[
https://issues.apache.org/jira/browse/FLINK-10317?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16830265#comment-16830265
]
Nico Kruber commented on FLINK-10317:
-------------------------------------
I'm actually unsure whether we should set a default now,
It seems I was mistaken by thinking that GC is only triggered if
{{MaxMetaspaceSize}} is set, but it seems that this is not true and indeed
first GC starts with {{MetaspaceSize}} and then depending on the GC itself.
Therefore, setting {{MaxMetaspaceSize}} would ease troubleshooting by failing
earlier in case of class loader leaks but wouldn't solve anything on its own.
> Configure Metaspace size by default
> -----------------------------------
>
> Key: FLINK-10317
> URL: https://issues.apache.org/jira/browse/FLINK-10317
> Project: Flink
> Issue Type: Bug
> Components: Deployment / Scripts
> Affects Versions: 1.5.3, 1.6.0, 1.7.0
> Reporter: Stephan Ewen
> Priority: Major
> Fix For: 1.7.3, 1.6.5
>
> Attachments: Screen Shot 2019-01-07 at 10.08.39.png, Screen Shot
> 2019-01-07 at 10.08.47.png, Screenshot 2018-12-18 at 12.14.11.png
>
>
> We should set the size of the JVM Metaspace to a sane default, like
> {{-XX:MaxMetaspaceSize=256m}}.
> If not set, the JVM offheap memory will grow indefinitely with repeated
> classloading and Jitting, eventually exceeding allowed memory on docker/yarn
> or similar setups.
> It is hard to come up with a good default, however, I believe the error
> messages one gets when metaspace is too small are easy to understand (and
> easy to take action), while it is very hard to figure out why the memory
> footprint keeps growing steadily and infinitely.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
