You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Eric Yang (JIRA)" <ji...@apache.org> on 2008/11/05 06:59:44 UTC

[jira] Assigned: (HADOOP-4434) Improve log rotation and reduce communication between agent and logging process, make agent more fault tolerant

     [ https://issues.apache.org/jira/browse/HADOOP-4434?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Eric Yang reassigned HADOOP-4434:
---------------------------------

    Assignee: Eric Yang

> Improve log rotation and reduce communication between agent and logging process, make agent more fault tolerant
> ---------------------------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-4434
>                 URL: https://issues.apache.org/jira/browse/HADOOP-4434
>             Project: Hadoop Core
>          Issue Type: Improvement
>         Environment: Redhat Enterprise Linux 4.5, Java 6
>            Reporter: Eric Yang
>            Assignee: Eric Yang
>   Original Estimate: 96h
>  Remaining Estimate: 96h
>
> - Chukwa Agent does not handle rotated log correctly.  The current design requires the logging process to pause the agent, rotate the log, then resume the log streaming.  The three communication between the running process and chukwa agent could become a problem when either or one of the process is down.  The implementation should be able to detect log rotation in the chukwa agent.
> - Chukwa Agent should detect and prevent the same log being streamed more than once.
> - Chukwa Agent should synchronize data streaming addition when picking up the check point file.

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