[
https://issues.apache.org/jira/browse/AMBARI-7284?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14131898#comment-14131898
]
Matt Foley commented on AMBARI-7284:
------------------------------------
This is probably an issue with sys_logger.py. See the previous-generation
issue AMBARI-4435, which provided this functionality for Ambari 1.5.0 and
Hadoop-1.x.
> Hadoop cluster alerts have not been updated for Hadoop 2.4 and 2.5
> ------------------------------------------------------------------
>
> Key: AMBARI-7284
> URL: https://issues.apache.org/jira/browse/AMBARI-7284
> Project: Ambari
> Issue Type: Bug
> Affects Versions: 1.6.0
> Environment: Hadoop branch-2, release 2.4 or 2.5.
> Reporter: Nicholas Yao
>
> Many /var/log/message alerts we keyed off of previously are no longer working
> or valid. It appears that many hadoop 1.x terms such as jobtracker,
> tasktracker and templeton still exist, when Ambari is being used with hadoop
> 2.x.
> I believe existing rules need to be modified for the following service name
> changes:
> resourcemanager_process_down
> resourcemanager_process_down_ok
> resourcemanager_rpc_latency
> resourcemanager_rpc_latency_ok
> resourcemanager_cpu_utilization
> resourcemanager_cpu_utilization_ok
> nodemanagers_down
> nodemanagers_down_ok
> nodemanager_process_down
> nodemanager_process_down_ok
> webhcat_down
> webhcat_down_ok
> It also appears that existing messages are getting improperly matched as we
> see the following HADOOP_UNKNOWN_MSG in /var/log/messages:
> Jul 15 10:36:34 pitH1 nagios[35331]: Warning: Hadoop: HADOOP_UNKNOWN_MSG#
> Event Host=pitH1.td.teradata.com Service Description=HDFS::Percent DataNodes
> with space available(WARNING), WARNING: total:6, affected:1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
