[ 
https://issues.apache.org/jira/browse/YARN-3672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15420589#comment-15420589
 ] 

Jian He commented on YARN-3672:
-------------------------------

- How will the RM failover play together with   YARN-3673 ? Let’s say 
subCluster1(RM1, RM2), subCluster2(RM3, RM4). Looks like the implementation 
will ignore cluster intra-failover and do cluster inter-failover only ?
- question for such API. It asks for a specific subCluster info, do we still 
need the filterInactiveSubClusters flag ? Even if it’s required, the behavior 
for the if/else is inconsistent, the if case is honoring the flag, while the 
else doesn’t.
{code}
public SubClusterInfo getSubCluster(final SubClusterId subClusterId,
    final boolean filterInactiveSubClusters) throws YarnException {
  if (isCachingEnabled()) {
    return getSubClusters(filterInactiveSubClusters).get(subClusterId);
  } else {
    return stateStore
        .getSubCluster(GetSubClusterInfoRequest.newInstance(subClusterId))
        .getSubClusterInfo();
  }
}
{code}
- I think we should not reuse these two configs for retry, the default value of 
both is zero.
{code}
conf.getInt(YarnConfiguration.CLIENT_FAILOVER_RETRIES,
    YarnConfiguration.DEFAULT_CLIENT_FAILOVER_RETRIES),
conf.getLong(
    YarnConfiguration.CLIENT_FAILOVER_RETRIES_ON_SOCKET_TIMEOUTS,
    YarnConfiguration.DEFAULT_CLIENT_FAILOVER_RETRIES_ON_SOCKET_TIMEOUTS),
{code}

> Create Facade for Federation State and Policy Store
> ---------------------------------------------------
>
>                 Key: YARN-3672
>                 URL: https://issues.apache.org/jira/browse/YARN-3672
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: nodemanager, resourcemanager
>            Reporter: Subru Krishnan
>            Assignee: Subru Krishnan
>         Attachments: YARN-3672-YARN-2915-v1.patch, 
> YARN-3672-YARN-2915-v2.patch, YARN-3672-YARN-2915-v3.patch
>
>
> This JIRA proposes creating a facade for Federation State and Policy Store to 
> simply access and have a common place for cache management etc that can be 
> used by both Router & AMRMProxy



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

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to