[
https://issues.apache.org/jira/browse/PHOENIX-3970?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16058865#comment-16058865
]
Lars Hofhansl commented on PHOENIX-3970:
----------------------------------------
Why does it have to be? I think this is as good as it gets without a proper
work queue. With remote RPCs done synchronously from handler threads we always
have the possibility of deadlocks when all threads are exhausted.
The work that the index rebuild was doing on the default RPC handlers had
caused the problem. At least this make it so that long running tasks are not
performed on the default RPC handlers.
> Ensure that automatic partial index rebuilds are served from the index
> handler pool
> -----------------------------------------------------------------------------------
>
> Key: PHOENIX-3970
> URL: https://issues.apache.org/jira/browse/PHOENIX-3970
> Project: Phoenix
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Attachments: 3970.txt, 3970-v2.txt
>
>
> This (and other issues) have rendered multiple larger cluster inoperable.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
