Constant hourglass on some collections. Colleval.log errors


  • Getting a SQL stored proceedure error in the Colleval.log.  Seems to be stuck updating collections.  We are getting a constant hourglass on a collection.  Any ideas? 

    PF: [Auxiliary Evaluator] starts to evaluate collection [XXX00029]

    PF: Refreshing collection XXX00029

    *** exec dbo.sp_TransferMembership 'XXX00029', 0

    *** [23000][547][Microsoft][SQL Server Native Client 11.0][SQL Server]The INSERT statement conflicted with the CHECK constraint "ClientOfferStatus_ItemKey_Partition_CK". The conflict occurred in database "CM_MKE", table "dbo.ClientOfferStatus", column 'ItemKey'. : CollectionMembers_ins_upd_del

    CCollectionSource_SQL::RefreshResults - could not execute SQL cmd exec dbo.sp_TransferMembership 'XXX00029', 0

    STATMSG: ID=601 SEV=E LEV=M SOURCE="SMS Server" COMP="SMS_COLLECTION_EVALUATOR_AUXILIARY_EVALUATOR" SITE=MKE PID=4648 TID=3612 GMTDATE=Tue Feb 25 14:08:42.671 2014 ISTR0="exec dbo.sp_TransferMembership 'XXX00029', 0" ISTR1="" ISTR2="" ISTR3="" ISTR4="" ISTR5="" ISTR6="" ISTR7="" ISTR8="" ISTR9="" NUMATTRS=0

    [Auxiliary Evaluator] Error refreshing collection, will retry momentarily [XXX00029, Error 0xb]


    Tuesday, February 25, 2014 2:29 PM


All replies

  • Editing the database manually is not supported at all so you might consider calling Microsoft support (CSS) for help.
    Is the error just happening for collection XXX00029? If so: just delete and recreate it.

    Torsten Meringer |

    Tuesday, February 25, 2014 6:18 PM
  • Hello

    Try to check XXX00029 collection for Unknown Computers in it and remove them. This should make collection updates run smooth.

    Friday, July 11, 2014 8:31 AM
  • Sorry to hear you are having this issue, it is extremely frustrating to have this happen.  I've seen these symptoms twice before, and in our case, we were experiencing SQL blocking.  There was a procedure running that was locking those given tables and collections were not updating.  Fortunately, we have a wonderful DBA at this client who was able to quickly come up with that hypothesis and traced the blocking process. 

    We ultimately decided to forcibly end that task on the DB side (be VERY careful if you choose to do this).  In our first instance, the task locking everything was the Client Operations Summarizer.  In the 2nd instance, the Rebuild indexes was still running but not actually doing anything.  In that situation we opted to stop SMSExec (this can take quite a while, be patient).  Once it and the other associate SCCM component services were stopped, we watched the database and the blocking sessions disappeared excited for the index rebuild.  We killed index process and restarted SCCM and everything came back.

    I don't know if this is what is happening, but it could be worth investigating.  Here is my write-up on the first instance from a few months ago.  Good luck!

    Friday, July 11, 2014 7:07 PM