[
https://issues.apache.org/jira/browse/HIVE-18663?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16371729#comment-16371729
]
Sahil Takiar commented on HIVE-18663:
-------------------------------------
[~vihangk1] could you take a look? Its a small change. In {{SparkTask}} some of
our log statements are confusing the Spark job id with an unrelated UUID. In
{{SparkTask}} we treat {{jobRef.getJobId()}} as the {{sparkJobID}}, but
{{jobRef.getJobId()}} returns a UUID that is unrelated to the actual Spark job
id. Instead we should be using {{sparkJobStatus.getJobId()}} which returns the
actual Spark job id.
> Logged Spark Job Id contains a UUID instead of the actual id
> ------------------------------------------------------------
>
> Key: HIVE-18663
> URL: https://issues.apache.org/jira/browse/HIVE-18663
> Project: Hive
> Issue Type: Sub-task
> Components: Spark
> Reporter: Sahil Takiar
> Assignee: Sahil Takiar
> Priority: Major
> Attachments: HIVE-18663.1.patch
>
>
> We have logs like {{Spark Job[job-id]}} but the {{[job-id]}} is set to a UUID
> that is created by the RSC {{ClientProtocol}}. It should be pretty easy to
> print out the actual job id instead.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
