[
https://issues.apache.org/jira/browse/SPARK-53492?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Herman van Hövell resolved SPARK-53492.
---------------------------------------
Fix Version/s: 4.1.0
Resolution: Fixed
Issue resolved by pull request 52232
[https://github.com/apache/spark/pull/52232]
> Disallow duplicate operation id within the same session
> -------------------------------------------------------
>
> Key: SPARK-53492
> URL: https://issues.apache.org/jira/browse/SPARK-53492
> Project: Spark
> Issue Type: Improvement
> Components: Connect
> Affects Versions: 4.0.0
> Reporter: Niranjan Jayakar
> Assignee: Niranjan Jayakar
> Priority: Major
> Labels: pull-request-available
> Fix For: 4.1.0
>
>
> Sometimes, network blips can cause ExecutePlan requests to be lost and arrive
> late. In the meantime, the client might give up and retry causing a second
> request. This causes two similar ExecutePlan requests to arrive on the server.
> If the second request arrives while the first is still being processed, the
> client will simply ReattachExecute. However, in some cases, the request can
> arrive after the query is complete (for very short queries). If the query
> contains a WRITE operation, then there are now two write operations executed.
>
> Furthermore, SPARK-51425 introduced the ability for the client to customize
> the operation id. Clients can unintentionally send two queries with the same
> operation id, leading to confusing behaviour.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
