Issue Summary

      Performing development related activities such as creating commits, branches, pull requests will cause the related issue to disappear from the active sprint board.

      Please note that this behavior cannot be reproduced reliably and the steps to reproduce will need to be repeated several times. 

      As of 19.11.20 the bug is reproducible every time.

      Steps to Reproduce

      1. Start a sprint
      2. Create a new issue
      3. Add the issue to active sprint by editing sprint field on the issue
      4. Transition Issue
      5. Create a branch for issue via Jira
      6. Create a Pull request
      7. Transition Issue

      Steps to Reproduce 2

      1. Start a sprint
      2. Create issue
      3. Add issue to active sprint by editing sprint field on issue
      4. Transition issue
      5. Commit a new message with the issue key
      6. Check the sprint board

      Actual steps to reproduce

      1. Create a sprint
      2. Create an issue
      3. Commit a new message with the issue key on EVERY devstatus.applink:thread (either make sure with additional logging, or just do it multiple times with some time space between pushes). You have to make sure that the same thread will perform a commit before and after assigning the issue to a sprint.
      4. Assign the issue to the sprint
      5. Commit yet another message with the same issue key
      6. Check the sprint board

      Expected Results

      The issues should be visible on the active sprint board

      Actual Results

      Check the Sprint board - the issue disappears from the board

      1. In this example, we can see issue AH-3 is added to the Sprint field for AH Sprint 1 but we cannot see the issue on the active sprint board:

      2. Perform a project re-index and issue re-appears

      It is very hard to reproduce the issue reliably. After countless commits, we manage to reproduce the problem again ( AH-4 this time )

      As of 19.11.20 the issue can be reproduced reliably every time, as seen on the following screenshot: 

      Checking this SQL output, the database returns correctly which confirms the problem with the indexes :

      select cfv.issue as issueid, cfv.stringvalue as SprintNumber, ao."NAME" as SprintName, cf.cfname, p.pkey || '-' || jissue.IssueNum as IssueKey, cf.id As SprintCfid
      from
      CustomFieldValue cfv,
      CustomField cf,
      jiraissue jissue,
      project p,
      "AO_60DB71_SPRINT" ao
      where
      cf.cfname='Sprint'and
      cf.Id = cfv.CustomField and
      jissue.id = cfv.issue and
      jissue.project = p.Id and
      cf.cfname='Sprint' and
      ao."NAME" like '%<SPRINT_NAME>%'
      

      Workaround

      Use the reindex rest API as a work around for the issues that are impacted: https://docs.atlassian.com/software/jira/docs/api/REST/8.9.0/#api/2/reindex-reindexIssues

            [JRASERVER-71091] Development Activity breaks Sprint indexes

            Hello, a customer of ours uses on prem Jira 8.13.9.

            They observed that when one user, who is the admin of their instance, executes a "commit" on GitLab that the corresponding story vanishes from the board.

            The workaround for us is to start a reindex every night. However, this does not help with changes during the day. It merely mitigates long term problems but does not solve it. 

            According to this ticket here the problem is solved with Jira 8.5.5 / 8.6.0. Obviously this contradicts our observations. It still is an issue. 

            linkyard Team Account added a comment - Hello, a customer of ours uses on prem Jira 8.13.9. They observed that when one user, who is the admin of their instance, executes a "commit" on GitLab that the corresponding story vanishes from the board. The workaround for us is to start a reindex every night. However, this does not help with changes during the day. It merely mitigates long term problems but does not solve it.  According to this ticket here the problem is solved with Jira 8.5.5 / 8.6.0. Obviously this contradicts our observations. It still is an issue. 

            migrated from cloud to on-prem 8.17.0. having same issue sprints are empty all the issues are showing in backlog. even manual movement of the issues to the sprint doesn't work. re-indexing doesn't work .

            ahmad faraz jafri added a comment - migrated from cloud to on-prem 8.17.0. having same issue sprints are empty all the issues are showing in backlog. even manual movement of the issues to the sprint doesn't work. re-indexing doesn't work .

            Up voting as well. This happens every couple weeks and is very confusing for the team.

            James Kleist added a comment - Up voting as well. This happens every couple weeks and is very confusing for the team.

            Voting up this bug fix - very frustrating for our development and project management team.

            Library & Archives NT added a comment - Voting up this bug fix - very frustrating for our development and project management team.

            I'm seeing this more and more.

            Certain tickets won't stay on the active sprint board.

            I have to regularly re-index to get tickets back onto the active sprint board.

            Terry Child added a comment - I'm seeing this more and more. Certain tickets won't stay on the active sprint board. I have to regularly re-index to get tickets back onto the active sprint board.

            This affects my team as well. what should be an easy look at the board to find the ticket turns into having to search our backlog for the appropriate one.

            I ended up creating a separate project to store our backlog items due to this bug it's very tedious

            Neal Ruggles added a comment - This affects my team as well. what should be an easy look at the board to find the ticket turns into having to search our backlog for the appropriate one. I ended up creating a separate project to store our backlog items due to this bug it's very tedious

              Unassigned Unassigned
              astephen@atlassian.com Adrian Stephen
              Affected customers:
              43 This affects my team
              Watchers:
              34 Start watching this issue

                Created:
                Updated:
                Resolved: