Uploaded image for project: 'Jira Data Center'
  1. Jira Data Center
  2. JRASERVER-69826

Add null constraint for issuenum column to jiraissue table

XMLWordPrintable

    • 5
    • We collect Jira feedback from various sources, and we evaluate what we've collected when planning our product roadmap. To understand how this piece of feedback will be reviewed, see our Implementation of New Features Policy.

      Problem Definition

      In some cases due to JSWSERVER-16473 Jira ends up with a NULL value in the issuenum column of jiraissue table,

      Current schema:

                        Table "public.jiraissue"
        Column        |           Type           | Modifiers
----------------------+--------------------------+-----------
 id                   | numeric(18,0)            | not null
 pkey                 | character varying(255)   |
 issuenum             | numeric(18,0)            |
 project              | numeric(18,0)            |
...
Indexes:
    "pk_jiraissue" PRIMARY KEY, btree (id)
    "issue_assignee" btree (assignee)
    "issue_created" btree (created)
    "issue_duedate" btree (duedate)
    "issue_proj_num" btree (issuenum, project)
    "issue_proj_status" btree (project, issuestatus)
    "issue_reporter" btree (reporter)
    "issue_resolutiondate" btree (resolutiondate)
    "issue_updated" btree (updated)
    "issue_votes" btree (votes)
    "issue_watches" btree (watches)
    "issue_workflow" btree (workflow_id)

      Suggested Solution

      Prevent the data consistency problem by adding not null constraint for issuenum column for jiraissue table.

      Workaround

      None

            [JRASERVER-69826] Add null constraint for issuenum column to jiraissue table

            Gisela Lassahn added a comment -

            In my opinion it is a bug that Jira produces such nearly empty lines in jiraissue table and should be treated with a high priority as an critical bug because it can happen every day and keep administrators from reindexing.
            The workaround mentioned in JSWSERVER-16473 (to delete the corrupt lines from jiraissue table) is no real solution.

            We always avoid manual database manipulation because they are “open-heart operations” where you never know whether there are dependent records in other tables and you can get inconsistencies.

            So why is JSWSERVER-16473 solved by preventing “Jira from trying to index such corrupted issue entries” instead of solving the real bug (additionally!) which is obviously already known for at least 2,5 years?

            JRASERVER-69826 is a „suggestion“ to help to find the cause for the bug and has been created more than 2 years ago and still is in status “Gathering interest” which is totally unintelligible in my eyes.

            What about the “Atlassian Data Center and Server Bug Fix Policy”?:
            “We'll generally fix critical bugs in the next maintenance release”

            Gisela Lassahn added a comment - In my opinion it is a bug that Jira produces such nearly empty lines in jiraissue table and should be treated with a high priority as an critical bug because it can happen every day and keep administrators from reindexing. The workaround mentioned in  JSWSERVER-16473  (to delete the corrupt lines from jiraissue table) is no real solution. We always avoid manual database manipulation because they are “open-heart operations” where you never know whether there are dependent records in other tables and you can get inconsistencies. So why is  JSWSERVER-16473  solved by preventing “Jira from trying to index such corrupted issue entries” instead of solving the real bug (additionally!) which is obviously already known for at least 2,5 years? JRASERVER-69826  is a „suggestion“ to help to find the cause for the bug and has been created more than 2 years ago and still is in status “Gathering interest” which is totally unintelligible in my eyes. What about the “Atlassian Data Center and Server Bug Fix Policy”?: “We'll generally fix critical bugs in the next maintenance release”

            Andriy Yakovlev [Atlassian] added a comment -

            mdoar2
            since inserting NULL will fail, it will help find places where this happening

            Andriy Yakovlev [Atlassian] added a comment - mdoar2 since inserting NULL will fail, it will help find places where this happening

            Matt Doar added a comment -

            This is a good idea to stop the data corruption from JSWSERVER-16473 spreading, but what about fixing the underlying cause of this corruption?

            Matt Doar added a comment - This is a good idea to stop the data corruption from JSWSERVER-16473 spreading, but what about fixing the underlying cause of this corruption?

              Unassigned Unassigned
              ayakovlev@atlassian.com Andriy Yakovlev [Atlassian]
              Votes:
              15 Vote for this issue
              Watchers:
              17 Start watching this issue

                Created:
                Updated: