[
https://issues.apache.org/jira/browse/PHOENIX-4020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16086714#comment-16086714
]
Samarth Jain commented on PHOENIX-4020:
---------------------------------------
We have to be a little careful and make sure the change is backward compatible.
When the cluster is going through a rolling restart, the new servers will be
sending out RPCs with the new priority. And the destination servers which are
running the old code may not know how to handle the priority and may end up
falling back to using the regular handler pool. The situation would be resolved
though once the rolling restart is complete.
> Lower our medium and high priority numbers below HBase meta priority
> --------------------------------------------------------------------
>
> Key: PHOENIX-4020
> URL: https://issues.apache.org/jira/browse/PHOENIX-4020
> Project: Phoenix
> Issue Type: Bug
> Reporter: James Taylor
>
> We should lower our Phoenix medium/high priority numbers below HBase meta
> priority (which I believe is set to 200).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
