hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eli Collins (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-3870) QJM: add metrics to JournalNode
Date Sun, 02 Sep 2012 01:41:07 GMT

    [ https://issues.apache.org/jira/browse/HDFS-3870?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13446854#comment-13446854
] 

Eli Collins commented on HDFS-3870:
-----------------------------------

+1 looks great. Sync and lag are IMO the most interesting.  Only other useful metrics I can
think of are additions to the NN's JN metrics (NN -> Journal latency and failed Journal
operations) though those aren't QJM specific.
                
> QJM: add metrics to JournalNode
> -------------------------------
>
>                 Key: HDFS-3870
>                 URL: https://issues.apache.org/jira/browse/HDFS-3870
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>    Affects Versions: QuorumJournalManager (HDFS-3077)
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: hdfs-3870.txt
>
>
> The JournalNode should expose some basic metrics through the usual interface. In particular:
> - the writer epoch, accepted epoch,
> - the last written transaction ID and last committed txid (which may be newer in case
that it's in the process of catching up)
> - latency information for how long the syncs are taking
> Please feel free to suggest others that come to mind.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message