[
https://issues.apache.org/jira/browse/HIVE-5794?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13819877#comment-13819877
]
Prasad Mujumdar commented on HIVE-5794:
---------------------------------------
[~brocknoland] It might be a good idea to add server.stop() in the exception
block before retry. There could be other resource like metastore connection,
thrift transport etc that would get cleaned up before attempting next retry.
> HiveServer2 should retry startup
> --------------------------------
>
> Key: HIVE-5794
> URL: https://issues.apache.org/jira/browse/HIVE-5794
> Project: Hive
> Issue Type: Improvement
> Affects Versions: 0.13.0
> Reporter: Brock Noland
> Assignee: Brock Noland
> Attachments: HIVE-5794.patch, HIVE-5794.patch
>
>
> On startup the CLIService attempts to create scratch directories. If that
> fails, e.g. because the nn is also starting, HS2 fails to start. We should
> retry this operation this a few times with sleeps.
> It might be preferable todo this directly in HS2 main so that we can catch
> all errors.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
