hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Varun Saxena (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-3047) [Data Serving] Set up ATS reader with basic request serving structure and lifecycle
Date Sat, 21 Mar 2015 12:30:39 GMT

    [ https://issues.apache.org/jira/browse/YARN-3047?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14372708#comment-14372708
] 

Varun Saxena commented on YARN-3047:
------------------------------------

Thanks for the comments [~gtCarrera]

bq. NameValuePair duplicates the same file in ahs, timeline/NameValuePair. 
We can probably move it to yarn-api. Thoughts, [~zjshen].

bq. We're using Java 7 now so we can use switch statements for strings:
ok

bq. TimelineReaderServer, notice that the following lines:
ok, will fix.

bq. This about info seems to be confusing.
Will change it to Timeline Reader API.

bq. In the original design, there was a reader pool that holds a set of readers.
This will be handled in next phase(check YARN-3118). Right now, we will have a single daemon.

bq. In TimelineReaderStore, DEFAULT_LIMIT is never used.
It will be used by subclasses. This is like v1 where if the client didn't specify a limit/number
of records to be returned, this DEFAULT_LIMIT was used.

bq. Why are we catching Exceptions, rather than the precise exception readValue may throw?
Yes, we can catch them.

bq. Just want to double check that we do want to have the same end point for both timeline
readers and collectors. (Seems fine with me, since the reader process runs on different machines
as our collectors. )
Anyways I was using a different host and port configuration for reader. This has to be discussed
though as [~zjshen] thinks we can use the same v1 config.


bq. I would be very appreciate if you could upload a simple writeup for the reader design.
Yes, working on it. Sorry for the delay. Got busy with other work last week.

> [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
(v6.3.4#6332)

Mime
View raw message