paging...

    Or are we supposed to do our import method in a way that it uses that parameter? I've looked on the web but all examples just use this default...

    Any hints would be nice. The stopped-ma is a PITA. All regular logging shows no exceptions whatsoever. I got 18 objects returned but they dont show in the FIM sync run results and I get a stopped-ma

    Thursday, November 28, 2013 8:59 AM
  • After some more research I've found out that if you don't code anything specific the page size is indeed the maximum number of objects for a full import.

    changing the "default" in the code does not help in any way. You have the modify the value on the run profile if you want to bypass or temporary work around this by cranking up the page size.

    Obviously implementing paging is the only real good solution...

    Friday, November 29, 2013 7:16 AM