[ 
https://issues.apache.org/jira/browse/HBASE-19320?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16736706#comment-16736706
 ] 

wengliemiao commented on HBASE-19320:
-------------------------------------

Thank you [~stack] for the review. For the users whom are suffering by the 
similar hbase leak problem, except the solutions mentioned by the [~huaxiang], 
I have the other suggestions which uses the hbase client's attribute:
 * Config _hbase.rpc.client.impl_ to use _AsyncRpcClient_, which uses netty's 
pattern
 * Use _Connection#getTable(tableName, pool)_ method to custom batch pool, and 
limit the data size per write
 * Config _hbase.hconnection.threads.keepalivetime_ to change the timeout for 
batch pool, user can change this config according to the project's hbase write 
rate
 * For the CMS GC users, can config _-XX:__MaxTenuringThreshold_ to change 
Objects' age to promote to  Old generation

My suggestions are according to the conditions listed above, user can choose 
one or more to the project. If you think these suggestions are helpful, can add 
the useful one to the note of refguide, thx.

> document the mysterious direct memory leak in hbase 
> ----------------------------------------------------
>
>                 Key: HBASE-19320
>                 URL: https://issues.apache.org/jira/browse/HBASE-19320
>             Project: HBase
>          Issue Type: Improvement
>    Affects Versions: 1.2.6, 2.0.0
>            Reporter: huaxiang sun
>            Assignee: huaxiang sun
>            Priority: Critical
>         Attachments: HBASE-19320-master-v001.patch, Screen Shot 2017-11-21 at 
> 4.43.36 PM.png, Screen Shot 2017-11-21 at 4.44.22 PM.png
>
>
> Recently we run into a direct memory leak case, which takes some time to 
> trace and debug. Internally discussed with our [[email protected]], we 
> thought we had some findings and want to share with the community.
> Basically, it is the issue described in 
> http://www.evanjones.ca/java-bytebuffer-leak.html and it happened to one of 
> our hbase clusters.
> Create the jira first and will fill in more details later.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to