hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Karthik Kambatla (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-1028) Add FailoverProxyProvider like capability to RMProxy
Date Fri, 13 Dec 2013 22:54:07 GMT

    [ https://issues.apache.org/jira/browse/YARN-1028?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13848025#comment-13848025

Karthik Kambatla commented on YARN-1028:

Thanks Vinod. 

bq. Shouldn't RMProxy.createFailoverProxyProvider actually be in RMFailoverProxyProvider itself?
My initial patch had it that way, RMFailoverProxyProvider was an abstract class then. For
consistency with HDFS, Tom suggested we handle everything in ConfiguredRMFailoverProxyProvider
itself. Do you suggest changing it back to an abstract class. If yes, there are other methods
we can choose to move there. See yarn-1028-4.patch

bq. Is it possible to consolidate the configuration properties for both HA and non-HA case?
If not, we'll have to fix documentation in yarn-default for non-ha configs like connect.max-wait.ms
that they are not used in HA mode
Again, this is a trade-off between how much consistency we want with HDFS. My initial idea
was to infer these values from the existing yarn "connect." configs, but it kind of gets complicated.
Thinking more about it, we could compute numFailovers using connect.max-wait-ms when yarn.client.failover.max-attempts
is not specified?

> Add FailoverProxyProvider like capability to RMProxy
> ----------------------------------------------------
>                 Key: YARN-1028
>                 URL: https://issues.apache.org/jira/browse/YARN-1028
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>            Reporter: Bikas Saha
>            Assignee: Karthik Kambatla
>         Attachments: yarn-1028-1.patch, yarn-1028-2.patch, yarn-1028-3.patch, yarn-1028-4.patch,
yarn-1028-5.patch, yarn-1028-6.patch, yarn-1028-7.patch, yarn-1028-8.patch, yarn-1028-draft-cumulative.patch
> RMProxy layer currently abstracts RM discovery and implements it by looking up service
information from configuration. Motivated by HDFS and using existing classes from Common,
we can add failover proxy providers that may provide RM discovery in extensible ways.

This message was sent by Atlassian JIRA

View raw message