[
https://issues.apache.org/jira/browse/AURORA-236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13913922#comment-13913922
]
Kevin Sweeney commented on AURORA-236:
--------------------------------------
Apparently this already exists (set max_failures to -1). [1]
[1]
https://github.com/apache/incubator-aurora/blob/master/src/main/java/org/apache/aurora/scheduler/state/TaskStateMachine.java?source=c#L197-L200
> Document "Run until success" semantics for adhoc jobs
> -----------------------------------------------------
>
> Key: AURORA-236
> URL: https://issues.apache.org/jira/browse/AURORA-236
> Project: Aurora
> Issue Type: Story
> Components: Documentation, Scheduler
> Reporter: Kevin Sweeney
>
> Inspired by the thread here [1], it would be useful to specify that an adhoc
> job is "run until success" rather than require an explicit max_failures cap.
> This would allow it to free its cluster resources after it exits, but
> reliably be retried to completion (something like a "terminating service").
> [1]
> http://mail-archives.apache.org/mod_mbox/aurora-dev/201402.mbox/%3CCAL0VptO3wSg_8Qb_qP%3D9uYPoZbHDjhbr3mEVzV%3DH03_h59%3DBZA%40mail.gmail.com%3E
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
