Jira index size and disk space are not decreasing after archiving Issues or Projects or cleaning up
Platform Notice: Data Center Only - This article only applies to Atlassian products on the Data Center platform.
Note that this KB was created for the Data Center version of the product. Data Center KBs for non-Data-Center-specific features may also work for Server versions of the product, however they have not been tested. Support for Server* products ended on February 15th 2024. If you are running a Server product, you can visit the Atlassian Server end of support announcement to review your migration options.
*Except Fisheye and Crucible
Summary
After archiving Issues or Projects or cleaning up field configurations, the size of the Jira index doesn't decrease as expected, and the space allocated in the filesystem is also not deallocated as expected.
It's often observed the Index size to even increase after archiving or cleaning up Jira.
Refer to the Clean up your Jira instance article for best-practices that may improve maintainability and performance.
Diagnosis
Both Jira's internal health check on Indexing > Free Space and a disk space assessment tool show an Index size and space allocation that hasn't decreased as expected — or even increased.
Example output of du -h -s
1
2
3
$ du -h -s /var/atlassian/application-data/jira/caches/indexesV1
908M /var/atlassian/application-data/jira/caches/indexesV1
$Also, the Index may or may not be working as expected, with archived Issues no longer available on searches or gadgets across all nodes.
Cause
Jira relies on Apache Lucene for Indexing, and Lucene prioritizes Index performance and integrity over disk space deallocation.
This results in the Index indeed behaving as expected, but the disk space not being freed up eventually unless a forced reindex is issued.
Some further external material on the subject:
Solution
After archiving Issues or Projects or performing other cleanup routines in Jira, issue a Full re-index in one of the Jira nodes or a background re-index if it's a single node.
After a successful re-index, we should see the disk space being deallocated accordingly on the reindexed node and subsequently on the others as they update their own indexes with the latest Index snapshot.
Also, from the documentation on Archiving a project: Project data will be ignored in the index, and removed completely once you re-index Jira. This improves performance by removing data that is stored directly in Jira.
Was this helpful?