hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Boris Shkolnik (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-1408) Herriot NN and DN clients should vend statistics
Date Wed, 27 Oct 2010 16:08:23 GMT

    [ https://issues.apache.org/jira/browse/HDFS-1408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12925432#action_12925432
] 

Boris Shkolnik commented on HDFS-1408:
--------------------------------------

+1
small nits: 
1. looks like the only purpose of readAttr in Test is to get capacity - shouldn't it be named
like that?
2. Instead of overwriting isJmxEnabled() and getJmxPortNumber() you could create and overwrite
a method that gets and extra environment variable name.. So implementation of isJmxEnabled()
is something like:
isJmxEnabled(HADOO_OPT) || isJmxEnabled(getExtraEnvVarName())...
 this one is a matter of preference though....

> Herriot NN and DN clients should vend statistics
> ------------------------------------------------
>
>                 Key: HDFS-1408
>                 URL: https://issues.apache.org/jira/browse/HDFS-1408
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: test
>    Affects Versions: 0.22.0
>            Reporter: Al Thompson
>            Assignee: Konstantin Boudnik
>         Attachments: add_JMXlisteners.patch, HADOOP-6927.y20s.patch, HDFS-1408.patch,
HDFS-1408.patch, HDFS-1408.patch, jmx.patch, jmx.patch, jmx.patch, jmx.patch
>
>
> The HDFS web user interface serves useful information through dfshealth.jsp and dfsnodelist.jsp.
> The Herriot interface to the namenode and datanode (as implemented in NNClient and DNClient,
respectively) would benefit from the addition of some way to channel this information. In
the case of DNClient this can be an injected method that returns a DatanodeDescriptor relevant
to the underlying datanode.
> There seems to be no analagous NamenodeDescriptor. It may be useful to add this as a
facade to a visitor that aggregates values across the filesystem datanodes. These values are
(from dfshealth JSP):
> Configured Capacity
> DFS Used
> Non DFS Used
> DFS Remaining
> DFS Used%
> DFS Remaining%
> Live Nodes
> Dead Nodes
> Decommissioning Nodes
> Number of Under-Replicated Blocks
> Attributes reflecting the web user interface header may also be useful such as When-Started,
Version, When-Compiled, and Upgrade-Status.
> A NamenodeDescriptor would essentially "push down" the code in dfshealth web UI behind
a more general abstraction. If it is objectionable to make this class available in HDFS, perhaps
this could be packaged in a Herriot specific way.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message