locked
Sync Framework sends 293 updates to the client on every sync RRS feed

  • Question

  • Here is my scenario...

    I have restored a live database on my local machine.  I de-provision it and re-provision it.  I set up the proxy from my client to point at my local WCF services.  I delete any local CE database and do a full sync.  The appropriate records are synced to the client as expected.  

    On subsequent syncs, 293 records are updated to the client every time I sync.  I can see this from the server logs...

    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:22:121,    ----- Updates for Table "InspectionHeader" -----
    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:23:287, Applied 75 of 75 rows with bulk command BulkUpdateCommand
    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:23:298, Applied 75 of 75 rows with bulk command BulkUpdateCommand
    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:23:308, Applied 75 of 75 rows with bulk command BulkUpdateCommand
    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:23:321, Applied 68 of 68 rows with bulk command BulkUpdateCommand
    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:23:326,       293 Updates Applied
    INFO   , WcfSvcHost, 10, 08/28/2013 18:08:23:327,    --- End Updates for Table "InspectionHeader" ---

    This happens every time I sync.  This happens with other scopes and tables, I chose this scope/table arbitrarily.  

    This also seems to only happen when I am syncing against my local machine.

    I've done some profiling and when sync framework does its Select Changes, it sends up a timestamp of 18354513 (0x1181151) which when it compares this to the local_update_peer_timestamp in the table, it is much lower so the row is returned in Select Changes.  Here's one of the local_update_peer_timestamps in one of the records that is returned.  (0x119563D). 

    Unless I am mistaken, I was thinking that SelectChanges would only return rows that have been updated?  Does anyone know what I am doing wrong?

    Monday, September 16, 2013 3:20 PM