Environment:
- Windchill 11.0 M020
- Windows 2012 master
- Windows 2008 replica server
I have a recurring problem replicating very large files to one specific replica server.
When I 'Reset Undelivered Replication Items' to that site, and kick off a replication job, the job usually processes a few small files successfully, before it starts transmitting a single large file (usually over 2.5 GB). This large file will begin to replicate and make good progress for the first couple hours, but eventually all replication activity to that site stops. Sometimes the file being transferred will be moved to the current vault folder, and then into the .unreferenced folder within. I know this based on monitoring the file activity on the replica server and watching the wrmfTransport folder on the replica.
Here is evidence of the same file being pushed to the replica, and winding up '.unreferenced'
When this happens, the Replication job continues to say 'Executing' and the wt.router.7.x queue continues to have 1 executing item.
As of now, the only 'fix' I have is to cancel the replication job, restart ALL of the wt.router queues, and create a new replication job. However, this job too will eventually 'time out'.
I'm thinking that the wt.router queue is timing out, but haven't been successful in making changes to the queue parameters.
Does anyone else have similar issues?
I have a PTC case open, but of course, no response for a very long time.
