Hello,

Thank you for submitting this suggestion. We appreciate you taking the time to share your ideas for improving our products, as many features and functions come from valued customers such as yourself.

Atlassian is committed to enhancing the security and compliance of our Data Center products, with an emphasis on sustainable scalability and improving the product experience for both administrators and end-users. We periodically review older suggestions to ensure we're focusing on the most relevant feedback. This suggestion is being closed due to a lack of engagement in the last four years, including no new watchers, votes, or comments. This inactivity suggests a low impact. Therefore, this suggestion is not in consideration for our future roadmap.

Please note the comments on this thread are not being monitored.

You can read more about our approach to highly voted suggestions here and how we prioritize what to implement here.

To learn more about our recent investments in Bitbucket Data Center, please check our public roadmap and our dashboards, which contain recently resolved issues.

Kind regards,
Bitbucket Data Center

Would the following fix this issue ?

1. Create a backup before continuing.
2. Shut down Bitbucket Server.
3. Remove the contents of the <BITBUCKET_HOME>/shared/data/snapshots directory.
4. Restart Bitbucket Server.
5. Push to the repository with the JIRA issue key in the commit message to force a reindex of the data.

Source: Reindex JIRA issue keys

Problem still exists in 2.4.2. I think it's more than minor because there is no work around to cause previous commit messages to be rescanned. Deleting the idx-snapshot doesn't do it apparently. The other workaround at https://confluence.atlassian.com/display/STASHKB/Git+Commits+Are+Not+Linking+to+JIRA+Tickets+Due+to+Corrupted+Indexes (which loses pull request info) is very slow for more than a dozen repos.

Please provide a way to reindex each repo individually, and also a global repo reindex

Short-term we should allow users to customize the JIRA issue key pattern through STASH-HOME/stash-config.properties as well. This would make upgrades easier; no need to modify the setenv.sh.

A partial workaround is to delete the files in STASH_HOME/caches/idx-snapshots/<repo-id>. This will force the indexer to start indexing from the first commit again. Unfortunately, this is not the same as a real 're-index' because previously indexed JIRA keys are not removed.

You can retrieve the repository ID through the REST api: https://stash.yourdomain.com/rest/api/latest/projects/<project-key>/repos/<repo-slug>

Long-term we should:

• retrieve the regex pattern(s) from connected JIRA servers. When we get a pattern that is not the default pattern, we should re-index all(?) repositories or add a banner to the admin screens.
• add the option to re-index a repository from the Repository Settings and REST.

Problem still exists in 2.1.2.

We forgot to change the setenv script after upgrading stash.
We made some commits before noticing the issue.

After changing the setenv script the new pattern is used; new commits are linked to the jira issues.
Old commits are still not linked.

Is there a workaround to force relinking of old commits?

mheemskerk, is this related to the work that you've done recently on indexing the changeset? Does it fix this issue or do we require re-indexing for this to work.