chukwa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jerome Boulon (JIRA)" <>
Subject [jira] Commented: (CHUKWA-303) collector gets stuck when unable to write to hdfs
Date Wed, 17 Jun 2009 18:14:07 GMT


Jerome Boulon commented on CHUKWA-303:

So it's down to one line for the collector and we are running 0.20 instead of 0.18
CollectorStub change: jettyServer.setStopAtShutdown(true); this is require for localWriter
to exit cleanly. 
For SeqWriter, it doesn't matter since HDFS shutdownHook is executed before SeqWriter had
a chance to rename the file on HDFS.

/etc/init.d/chukwa-collector start/stop are working properly on my system. Is it the case
for everyone?

> collector gets stuck when unable to write to hdfs
> -------------------------------------------------
>                 Key: CHUKWA-303
>                 URL:
>             Project: Hadoop Chukwa
>          Issue Type: Bug
>          Components: data collection
>    Affects Versions: 0.2.0
>            Reporter: Eric Yang
>            Assignee: Jerome Boulon
>            Priority: Blocker
>             Fix For: 0.2.0
> The recent change to collector may have problems.  Last night, I shutdown hdfs, and expecting
the collectors to bail out by themselves, and restart by the watchdog.  However, this morning,
the collector was still running and sending 0 bytes to the system.  Port 4080 isn't accepting
any more HTTP post.  This indicates that the jettyCollector may have been shutdown but the
java program didn't exist.  The only change to collector is the LocalWriter enhancement, and
it isn't enabled.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message