ambari-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sumit Mohanty (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (AMBARI-1994) Adding component to Host should should wire-up + adjust associated Nagios alerts
Date Tue, 23 Apr 2013 04:01:15 GMT

    [ https://issues.apache.org/jira/browse/AMBARI-1994?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13638758#comment-13638758
] 

Sumit Mohanty commented on AMBARI-1994:
---------------------------------------

The issue is as follows:
* Both DATANODE and TASKTRACKER component hosts are represented by the same agent side param
"slave_hosts" populated by BE. If there is a mismatch between the list of hosts for DATANODE
and TASKTRACKER components, then the last list wins (TASKTRACKER it is). So the list of hosts
sent through param "slave_hosts" does not list all hosts that has DATANODE
* We need to introduce separate parameters for DATANODE and TASKTRACKER (similar to "hbase-rs_hosts"
for REGIONSERVER)
* We need to wire-up the nagios alerts to appropriate parameters

Most changes are to the puppet files with no unit tests. However, the scenario is manually
verified with progressive addition of DATANODE, TASKTRACKER, and HBASE_REGIONSERVER.
                
> Adding component to Host should should wire-up + adjust associated Nagios alerts
> --------------------------------------------------------------------------------
>
>                 Key: AMBARI-1994
>                 URL: https://issues.apache.org/jira/browse/AMBARI-1994
>             Project: Ambari
>          Issue Type: Bug
>    Affects Versions: 1.3.0
>            Reporter: Sumit Mohanty
>            Assignee: Sumit Mohanty
>             Fix For: 1.3.0
>
>         Attachments: AMBARI-1994.patch
>
>
> FOR DATANODE:
> 1. Add DataNode component to a host
> 2. This should also:
> - setup Nagios alerts for that host for "datanode process down" and "datanode storage
full"
> - on namenode, Nagios alerts for datanodes should be adjusted "Percent DataNodes down"
and "percent DataNodes storage full"
> FOR TASKTRACKER:
> 1. Add TaskTracker component to a host
> 2. This should also:
> - setup Nagios alert for that host for "TaskTracker process down"
> - on namenode, Nagios alerts for tasktrackers should be adjusted "Percent TaskTracker
down"
> FOR REGIONSERVER:
> 1. Add RegionServer component to a host
> 2. This should also:
> - setup Nagios alert for that host for "RegionServer process down"
> - on namenode, Nagios alerts for regionservers should be adjusted "Percent RegionServers
down"

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message