You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@myriad.apache.org by "Swapnil Daingade (JIRA)" <ji...@apache.org> on 2015/09/09 02:07:45 UTC

[jira] [Created] (MYRIAD-132) Support active-standby RM in Myriad HA implementation

Swapnil Daingade created MYRIAD-132:
---------------------------------------

             Summary: Support active-standby RM in Myriad HA implementation
                 Key: MYRIAD-132
                 URL: https://issues.apache.org/jira/browse/MYRIAD-132
             Project: Myriad
          Issue Type: Improvement
          Components: Scheduler
    Affects Versions: Myriad 0.1.0
            Reporter: Swapnil Daingade
            Assignee: Swapnil Daingade


Currently if an RM goes down, we expect an external entity (like Marathon) to launch a new instance of the RM. This new instance of RM then recovers state from the state store left by the previous instance. 

Yarn currently supports a active-standby RM configuration. To support this configuration, we will need to delay Myriad initialization till the RM becomes active. 



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