[
https://issues.apache.org/jira/browse/HADOOP-5301?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12676095#action_12676095
]
Ari Rabkin commented on HADOOP-5301:
------------------------------------
I don't love using miliseconds as a nonce; it seems kinda ugly. But I can't
think of anything better. +1
> Hadoop Metrics Log file should be used by only one process
> ----------------------------------------------------------
>
> Key: HADOOP-5301
> URL: https://issues.apache.org/jira/browse/HADOOP-5301
> Project: Hadoop Core
> Issue Type: Bug
> Components: contrib/chukwa
> Reporter: Jerome Boulon
> Assignee: Jerome Boulon
> Attachments: HADOOP-5301.patch
>
>
> All JVM info are going to chukwa-<user>-jvm.log so more than one process is
> writing to this log file.
> The problem is that each new process will register a new adaptor starting at
> the beginning of the file.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
