[
https://issues.apache.org/jira/browse/HAWQ-268?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15070804#comment-15070804
]
ASF GitHub Bot commented on HAWQ-268:
-------------------------------------
Github user liming01 commented on the pull request:
https://github.com/apache/incubator-hawq/pull/214#issuecomment-167076757
Good fix.
BTW, if we can auto detect it in recovery mode, then wait until it
finished, instead of hard code to sleep 5 minutes, it would be much better in
case of some scenario that the recovery may take more than 5 seconds to finish.
> HAWQ activate standby fails
> ---------------------------
>
> Key: HAWQ-268
> URL: https://issues.apache.org/jira/browse/HAWQ-268
> Project: Apache HAWQ
> Issue Type: Bug
> Components: Command Line Tools
> Reporter: Radar Lei
> Assignee: Lei Chang
>
> Since we changed the error out logic of 'hawq stop/start', now hawq activate
> standby will fail due to stop/start cluster error.
> Now we should change to :
> 1. While doing activate standby, hawq will try to stop master/standby node,
> but not fail out if they can't get stopped. Need to make sure no standby
> syncmaster running.
> 2. We should consider if it's necessary to stop/start segments during the
> activate progress.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
