hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nemo Chen (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-5507) Several log refactoring/improvement suggestion in HDFS
Date Wed, 17 Aug 2016 19:38:21 GMT

    [ https://issues.apache.org/jira/browse/YARN-5507?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15425219#comment-15425219
] 

Nemo Chen commented on YARN-5507:
---------------------------------

Hi [~vrushalic] and [~Naganarasimha], 

I already set the YARN-5507 to supercede YARN-5508 and YARN-5511. Feel free to combine patches.

> Several log refactoring/improvement suggestion in HDFS
> ------------------------------------------------------
>
>                 Key: YARN-5507
>                 URL: https://issues.apache.org/jira/browse/YARN-5507
>             Project: Hadoop YARN
>          Issue Type: Bug
>    Affects Versions: 2.7.2
>            Reporter: Nemo Chen
>            Assignee: Vrushali C
>              Labels: easyfix, easytest
>         Attachments: YARN-5507.001.patch
>
>
> Inconsistent logging content and logging level for server nodemanager:
> Similar to a fix for MAPREDUCE-2907, in file: 
> hadoop-rel-release-2.7.2/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/test/java/org/apache/hadoop/yarn/server/nodemanager/DummyContainerManager.java
> In line 96, the logging text indicates this is a DEBUG level log, but the level is set
to info.
> {code:borderStyle=solid}
> LOG.info("DEBUG: " + req + ":" + rsrcReqs.getContainer().getContainerId());
> {code}
> ----
> Log refactoring: method invocation should be replaced by variable in yarn server
> Similar to the fix for HDFS-409. In file:
> hadoop-rel-release-2.7.2/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/security/NMTokenSecretManagerInNM.java
> {code:borderStyle=solid}
> ...
> ApplicationAttemptId appAttemptId = identifier.getApplicationAttemptId();
> ...
> LOG.debug("NMToken key updated for application attempt : "
>           + identifier.getApplicationAttemptId().toString());
> {code}
> In line 226, the method invocation identifier.getApplicationAttemptId().toString()) can
be replaced by appAttemptId.
> ----
> Exception should be recorded as caught rather than thrown
> Similar to the fix to HADOOP-657. In file:
> hadoop-rel-release-2.7.2/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-sharedcachemanager/src/main/java/org/apache/hadoop/yarn/server/sharedcachemanager/CleanerTask.java
> In line 221, the exception was caught rather than thrown.
> {code: borderStyle=solid}
> try {
>         store.cleanResourceReferences(key);
> } catch (YarnException e) {
>         LOG.error("Exception thrown while removing dead appIds.", e);
>    }
> {code}



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

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message