[jira] [Commented] (YARN-4083) Add a discovery mechanism for the scheduler address

["Vinod Kumar Vavilapalli (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/YARN-4083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15178423#comment-15178423 ] 

Vinod Kumar Vavilapalli commented on YARN-4083:
-----------------------------------------------

bq. Today many apps like Distributed Shell, REEF, etc rely on the fact that the HADOOP_CONF_DIR of the NM is on the classpath to discover the scheduler address.
Please see my comment [here|https://issues.apache.org/jira/browse/YARN-4650?focusedCommentId=15176322&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15176322] on YARN-4650. After rolling-upgrades (YARN-666), for correctness sake, we require all apps to *not* depend on server side config files (which may change during upgrades).

bq. I feel an environment variable should be accessible by linux, windows and other containers.
I think [~aw] is making the same comment I made in the design doc (in section 3.2.5) for YARN-4692. Pasting that comment below:
{quote}
All of our platform-­to­-application communication currently is only through process environment variables: for e.g. A​pplicationContants.NM_HOST.​ With things like Linux CGroups, containerization through docker etc, it is now possible to launch multi­-process containers where the solution of environmental­ variables breaks down. [(Task) We need better ways of propagating important information down to the containers ­ information like container’s resource size, local­-dirs and log­-dirs available for writing etc.

{quote}

> Add a discovery mechanism for the scheduler address
> ---------------------------------------------------
>
>                 Key: YARN-4083
>                 URL: https://issues.apache.org/jira/browse/YARN-4083
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: nodemanager, resourcemanager
>            Reporter: Subru Krishnan
>            Assignee: Subru Krishnan
>
> Today many apps like Distributed Shell, REEF, etc rely on the fact that the HADOOP_CONF_DIR of the NM is on the classpath to discover the scheduler address. This JIRA proposes the addition of an explicit discovery mechanism for the scheduler address



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