[
https://issues.apache.org/jira/browse/HIVE-9253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14264735#comment-14264735
]
Brock Noland commented on HIVE-9253:
------------------------------------
Hi,
Thank you [~dongc] for your excellent investigation! Because of the issue and
the fact that clients could be legitimately idle for long periods, I was
thinking we could do the following:
1) We add the client timeout property as a new optional arg to the request
going to the server. We'd have to look at the best way to do this. Might be on
a per-request basis or a session-level, e.g. create a new request
{{SetSessionPropertiesRequest}}.
2) We store that value and the start time of a request in a thread local of a
new class "RuntimeTimeout"
3) In places where we know we can be processing long running requests we check
this timeout periodically by calling a method {{RuntimeTimeout.checkTimeOut}}
If we've timeout we throw a new exception {{RuntimeTimeoutException}}.
Thoughts?
> MetaStore server should detect client timeout
> ---------------------------------------------
>
> Key: HIVE-9253
> URL: https://issues.apache.org/jira/browse/HIVE-9253
> Project: Hive
> Issue Type: Sub-task
> Components: Metastore
> Reporter: Dong Chen
> Assignee: Dong Chen
> Attachments: HIVE-9253.patch
>
>
> In the description of HIVE-7195, one issue is that MetaStore client timeout
> is quite dumb. The client will timeout and the server has no idea the client
> is gone.
> The server should detect client timeout.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
