You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@aurora.apache.org by "Maxim Khutornenko (JIRA)" <ji...@apache.org> on 2014/05/17 02:06:14 UTC

[jira] [Commented] (AURORA-443) Aurora admin perform_maintenance_hosts must bail out if no hosts found

    [ https://issues.apache.org/jira/browse/AURORA-443?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14000524#comment-14000524 ] 

Maxim Khutornenko commented on AURORA-443:
------------------------------------------

A possible solution here could be returning a list of known (to the scheduler) hosts with successful maintenance attribute set and let client work against that list rather than the original one.

> Aurora admin perform_maintenance_hosts must bail out if no hosts found
> ----------------------------------------------------------------------
>
>                 Key: AURORA-443
>                 URL: https://issues.apache.org/jira/browse/AURORA-443
>             Project: Aurora
>          Issue Type: Bug
>          Components: Client
>            Reporter: Maxim Khutornenko
>
> As currently implemented, a mistyped but valid cluster name in perform_maintenance_hosts command results in skipping the maintenance mode for the provided hosts but correctly executes post drain script on them. This results in lost tasks rather than graceful draining.
> Example: "admin perform_maintenance_hosts cstrA --hosts=host1-cstrB-prod" does not fail to set maintenance and proceed straight to post drain script.  



--
This message was sent by Atlassian JIRA
(v6.2#6252)