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 2015/01/20 17:43:35 UTC
[jira] [Updated] (AMBARI-8749) Parallel Service Actions
[ https://issues.apache.org/jira/browse/AMBARI-8749?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Hari Sekhon updated AMBARI-8749:
--------------------------------
Description:
Ambari is still doing independent service actions in serial rather than in parallel. This is basically wasting significant time waiting for different services to restart one after another when one service starting back up doesn't depend on the other service behind which they are waiting.
Each service's actions should not wait on any other service except when there is a startup dependency such as HBase, SolrCloud, HA ZKFC waiting on ZooKeeper for example.
Related to doing a cluster wide restart (
Given that HDFS Balancer action doesn't block an adjacent service restart there must already be some mechanism for this parallelism in the code?
Regards,
Hari Sekhon
http://www.linkedin.com/in/harisekhon
was:
Ambari is still doing independent service actions in serial rather than in parallel. This is basically wasting significant time waiting for different services to restart one after another when one starting back up doesn't depend on the other service behind which they are waiting.
Each service's actions should not wait any other service except when there is a startup dependency such as HBase, SolrCloud, HA ZKFC waiting on ZooKeeper for example.
Given that HDFS Balancer action doesn't block an adjacent service restart there must already be some mechanism for this parallelism in the code?
Regards,
Hari Sekhon
http://www.linkedin.com/in/harisekhon
> Parallel Service Actions
> ------------------------
>
> Key: AMBARI-8749
> URL: https://issues.apache.org/jira/browse/AMBARI-8749
> Project: Ambari
> Issue Type: Improvement
> Affects Versions: 1.7.0
> Environment: HDP 2.1
> Reporter: Hari Sekhon
>
> Ambari is still doing independent service actions in serial rather than in parallel. This is basically wasting significant time waiting for different services to restart one after another when one service starting back up doesn't depend on the other service behind which they are waiting.
> Each service's actions should not wait on any other service except when there is a startup dependency such as HBase, SolrCloud, HA ZKFC waiting on ZooKeeper for example.
> Related to doing a cluster wide restart (
> Given that HDFS Balancer action doesn't block an adjacent service restart there must already be some mechanism for this parallelism in the code?
> Regards,
> Hari Sekhon
> http://www.linkedin.com/in/harisekhon
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)