[
https://issues.apache.org/jira/browse/AMBARI-14131?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hari Sekhon updated AMBARI-14131:
---------------------------------
Description:
Ambari agent was today using 15GB of RAM. I've noticed this before and
restarting the agent brings it back down to normal around 1GB.
I've noticed this before at another client but can't reproduce it, it seems
random. In fact on larger production sized clusters this goes largely unnoticed
but on smaller test clusters it actually impacts other components and causes
outages.
There is nothing in the Ambari agent log to explain why the agent ever needs to
take 15GB of ram, but I suspect it's a coding error / memory leak etc.
was:
Ambari agent was today using 15GB of RAM. I've noticed this before and
restarting the agent brings it back down to normal around 1GB.
I've noticed this before but can't reproduce it, it seems random. In fact on
larger production sized clusters this goes largely unnoticed but on smaller
test clusters it actually impacts other components and causes outages.
There is nothing in the Ambari agent log to explain why the agent ever needs to
take 15GB of ram, but I suspect it's a coding error / memory leak etc.
> Ambari Agent using 15GB RAM (memory leak??)
> -------------------------------------------
>
> Key: AMBARI-14131
> URL: https://issues.apache.org/jira/browse/AMBARI-14131
> Project: Ambari
> Issue Type: Bug
> Components: ambari-agent
> Affects Versions: 2.1.0
> Environment: HDP 2.3
> Reporter: Hari Sekhon
> Priority: Critical
>
> Ambari agent was today using 15GB of RAM. I've noticed this before and
> restarting the agent brings it back down to normal around 1GB.
> I've noticed this before at another client but can't reproduce it, it seems
> random. In fact on larger production sized clusters this goes largely
> unnoticed but on smaller test clusters it actually impacts other components
> and causes outages.
> There is nothing in the Ambari agent log to explain why the agent ever needs
> to take 15GB of ram, but I suspect it's a coding error / memory leak etc.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
