none
FIM Reporting Job fails with timeout RRS feed

  • Question

  • Hi,

    We just migrated a FIM to MIM instance.

    The initial MIM reporting jobs ran fine, but we brought over the production data and are now getting a timeout error on the incremental reporting job:

    "This reporting job has been cancelled because the FIM Service instance handling the job has failed to respond within the pre-configured timeout window."

    Anyone know how to find out which FIM Service instance is handling the job? And how to extend the timeout window?

    Or any other ideas?

    Thank you for any help!

    Sunday, December 4, 2016 4:38 PM

Answers

  • SamiVV

    If this is a new setup I would look at the timeout value we can configure on the fim service configuration below. Factors like network, VM backup, Service Manager performance due to low memory resources, or I/O on the SQL server, or the Service Manger server itself having insufficient resources to complete all operations in a timely manner

     reportingDataReadTimeoutInSeconds               300                                         Reporting operations that read from the FIM Service
                                                                                                        are tied to this timeout to allow these operations extra
                                                                                                        time to complete if there are a large number of
                                                                                                        objects / values in a batch.                                                                                                   
                                                                                                        Range: 0-max(int16)       

      reportingDataWriteTimeoutInSeconds              300                                         Reporting operations that write to the FIM Service
                                                                                                        are tied to this timeout to allow these operations extra
                                                                                                        time to complete if there are a large number of
                                                                                                        objects / values in a batch.                                                                                                   
                                                                                                        Range: 0-max(int16)       

    Also I would Increase the frequency in which the "FIM_ScheduleReportingIncrementalSynchronizationJob"  job runs. In effect, the more frequent the jobs runs, the less data has to send at each interval.

    • Marked as answer by SamiVV Tuesday, December 6, 2016 11:37 AM
    Monday, December 5, 2016 6:33 PM
    Moderator

All replies

  • SamiVV

    If this is a new setup I would look at the timeout value we can configure on the fim service configuration below. Factors like network, VM backup, Service Manager performance due to low memory resources, or I/O on the SQL server, or the Service Manger server itself having insufficient resources to complete all operations in a timely manner

     reportingDataReadTimeoutInSeconds               300                                         Reporting operations that read from the FIM Service
                                                                                                        are tied to this timeout to allow these operations extra
                                                                                                        time to complete if there are a large number of
                                                                                                        objects / values in a batch.                                                                                                   
                                                                                                        Range: 0-max(int16)       

      reportingDataWriteTimeoutInSeconds              300                                         Reporting operations that write to the FIM Service
                                                                                                        are tied to this timeout to allow these operations extra
                                                                                                        time to complete if there are a large number of
                                                                                                        objects / values in a batch.                                                                                                   
                                                                                                        Range: 0-max(int16)       

    Also I would Increase the frequency in which the "FIM_ScheduleReportingIncrementalSynchronizationJob"  job runs. In effect, the more frequent the jobs runs, the less data has to send at each interval.

    • Marked as answer by SamiVV Tuesday, December 6, 2016 11:37 AM
    Monday, December 5, 2016 6:33 PM
    Moderator
  • Thank you! I made some changes to the SQL job to run every 4 hours instead of every 8.

    The timeout issue seemed to resolve itself when I restarted the FIM Service on one of the load balanced Service and Portal servers. (It was responding to the portal, but not the ResourceManagementService/MEX url.)

    I appreciate your help.

    Tuesday, December 6, 2016 11:39 AM