You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ambari.apache.org by "Hari Sekhon (JIRA)" <ji...@apache.org> on 2014/08/01 13:02:38 UTC

[jira] [Created] (AMBARI-6706) Rolling restarts for HA Masters eg HDFS NameNodes

Hari Sekhon created AMBARI-6706:
-----------------------------------

             Summary: Rolling restarts for HA Masters eg HDFS NameNodes
                 Key: AMBARI-6706
                 URL: https://issues.apache.org/jira/browse/AMBARI-6706
             Project: Ambari
          Issue Type: Improvement
    Affects Versions: 1.6.1
         Environment: Hortonworks HDP 2.1 with Ambari 1.6.1
            Reporter: Hari Sekhon


Rolling restarts currently only work for slaves such as Datanodes in Web UI when doing Service Actions -> Restart Datanodes. When doing a Restart All it doesn't give any option to do a rolling restart across all components including masters.

I usually configure HDFS NameNode HA but this means that in Ambari there is downtime in reconfiguration even with NN HA enabled since it doesn't do rolling restarts of the NameNodes in order to keep the cluster up.

This is needed to actually make the service really HA including HA during routine maintenance and reconfiguration. Maybe I could workaround to shut master components one by one manually but would be nice if Ambari could support this directly.

Regards,

Hari Sekhon
http://www.linkedin.com/in/harisekhon



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