[
https://issues.apache.org/jira/browse/HADOOP-16705?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16973448#comment-16973448
]
Lukas Majercak edited comment on HADOOP-16705 at 11/13/19 3:38 PM:
-------------------------------------------------------------------
Attached is a screenshot from a 5minute java flight recording (HDFS DataNode
process). It shows that during these 5 minutes, more than 45GB of memory was
created from
{code:java}
MbeanInfoBuilder.get()
{code}
was (Author: lukmajercak):
Attached is a screenshot from a 5minute java flight recording. It shows that
during these 5 minutes, more than 45GB of memory was created from
{code:java}
MbeanInfoBuilder.get()
{code}
> MBeanInfoBuilder puts unnecessary memory pressure on the system with a debug
> log
> --------------------------------------------------------------------------------
>
> Key: HADOOP-16705
> URL: https://issues.apache.org/jira/browse/HADOOP-16705
> Project: Hadoop Common
> Issue Type: Improvement
> Components: metrics
> Affects Versions: 2.9.2
> Reporter: Lukas Majercak
> Assignee: Lukas Majercak
> Priority: Major
> Attachments: mbeaninfobuilder.JPG
>
>
> MBeanInfoBuilder's get() method DEBUG logs all the MBeanAttributeInfo
> attributes that it gathered. This can have a high memory churn that can be
> easily avoided.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
