Hi everyone,
on my openBIS instance, I use a segmented datastore consisting of 3 shares. There is a daily maintenance task using the simple store shuffler to drain share 1 and move all present datasets to shares 2 and 3. The maintenance task is using the default configuration as listed in the docs ( Installation and Administrators Guide of the openBIS Data Store Server - openBIS Documentation Rel. 20.10 - Confluence ).
Currently I am facing the problem, that there are quite a few “locked” datasets:
INFO [store-shuffler - Maintenance Plugin] OPERATION.SimpleShuffling - Await for data set 20251209132851349-1547884 to be unlocked.
Once the maintenance task tries to move a locked dataset, the process hangs, since the timeout to wait for a dataset to be unlocked seems to be 84600s:
ERROR [store-shuffler - Maintenance Plugin] OPERATION.ShareIdManager - Timeout: Lock for data set 20251209132851349-1547884 is held by threads 'Data Set Command Execution' for 84600 seconds.
ERROR [store-shuffler - Maintenance Plugin] OPERATION.EagerShufflingTask - Couldn't move data set 20251209132851349-1547884 to share 2.
ch.systemsx.cisd.common.exceptions.EnvironmentFailureException: Lock for data set 20251209132851349-1547884 hasn't been released after time out of 84600 seconds.
Is there any way to reduce this timeout, so the process won’t wait for such a long time and instead try the next dataset?
Also, has anyone experience with why datasets are being locked for such a long time?
Thanks a lot and best regards!
Mirko