[jira] [Updated] (HIVE-5268) HiveServer2 accumulates orphaned OperationHandle objects when a client fails while executing query

["Thiruvel Thirumoolan (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HIVE-5268?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Thiruvel Thirumoolan updated HIVE-5268:
---------------------------------------

    Attachment: HIVE-5268_prototype.patch

Attaching a preliminary patch for branch 12. As mentioned before, this patch is aggressive (was a start) in cleaning up resources on server side. As soon as a client disconnects the resources are cleaned up on HS2 (if a query is running during disconnection, the resources are cleaned up at the end of the query). This approach was designed for Hive10 and I am working on porting it to trunk and a patch will be available for Hive12 too. The newer approach will handle disconnects during async query execution and also have timeouts after which handles/sessions will be cleaned up instead of the existing aggressive approach.

Vaibhav, can I assign this to myself if you arent working on this? Thanks!

> HiveServer2 accumulates orphaned OperationHandle objects when a client fails while executing query
> --------------------------------------------------------------------------------------------------
>
>                 Key: HIVE-5268
>                 URL: https://issues.apache.org/jira/browse/HIVE-5268
>             Project: Hive
>          Issue Type: Bug
>          Components: HiveServer2
>            Reporter: Vaibhav Gumashta
>            Assignee: Vaibhav Gumashta
>             Fix For: 0.13.0
>
>         Attachments: HIVE-5268_prototype.patch
>
>
> When queries are executed against the HiveServer2 an OperationHandle object is stored in the OperationManager.handleToOperation HashMap. Currently its the duty of the JDBC client to explicitly close to cleanup the entry in the map. But if the client fails to close the statement then the OperationHandle object is never cleaned up and gets accumulated in the server.
> This can potentially cause OOM on the server over time. This also can be used as a loophole by a malicious client to bring down the Hive server.



--
This message was sent by Atlassian JIRA
(v6.1#6144)