Jira 9.12.11 full re-index fails

[Ekaterina Berezutskaya]

I'm trying to run the full re-index with Jira 9.12.11 but I'm always getting the following error:

The logs are only showing some errors: 

025-02-12 13:13:42,886+0300 JiraTaskExecutionThread-1 ERROR evberez     [c.a.jira.task.TaskManagerImpl] Task 'Jira Indexing' failed.

com.atlassian.jira.index.IndexingFailureException: Indexing completed with 1 errors

    at com.atlassian.jira.util.index.CompositeIndexLifecycleManager.reIndexAll(CompositeIndexLifecycleManager.java:88)

    at com.atlassian.jira.util.index.CompositeIndexLifecycleManager.reIndexAll(CompositeIndexLifecycleManager.java:51)

Also, there is "NonUniqueResultException" in the logs:

2025-02-12 11:43:45,403+0300 JiraTaskExecutionThread-1 WARN xxx_user     [c.a.jira.index.AccumulatingResultBuilder] java.util.concurrent.ExecutionException: com.querydsl.core.NonUniqueResultException: Only one result is allowed for fetchOne calls

java.lang.RuntimeException: java.util.concurrent.ExecutionException: com.querydsl.core.NonUniqueResultException: Only one result is allowed for fetchOne calls

    at com.atlassian.jira.index.FutureResult.await(FutureResult.java:29)

    at com.atlassian.jira.index.AccumulatingResultBuilder.collectResult(AccumulatingResultBuilder.java:93)

......

I tried all solutions from this article 

https://confluence.atlassian.com/jirakb/troubleshoot-a-reindex-failure-in-jira-server-429917142.html

but nothing helps.

Has anyone had the same problem?

[John Funk]

Hi Ekaterina - did you ever get this solved? If so, what was the final solution?

[Ekaterina Berezutskaya]

Hi John. Yes, I did.  Finally, I used Jira's index analyzer that helped me to find 5 broken issues. I could not delete these issues. So I  was forced to delete records for 5 issues in the database (jiraissue, changeitem, changegroup, customfieldvalue, issuelink, os_currentstep tables). After this, the reindex was successful.

[John Funk]

Great! glad you found that solution. Why don't you post that as a separate answer so we can approve it and close this post out for future readers.  :-)

[Matt Doar _Adaptavist_]

I would dig into step 5 about a plugin. In a test instance, disable all customer installed plugins and run a reindex. See if it fails. If it does, then there is corrupted data in the database. If it works, then re-enable the plugins one at a time and run a reindex after each one is re-enabled. This is very tedious, but is one way to narrow down the problematic plugin.

If the error is in the DB, then look at increasing the log level around reindexing to see if you can work out which issue (or DB table and field) is the problem. I think I'd follow up with Atlassian support after that.

[Ekaterina Berezutskaya]

Thank you. I disabled all customer plugins but a reindex failed again. Now I'am trying to find DB error...