Bug #1525
System metadata replication errors
100%
Subtasks
History
#1 Updated by Ben Leinfelder almost 13 years ago
While debugging ObjectFormat errors during replication I noticed primary key constraint violations being thrown when sysmeta was replicated. This seems to be occurring when there is a replication "update" action triggered on the document and the sysmetadata code that handles this replication request on the target Metacat just treats it as an "insert". Also observed during debugging is that this "update" replication request is arising from a setaccess call on the auto-generated sysmeta (which happens just after the initial insert of the sysmeta).
#2 Updated by Ben Leinfelder almost 13 years ago
- Status changed from New to Closed
- Assignee set to Ben Leinfelder
Replication "update" requests do not raise unique key violations for sysmeta files now.
There are still messages about "doc id cannot be 1" because it's not a true "update" in that the rev does not get incremented on a setAccess call but this is known behavior for all documents in Metacat.
Note: permissions on sysMeta are the same as those on the corresponding sciMeta document and both these rulesets are replicated successfully (this had been a question during the standup call today).
