hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Allen Wittenauer (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-913) Add a way to register long-lived services in a YARN cluster
Date Fri, 19 Sep 2014 21:50:36 GMT

    [ https://issues.apache.org/jira/browse/YARN-913?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14141363#comment-14141363
] 

Allen Wittenauer commented on YARN-913:
---------------------------------------

* I have some concern around 'naked' zookeeper.* config options.  If another part of Hadoop
uses ZK, would they be expected to use the same ZK options?  e.g., what if I want a different
ZK setup for a YARN registrar vs. the NN?  

* Is there any risk about a user writing to a shared ZK with the RM? i.e., if a user kills
the ZK used for app registry through some action, what happens to the RM and other user's
bits that are running?

 * Why doesn't the hostname component allow for FQDNs? 

* Are we prepared for more backlash when another component requires working DNS? :)

* Is ZK actually the right thing to use here?
{code}
+Zookeeper has a default limit of 1MB/node. If all endpoints of a service or
+component are stored in JSON attached to that node, then there is a total limit
+of 1MB of all endpoint registration data.
{code}

If we are reducing certain fields out of fear of hitting this limit, this seems to indicate
that ZK is a bad fit and/or we are using ZK incorrectly.  It could be argued that ZK should
contain a pointer to the actual data stored on HDFS.  This allows for future growth without
having to worry about blowing ZK out.

> Add a way to register long-lived services in a YARN cluster
> -----------------------------------------------------------
>
>                 Key: YARN-913
>                 URL: https://issues.apache.org/jira/browse/YARN-913
>             Project: Hadoop YARN
>          Issue Type: New Feature
>          Components: api, resourcemanager
>    Affects Versions: 2.5.0, 2.4.1
>            Reporter: Steve Loughran
>            Assignee: Steve Loughran
>         Attachments: 2014-09-03_Proposed_YARN_Service_Registry.pdf, 2014-09-08_YARN_Service_Registry.pdf,
RegistrationServiceDetails.txt, YARN-913-001.patch, YARN-913-002.patch, YARN-913-003.patch,
YARN-913-003.patch, YARN-913-004.patch, YARN-913-006.patch, YARN-913-007.patch, YARN-913-008.patch,
yarnregistry.pdf, yarnregistry.tla
>
>
> In a YARN cluster you can't predict where services will come up -or on what ports. The
services need to work those things out as they come up and then publish them somewhere.
> Applications need to be able to find the service instance they are to bond to -and not
any others in the cluster.
> Some kind of service registry -in the RM, in ZK, could do this. If the RM held the write
access to the ZK nodes, it would be more secure than having apps register with ZK themselves.



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

Mime
View raw message