[jira] [Commented] (MESOS-2522) Add reason field for framework errors

["Benjamin Bannier (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MESOS-2522?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15177776#comment-15177776 ] 

Benjamin Bannier commented on MESOS-2522:
-----------------------------------------

The master currently responds with the following {{FrameworkErrorMessage}} contents:

||Master method ||{{FrameworkErrorMessage}} content||
| {{Master::exceededCapacity}}    |  Message {{$MESSAGE_NAME}} dropped: capacity ({{$CAPACITY}}) exceeded |
| {{Master::registerFramework}}   |  Registered with 'id' already set |
| {{Master::reregisterFramework}} |  Re-registering without an 'id' |
| {{Master::subscribe}}           |  Role '{{$FRAMEWORK_ROLE}}' is not present in the master's --roles |
|                             |  User 'root' is not allowed to run frameworks without --root_submissions set |
|                             |  Framework has been removed |
| {{Master::_subscribe}}          |  Authorization failure: {{$AUHTORIZATION_FAILURE}} |
|                             |  Not authorized to use role '{{$FRAMEWORK_ROLE}}' |
|                             |  Framework is already connected |
|                             |  Framework failed over |
| {{Master::failoverFramework}}   |  Framework failed over |

AFAIK currently the only message requiring a new {{FrameworkID}} is {{Framework has been removed}}.


> Add reason field for framework errors
> -------------------------------------
>
>                 Key: MESOS-2522
>                 URL: https://issues.apache.org/jira/browse/MESOS-2522
>             Project: Mesos
>          Issue Type: Improvement
>          Components: master
>    Affects Versions: 0.22.0
>            Reporter: Connor Doyle
>            Priority: Minor
>              Labels: mesosphere, newbie
>
> Currently, the only insight into framework errors is a message string.  Framework schedulers could probably be smarter about how to handle errors if the cause is known.  Since there are only a handful of distinct cases that could trigger an error, they could be captured by an enumeration.
> One specific use case for this feature follows. Frameworks that intend to survive failover typicaly persist the FrameworkID somewhere.  When a framework has been marked completed by the master for exceeding its configured failover timeout, then re-registration triggers a framework error.  Probably, the scheduler wants to disambiguate this kind of framework error from others in order to invalidate the stashed FrameworkID for the next attempt at (re)registration.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)