hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zhijie Shen (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-3047) [Data Serving] Set up ATS reader with basic request serving structure and lifecycle
Date Thu, 19 Mar 2015 06:01:40 GMT

    [ https://issues.apache.org/jira/browse/YARN-3047?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14368563#comment-14368563

Zhijie Shen commented on YARN-3047:

bq. Probably store related code can be removed from this JIRA and handled completely as part
of YARN-3051 to have a focussed review.

I'm okay with the plan.

bq. Yes, it is part of that. I have kept it inside hadoop-yarn-server-timelineservice

My bad, I've overlooked the package.

bq. Same config has been used by aggregator as well.

IMHO, that's something we need to fix. The aggregator should randomly pick an available port
to bind, and this port will be reported the client. Being based on config is safe now, but
will have the problem when we move to per-app aggregator mode. On a NM, there will be multiple
aggregators, which cannot bind to the same configured port. I'll file a separate ticket for

bq. TimelineReaderServer would use RPC as well for instance to serve request coming from YARN
CLI. Commands such as "yarn application" used to contact AHS if app was not found in RM. 

For some legacy problem. AHS exposes RPC interface. However, IMHO, we don't need to create
the RPC interface again in v2 as we're building the new system from the ground. What we can
do is to wrap over the REST APIs in the java client, and provide YARN CLI commands.

bq. This as per discussion with Sangjin will be done as part of YARN-3048.

To start a single web server, isn't it just some lines of commands in shell scripts? IMHO,
it doesn't need to be handled in a separate patch.

> [Data Serving] Set up ATS reader with basic request serving structure and lifecycle
> -----------------------------------------------------------------------------------
>                 Key: YARN-3047
>                 URL: https://issues.apache.org/jira/browse/YARN-3047
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: timelineserver
>            Reporter: Sangjin Lee
>            Assignee: Varun Saxena
>         Attachments: YARN-3047.001.patch, YARN-3047.02.patch
> Per design in YARN-2938, set up the ATS reader as a service and implement the basic structure
as a service. It includes lifecycle management, request serving, and so on.

This message was sent by Atlassian JIRA

View raw message