Uploaded image for project: 'Confluence Data Center'
  1. Confluence Data Center
  2. CONFSERVER-59325

LDAP auto-synchronisation stops after a Space Import

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Medium Medium
    • None
    • 6.10.1, 6.6.16, 6.13.7, 6.13.9, 6.15.9, 7.4.3, 7.4.8, 7.4.9, 7.4.11, 7.4.17
    • User - Management

      Issue Summary

      LDAP directories defined in Confluence Administration » User Directories stops auto-syncing on the synchronisation interval.

      Environment

      • Confluence connected to an LDAP directory with a sync interval of 2 minutes

      Steps to Reproduce / Diagnose:

      1. Install Confluence
      2. Connect Confluence to an LDAP directory with a sync interval of 2 minutes
      3. Enable the following DEBUG
        com.atlassian.confluence.importexport
        com.atlassian.scheduler.caesium.impl.CaesiumSchedulerService
        
      4. Watching atlassian-confluence.log, it will sync with LDAP every 2 minutes
      5. Drop the XML Space import zip into the Restore Directory
      6. Through Confluence Administration, perform the Space Import and immediately close the Administration page as soon as you press Restore

      (this is just a sample reproducible scenario)

      Expected Results

      1. We see in the logs that Confluence does unschedule the LDAP job but once the Space import is complete, it should re-enable the LDAP scheduled job

      Actual Results

      1. We see this in the log when the LDAP job is unscheduled
        2019-11-20 12:30:43,929 DEBUG [Long running task: Importing data] [scheduler.caesium.impl.CaesiumSchedulerService] unscheduleJob unscheduleJob: com.atlassian.crowd.manager.directory.monitor.poller.DirectoryPollerManager.1015809
        
      2. It never gets added back in by Confluence even and hence, the LDAP job no longer syncs on the configured sync interval

      Workaround

      • Restart Confluence to reset the LDAP job

      If Confluence is running as Data Center, shutdown all nodes, then start up each node, one at a time

              Unassigned Unassigned
              hlam@atlassian.com Eric Lam
              Votes:
              32 Vote for this issue
              Watchers:
              38 Start watching this issue

                Created:
                Updated: