Re-indexing failed in Jira Data Center
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
Problem
Re-indexing failed on one of your nodes and the following error was displayed in Jira:
Re-indexing failed
Node node1 lost contact with the cluster while performing a re-index. The index on this node is most likely corrupted, and might affect the index on other nodes if it rejoins the cluster.
Ensure the node is turned off, or restart it, to prevent this.
Context
When you perform a full re-index in Jira Data Center, Jira deletes the index on the node on which you initiated the re-index, and rebuilds it from scratch. If you’re seeing this error message, it means that the node went down or lost contact with the cluster while re-indexing was still in progress. Until this failed re-index is accounted for, no further re-indexing tasks can be performed in the cluster.
Solution
You can fix this problem by clicking the Clean up failed tasks button below the error. This action will cancel the current re-indexing process and clean up any failed tasks, allowing you to re-index again. Once that’s done, you can safely re-index Jira using a different node.
In the meantime, you should also make sure that the failed re-indexing node actually went down.
Jira has no way of knowing if the node died or not. It can just recognize that it stopped communicating with the rest of the cluster. So it’s possible (although unlikely) that the node may be alive and attempting to re-join the cluster. You need to ensure that this doesn’t happen. The node’s index is probably corrupted, and if it’s alive and rejoins the cluster, it will try to distribute that bad index to the other nodes.
In the unlikely event that the failed re-indexing node is still alive, either shut it down or restart it. For more info on how to restart Jira, see Starting and stopping Jira applications.
Was this helpful?