You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@aurora.apache.org by "Maxim Khutornenko (JIRA)" <ji...@apache.org> on 2014/09/23 23:55:34 UTC

[jira] [Created] (AURORA-743) Figure out retention policy for scheduler job update data

Maxim Khutornenko created AURORA-743:
----------------------------------------

             Summary: Figure out retention policy for scheduler job update data
                 Key: AURORA-743
                 URL: https://issues.apache.org/jira/browse/AURORA-743
             Project: Aurora
          Issue Type: Story
          Components: Scheduler
            Reporter: Maxim Khutornenko


We need to come up with job update history retention policy sooner than later to avoid tipping over the snapshot creation timeout (large heap data == longer GC times == failover). Perhaps going after the Task retention approach (N items per job) is the easiest until we have a durable DB storage/backups and can afford longer history (e.g. time-based purging or infinite archive).



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)