You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@aurora.apache.org by "Bill Farner (JIRA)" <ji...@apache.org> on 2014/05/11 00:00:50 UTC

[jira] [Updated] (AURORA-406) SLA stats should not be in internal TSDB

     [ https://issues.apache.org/jira/browse/AURORA-406?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Bill Farner updated AURORA-406:
-------------------------------

    Description: 
The recently-added SLA stats are registered as regular gauges, which go into the in-memory TSDB.  Since we track SLA stats per job, this causes too much heap consumption for large job counts.

This can be fixed trivially by exporting stats through StatsProvider#untracked().  This will cause values to show on /vars and /vars.json, but not be tracked in the TSDB.

  was:The recently-added SLA stats are registered as regular gauges, which go into the in-memory TSDB.  Since we track SLA stats per job, this causes too much heap consumption for large job counts.


> SLA stats should not be in internal TSDB
> ----------------------------------------
>
>                 Key: AURORA-406
>                 URL: https://issues.apache.org/jira/browse/AURORA-406
>             Project: Aurora
>          Issue Type: Story
>          Components: Scheduler
>            Reporter: Bill Farner
>            Assignee: Maxim Khutornenko
>            Priority: Critical
>
> The recently-added SLA stats are registered as regular gauges, which go into the in-memory TSDB.  Since we track SLA stats per job, this causes too much heap consumption for large job counts.
> This can be fixed trivially by exporting stats through StatsProvider#untracked().  This will cause values to show on /vars and /vars.json, but not be tracked in the TSDB.



--
This message was sent by Atlassian JIRA
(v6.2#6252)