[
https://issues.apache.org/jira/browse/MAPREDUCE-7050?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16359243#comment-16359243
]
Igor Dvorzhak edited comment on MAPREDUCE-7050 at 2/10/18 4:29 AM:
-------------------------------------------------------------------
In case you plan to do a large scale change you may consider to use Log4j v2
with its support of lazy-logging with Java 8 lambdas
([https://logging.apache.org/log4j/2.x/manual/api.html#LambdaSupport):]
{code:java}
logger.trace("Some long-running operation returned {}", () ->
expensiveOperation());
{code}
was (Author: medb):
In case you plan to do a large scale change you can consider to use Log4j v2
with its support of lazy-logging with Java 8 lambdas
(https://logging.apache.org/log4j/2.x/manual/api.html#LambdaSupport):
{code:java}
logger.trace("Some long-running operation returned {}", () ->
expensiveOperation());
{code}
> Debug logs should be improvement
> --------------------------------
>
> Key: MAPREDUCE-7050
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-7050
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Reporter: Sen Zhao
> Priority: Minor
>
> I noticed that in Mapreduced code, there are many places to use Log.debug as
> follows:
> {code}
> LOG.debug("example " + example.getClass().getName() + "example" );
> {code}
> When the system loglevel set higher than Debug, although the system does not
> print these levels of logs, but each time will splicing parameters string,
> affect system performance.
> It should be:
> {code}
> if(LOG.isDebugEnabled()){
> LOG.debug("example " + example.getClass().getName() + "example" );
> }
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
