[
https://issues.apache.org/jira/browse/OOZIE-2108?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14319203#comment-14319203
]
Bowen Zhang commented on OOZIE-2108:
------------------------------------
[~shwethags], {quote}APIs either kill all jobs that match the filter or they do
not kill any. Assume you want to kill all jobs between some time range, you
need to make sure all jobs are eligible for kill. If all are not eligible, you
need to find smaller time ranges that match the kill pre-condition which is a
pain. Instead, can the API kill the jobs that match the pre-condition and leave
the rest as is. In this case, the returned result should also specify which
jobs were killed and which weren't (or just return the status of each). The
same applies for suspend and resume{quote}
makes sense. Will do that. It takes me forever to upload to Review board.
> bulk kill, suspend, resume jobs using existing filter, offset, len, and
> jobtype params
> --------------------------------------------------------------------------------------
>
> Key: OOZIE-2108
> URL: https://issues.apache.org/jira/browse/OOZIE-2108
> Project: Oozie
> Issue Type: New Feature
> Reporter: Bowen Zhang
> Assignee: Bowen Zhang
> Attachments: oozie-2108.patch
>
>
> Currently, there is no bulk write operations in "jobs" API. We would like to
> first introduce a bulk kill operation that kills all jobs which satisfy the
> filter. The desired usage will be {noformat}oozie jobs -oozie
> http://localhost:11000/oozie -kill -filter name=something{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
