Search Engine: Elastic

Article ID: 131911, created on Dec 14, 2017, last review on Dec 14, 2017

  • Applies to:
  • Operations Automation 7.0
  • Operations Automation 7.1
  • Operations Automation 7.2
  • Operations Automation 7.3

Symptoms

On OA 7.x Update named or Update PowerDNS tasks execution time gradually increases with time. The effect is more visible, when many DNS zones are synchronized at the same time.

/var/log/pa/core.log shows a large amount of such entries at the time of the task execution:

Dec 11 13:25:36.267 : DBG [task:58394544:744530 RequestProcessor-162838 pau 742366617]: c.p.p.tracer entry: com.parallels.pa.service.notifications.ejb.OnScreenNotificationManagerBean.getLastMessagesByContext(null, UPDATE_DNS_ZONE:197881*, inProgress)
Dec 11 13:25:36.267 : DBG [task:58394544:744530 RequestProcessor-162838 pau 742366617]: c.p.p.s.a.e.UserManagerBean locating user by UUID null
Dec 11 13:25:36.269 : DBG [task:58394544:744530 RequestProcessor-162838 pau 742366617]: c.p.p.s.a.e.UserManagerBean user with UUID null not found
Dec 11 13:25:37.432 : DBG [task:58394543:744509 RequestProcessor-162839 pau 742363379]: c.p.p.tracer exit: com.parallels.pa.service.notifications.ejb.OnScreenNotificationManagerBean.getLastMessagesByContext returning []

Cause

Since OA 7.0, the system keeps records for CCP v2 on-screen notifications that are checked on each DNS update task. The amount of records grows with time and does not get rotated, decreasing the speed of DNS synchronization.

The issue is planned to be improved in scope of requests PFR-1223 and PFR-1224.

Resolution

Contact Odin Technical Support to optimize the system performance.

dd79f16c76b9dca75205ab5e9f5465ea caea8340e2d186a540518d08602aa065 e12cea1d47a3125d335d68e6d4e15e07 5356b422f65bdad1c3e9edca5d74a1ae 1941880841f714e458ae4dc3d9f3062d 0871c0b47b3b86ae3b1af4c2942cd0ce 31987597efff5a3a9ce779cc203bbe5e 8c199f0ee4305da1a577740620df4a51

Email subscription for changes to this article
Save as PDF