Details
-
Bug
-
Status: Closed
-
Normal
-
Resolution: Fixed
-
4.1.2, 4.2.6, 4.2.7, 3.2.12
-
None
-
None
-
1
-
Turing Sprint 180, Turing Sprint 181
Description
When a cluster node has an active lock and is shutdown without releasing the lock. It's possible that the lock can't be reclaimed, whenever this server is up and running within the time that the lock should expire. When the server is up and running, the server keeps updating the expire time of the lock, while there is actually no process running.
The result is that no server is able to claim te lock, because the expire time will 'never' expire.
Attachments
Issue Links
- is cloned by
-
REPO-2006 [Backport 12.3] Lock manager unable to reclaim lock, because expire time still gets updated for 'running' jobs that no longer exist
- Closed
-
REPO-2007 [Backport 11.2] Lock manager unable to reclaim lock, because expire time still gets updated for 'running' jobs that no longer exist
- Closed
-
REPO-2008 [Backport 12.2] Lock manager unable to reclaim lock, because expire time still gets updated for 'running' jobs that no longer exist
- Closed
-
REPO-2009 [Backport 10.2] Lock manager unable to reclaim lock, because expire time still gets updated for 'running' jobs that no longer exist
- Closed