[
https://issues.apache.org/jira/browse/OOZIE-2792?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xiaobin Zheng updated OOZIE-2792:
---------------------------------
Attachment: OOZIE-2792-1.patch
Attaching patch 1.
This patch adds one more pattern for Hive2Main to parse job ID from log file in
hive-on-spark use case. I added a unit test for basic verification. The 2nd
regex from the original code seems not accurate, but I am more inclined to
leave it as is since there could be use cases that I am not aware of from
history.
> Hive2 action is not parsing Spark application ID from log file properly when
> Hive is on Spark
> ---------------------------------------------------------------------------------------------
>
> Key: OOZIE-2792
> URL: https://issues.apache.org/jira/browse/OOZIE-2792
> Project: Oozie
> Issue Type: Bug
> Components: action
> Reporter: Xiaobin Zheng
> Assignee: Xiaobin Zheng
> Priority: Minor
> Attachments: OOZIE-2792-1.patch
>
>
> When Hive2 is on Spark, hive action is not able to parse Spark application ID
> from log file as 'externalChildID' like Spark/MR actions. This makes it hard
> to tell which job hive launches from Oozie server for a particular workflow.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
