Uploaded image for project: 'Jira Platform Cloud'
  1. Jira Platform Cloud
  2. JRACLOUD-66664

Migrated pre-6.0 worklogs with mixed-case usernames are disassociated from authors

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Fixed
    • Icon: High High
    • None

      With the transition in JIRA 6 to user keys in lieu of user names, the pre-6.0 issue worklog is not being migrated correctly in the case of users who had LDAP-originating non-lowercase usernames. This results in the worklog entries not being associated with the user.

      For example, in the database in JIRA 5.x, the usernames are always stored in lowercase in the database in jiraissue.reporter and jiraissue.assignee. However, in the worklog.author column, the usernames are persisted in mixed case to match the LDAP user.

      This works fine in JIRA 5.x because JIRA is performing a case-insensitive match against the username. However, JIRA 6 is once again performing a case-sensitive match, since it apparently assumes that the user keys have all been normalized to lowercase (and so that there is no namespace conflict with the new "IDxxxxx" user keys).

      Unfortunately, this assumption fails in the case of the worklog table, since the worklog.author names are still left in mixed case. This yields in a failure to resolve the username when viewing the worklog for those users: it displays just the username (user key) without the display name or the corresponding inline dialog.

              Unassigned Unassigned
              7c60ab039b09 Scott Dudley [Inactive]
              Votes:
              0 Vote for this issue
              Watchers:
              11 Start watching this issue

                Created:
                Updated:
                Resolved: