Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Fri, 8 May 2015 15:23:00 +0000 (UTC)
From: "Kihwal Lee (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.12826986.1430761806000.50131.1431098580659@Atlassian.JIRA>
In-Reply-To: <JIRA.12826986.1430761806000@Atlassian.JIRA>
References: <JIRA.12826986.1430761806000@Atlassian.JIRA>
 <JIRA.12826986.1430761806680@arcas>
Subject: [jira] [Commented] (HDFS-8315) Datanode does not log reads
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HDFS-8315?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14534707#comment-14534707 ] 

Kihwal Lee commented on HDFS-8315:
----------------------------------

Changing log level of ClientTraceLog to debug will work for now since there is nothing else logging at debug.

> Datanode does not log reads
> ---------------------------
>
>                 Key: HDFS-8315
>                 URL: https://issues.apache.org/jira/browse/HDFS-8315
>             Project: Hadoop HDFS
>          Issue Type: Bug
>            Reporter: Kihwal Lee
>            Assignee: Kihwal Lee
>            Priority: Critical
>              Labels: BB2015-05-TBR
>         Attachments: HDFS-8315.patch
>
>
> HDFS-6836 made datanode read request logging DEBUG.  There is a good reason why it was at INFO for so many years.  This is very useful in debugging load issues. This jira will revert HDFS-6836.
> We haven't seen it being a bottleneck on busy hbase clusters, but if someone thinks it is a serious overhead, please make it configurable in a separate jira.


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)