Harsh J created OOZIE-964:
-----------------------------
Summary: A mechanism to kill and retry workflows from coordinators
Key: OOZIE-964
URL: https://issues.apache.org/jira/browse/OOZIE-964
Project: Oozie
Issue Type: New Feature
Components: coordinator
Reporter: Harsh J
Priority: Minor
One use-case of coordinator app seems to be to run a WF periodically, for say,
generating updates into a HBase table from some MR input. In this case, if the
MR job thats running for one of the WFs, has still not finished before the next
instance is launched by the coordinator, it may be beneficial to kill the
earlier jobs to free up slots so that the next one can proceed faster and be
able to digest both the previous and the current inputs in its run.
We could provide a mechanism to auto-kill the WF instances launched from a
coordinator and all its sub-jobs, to facilitate this at the scheduler level
itself. The kill can happen based on a provided timeout value in the coord app,
as one example way to do it.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
