[
https://issues.apache.org/jira/browse/HADOOP-4482?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tom White updated HADOOP-4482:
------------------------------
Attachment: hadoop-4482.patch
The problem with the previous patch was that the JMX object name had an illegal
':' character in it, and this error was being masked by MBeanUtil which logs
exceptions, but doesn't throw them. This patch replaces the ':' with a '-', and
I verified that TestDataNodeMetrics doesn't log an exception. I've also opened
HADOOP-5237 to address the exception logging (and handling).
> Hadoop JMX usage makes Nagios monitoring impossible
> ---------------------------------------------------
>
> Key: HADOOP-4482
> URL: https://issues.apache.org/jira/browse/HADOOP-4482
> Project: Hadoop Core
> Issue Type: Improvement
> Components: metrics
> Affects Versions: 0.18.1
> Reporter: Brian Bockelman
> Assignee: Brian Bockelman
> Fix For: 0.19.1, 0.20.0
>
> Attachments: hadoop-4482.patch, jmx_name.patch,
> jmx_name_replaced.patch
>
>
> When Hadoop reports Datanode information to JMX, the bean uses the name
> "DataNode-" + storageid. The storage ID incorporates a random number and is
> unpredictable.
> This prevents me from monitoring DFS datanodes through Hadoop using the JMX
> interface; in order to do that, you must be able to specify the bean name on
> the command line.
> The fix is simple, patch will be coming momentarily. However, there was
> probably a reason for making the datanodes all unique names which I'm unaware
> of, so it'd be nice to hear from the metrics maintainer.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
