[
https://issues.apache.org/jira/browse/HADOOP-7924?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Eli Collins updated HADOOP-7924:
--------------------------------
Attachment: hadoop-7924.txt
Thanks for the good feedback. Updated patch attached, addresses your comments,
also adds an "haadmin" hook into bin/hadoop.
#1 Lemme know if my comment wrt naming made sense.
#2 Agree. Was originally thinking in this case the admin would just make the
standby active directly, but we need to run through the whole failover in this
case for fencing. Updated code and tests. For sanity, I ran bin/hadoop haadmin
failover X Y and verified that it works (and prints something reasonable) when
X isn't running, and fails when Y isn't yet running (and A remains active of
course). Note that this means if we failover from X to Y and there's a typo in
X then you might unexpectedely end up with two actives (added
testManualFailoverCanResultInTwoActives). Another jira for warning if X is not
specified as one of the options in dfs.ha.namenodes or we add a flag that's
required to failover from a service we can't connect to?
#3 Done.
#4 Done. Also fixed up the other HA exceptions to be consistent.
>
FailoverController for client-based configuration
> --------------------------------------------------
>
> Key: HADOOP-7924
> URL: https://issues.apache.org/jira/browse/HADOOP-7924
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: ha
> Affects Versions: HA Branch (HDFS-1623)
> Reporter: Eli Collins
> Assignee: Eli Collins
> Attachments: hadoop-7924.txt, hadoop-7924.txt
>
>
> Basic FailoverController to coordinate fail-over using a client-based config
> (ie fail-over from NameNode x to NameNode y).
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
