[
https://issues.apache.org/jira/browse/HBASE-15525?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15214526#comment-15214526
]
stack commented on HBASE-15525:
-------------------------------
So in your patch, ByteBufferPool is a FixedSizeBBP?
So we allocate in a step function adding a new BB of fixed size to the OS to
fill? If FixedSizeBBP is full, we allocate from Heap. Sounds good.
Tie the new pool to the new Stream? Give them same prefix. They work together.
Looks good [~anoop.hbase]
> OutOfMemory could occur when using BoundedByteBufferPool during RPC bursts
> --------------------------------------------------------------------------
>
> Key: HBASE-15525
> URL: https://issues.apache.org/jira/browse/HBASE-15525
> Project: HBase
> Issue Type: Bug
> Components: IPC/RPC
> Reporter: deepankar
> Assignee: Anoop Sam John
> Priority: Critical
> Attachments: WIP.patch
>
>
> After HBASE-13819 the system some times run out of direct memory whenever
> there is some network congestion or some client side issues.
> This was because of pending RPCs in the RPCServer$Connection.responseQueue
> and since all the responses in this queue hold a buffer for cellblock from
> BoundedByteBufferPool this could takeup a lot of memory if the
> BoundedByteBufferPool's moving average settles down towards a higher value
> See the discussion here
> [HBASE-13819-comment|https://issues.apache.org/jira/browse/HBASE-13819?focusedCommentId=15207822&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15207822]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
