hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Purtell (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17780) BoundedByteBufferPool "At capacity" messages are not actionable
Date Tue, 14 Mar 2017 17:17:41 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17780?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15924599#comment-15924599
] 

Andrew Purtell commented on HBASE-17780:
----------------------------------------

Going to commit trivial patch later today unless objection

> BoundedByteBufferPool "At capacity" messages are not actionable
> ---------------------------------------------------------------
>
>                 Key: HBASE-17780
>                 URL: https://issues.apache.org/jira/browse/HBASE-17780
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 1.3.0
>            Reporter: Andrew Purtell
>            Assignee: Andrew Purtell
>            Priority: Minor
>             Fix For: 2.0.0, 1.4.0, 1.31
>
>         Attachments: HBASE-17780-branch-1.patch, HBASE-17780.patch
>
>
> This comment in BoundedByteBufferPool talks about "At capacity ..." warnings from this
class that may appear in logs when under load:
> {code}
>  * If a ByteBuffer is bigger than the configured threshold, we will just let the ByteBuffer
go
>  * rather than add it to the pool. If more ByteBuffers than the configured maximum instances,
>  * we will not add the passed ByteBuffer to the pool; we will just drop it
>  * (we will log a WARN in this case that we are at capacity).
> {code}
> First, dropping buffers when the pool is full is obviously an expected and normal condition.
Second, there is nothing actionable about that warning. Might be useful for developers, perhaps.
Drop it to DEBUG. 



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message