[jira] [Commented] (BEAM-9065) Spark runner accumulates metrics (incorrectly) between runs

[Etienne Chauchot (Jira)]

    [ 
https://issues.apache.org/jira/browse/BEAM-9065?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17011875#comment-17011875
 ] 

Etienne Chauchot commented on BEAM-9065:
----------------------------------------

[~udim] If 2.18 is not released when the patch is ready I prefer that it is 
cherry picked to 2.18

Also I'm still investigating in the precise environment to reproduce this 
observed behavior so it might not come before 2.19

> Spark runner accumulates metrics (incorrectly) between runs
> -----------------------------------------------------------
>
>                 Key: BEAM-9065
>                 URL: https://issues.apache.org/jira/browse/BEAM-9065
>             Project: Beam
>          Issue Type: Bug
>          Components: runner-spark
>            Reporter: Etienne Chauchot
>            Assignee: Etienne Chauchot
>            Priority: Major
>             Fix For: 2.18.0
>
>          Time Spent: 40m
>  Remaining Estimate: 0h
>
> When pipeline.run() is called, MetricsAccumulator (wrapper of 
> MetricsContainerStepMap spark accumulator) is initialized. Spark needs this 
> class to be a singleton for failover. The problem is that when several 
> pipelines are run inside the same JVM, the initialization of 
> MetricsAccumulator singleton does not reset the underlying spark accumulator  
> causing metrics to be accumulated between runs.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)