hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zhe Zhang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-8920) Erasure Coding: when recovering lost blocks, logs can be too verbose and hurt performance
Date Thu, 20 Aug 2015 22:50:47 GMT

    [ https://issues.apache.org/jira/browse/HDFS-8920?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14705920#comment-14705920
] 

Zhe Zhang commented on HDFS-8920:
---------------------------------

Thanks for performing the tests and reporting the issue [~lirui].

I agree with Colin. Changing the log level will affect non-EC I/O operations, because in that
case {{chosenNode == null}} does indicate a pretty high chance of data loss. A non-ideal solution
is to check whether {{block}} is striped ({{BlockIdManager#isStripedBlockID}} and then determine
whether to warn or debug. I'm OK with that solution for now.

> Erasure Coding: when recovering lost blocks, logs can be too verbose and hurt performance
> -----------------------------------------------------------------------------------------
>
>                 Key: HDFS-8920
>                 URL: https://issues.apache.org/jira/browse/HDFS-8920
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Rui Li
>            Assignee: Rui Li
>
> When we test reading data with datanodes killed, {{DFSInputStream::getBestNodeDNAddrPair}}
becomes a hot spot method and effectively blocks the client JVM. This log seems too verbose:
> {code}
>     if (chosenNode == null) {
>       DFSClient.LOG.warn("No live nodes contain block " + block.getBlock() +
>           " after checking nodes = " + Arrays.toString(nodes) +
>           ", ignoredNodes = " + ignoredNodes);
>       return null;
>     }
> {code}



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

Mime
View raw message