hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aroop Maliakkal (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-2069) DFS Summary Page
Date Fri, 30 Nov 2007 10:12:43 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-2069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12547065
] 

Aroop Maliakkal commented on HADOOP-2069:
-----------------------------------------

We are building a dash board in which we want to put the "DFS Summary" also. We *don't* want
to show Live and Dead Nodes and just the "DFS Summary" is enough. One way we can do is parsing
 the dfshealth.jsp and put the output in the dashboard. The problem is that every time if
we are loading the dfshealth.jsp for a 2000 nodes cluster, it will put some extra load on
the NameNode which we obviously want to avoid it. The current dfshealth.jsp is quite big and
it takes long time to load. We are actually interested in only in "Cluster Summary" section
in dfshealth.jsp page. So if you can give a xml output of *just* DFS Summary, we can include
it in our dash board without putting any load on namenode and we can use the same info for
some of our monitoring scripts also for checking how many nodes are *live/dead* .

Let me know if it is not clear.

> DFS Summary Page
> ----------------
>
>                 Key: HADOOP-2069
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2069
>             Project: Hadoop
>          Issue Type: Improvement
>          Components: util
>            Reporter: Aroop Maliakkal
>
> DFS Summary page should consists of the DFS summary data and *links* to Live Datanodes
and Dead datanodes instead of loading "all nodes" information in one page. It would be great
if you can give the summary in XML format also.

-- 
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