Uploaded image for project: 'ZABBIX BUGS AND ISSUES'
  1. ZABBIX BUGS AND ISSUES
  2. ZBX-9149

When monitoring intervals are same or almost the same, there is a possibility that wrong evaluation is done.

XMLWordPrintable

    • Icon: Incident report Incident report
    • Resolution: Workaround proposed
    • Icon: Critical Critical
    • None
    • 2.0.12
    • Server (S)

      item1
      key_ :proc.num[ntpd]
      interval:30

      item2:
      key_ :vfs.file.exist[/tmp/maintenance]
      interval:30

      trigger
      Expression:{TEST_SERVER:proc.num[ntpd].count(#2,1,lt)}=2 & {TEST_SERVER:vfs.file.exist[/tmp/maintenance].last(0)}=0
      Event generation:Normal

      When doing setting like the above, an event is notified twice, and there is a possibility that wrong evaluation is done.
      When the same trigger is estimated by HistorySyncer or Timer twice at the same time, this problem should happen.

      The export file made for a test is attached.
      If the problem is not reproduced, please copy the host. It should be easier to reproduce.

        1. zbx_export_hosts.xml
          715 kB
        2. event.jpg
          event.jpg
          219 kB
        3. event_detail.jpg
          event_detail.jpg
          208 kB

            Unassigned Unassigned
            zinten Yoshinori Komuro
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated:
              Resolved: