www-infrastructure-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jukka Zitting (JIRA)" <j...@apache.org>
Subject [jira] Resolved: (INFRA-2180) mapreduce git repository not synch'ing commits from the svn repository
Date Thu, 19 Nov 2009 09:01:41 GMT

     [ https://issues.apache.org/jira/browse/INFRA-2180?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Jukka Zitting resolved INFRA-2180.

    Resolution: Fixed

An earlier update process had apparently died and left a stale lock behind. I cleared the
lock and triggered the sync.

If you want to help prevent similar problems in the future, I'm looking for ways to improve
the lock handling in the bin/update-mirror.sh script found in


The main problem is that currently the script is triggered by incoming email (and sometimes
manually), which forces us to deal with concurrent requests. There's a new svnpubsub mechanism
that would allow us to listen for changes from a single daemon process, which would make it
easier to control concurrency. For more details about svnpubsub, see


More details on infra-dev@

> mapreduce git repository not synch'ing commits from the svn repository
> ----------------------------------------------------------------------
>                 Key: INFRA-2180
>                 URL: https://issues.apache.org/jira/browse/INFRA-2180
>             Project: Infrastructure
>          Issue Type: Bug
>      Security Level: public(Regular issues) 
>          Components: Git
>            Reporter: Vinod K V
>            Assignee: Jukka Zitting
> HADOOP-4491 was committed nearly 20hrs back to mapreduce svn (http://svn.apache.org/repos/asf/hadoop/mapreduce/trunk),
but it is still not getting reflected in the git mirror(http://git.apache.org/hadoop-mapreduce.git)

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

View raw message