Background – user claimed that Pro/E is crashing when opening object (drw, prt, asm). In short, we proved that this is not Pro/E WF4 m140 issue but during transport (synchronization, replication) between Master site and replicas something really bad happened with Windchill 9.1 m50. Went thru long process of identifying hashed names of the files “synchronized” on replica and master and the conclusion was that even files were exactly the same size when we did binary comparison they were different!!! PTC recommended to change property on all nodes including RFS from wt.wrmf.transport.httpResumption.totalAttempts=5 to wt.wrmf.transport.httpResumption.totalAttempts=1 which per my understanding will retry the whole object to synch or replicate rather than resume if transfer was disrupted by network. We have WAN accelerators in place and no other changes besides WAN accelerator upgrade were happening recently.
PTC is helping with identifying what objects got corrupted. Still not sure what is the extent of the damage (10, 100,10K objects?) but engineers are reporting crashing Pro/E on some objects checked out from Windchill. On the positive side our CAD expert believes that there are good copies of the corrupted files on replicas. When engineer preferred vault = master and corruption occurs engineer switches preferred vault to replica (we have 9 but 1 is suspected as of now) and is able to get good file.
I wanted to share on this forum and find out if someone had similar horrific experience in the past. Of course any hint will be aBIG help.
TIA,
Artur
