[
https://issues.apache.org/jira/browse/KAFKA-1723?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14263180#comment-14263180
]
Jay Kreps commented on KAFKA-1723:
----------------------------------
The group/tags approach seems reasonable. I left some comments on the rb.
Questions: Can we define what group and tags in a way outside of JMX? Are these
required fields for all metrics? What is the data model of various metric
collection systems (graphite or whatever?) and would these map to that?
Should group be optional with some reasonable default? Should tags?
When reporting stats on lower-level facilities such as Selector or BufferPool
we often want to report these as associated with the producer. However we can't
add client id directly into the Selector or BufferPool and neither can we hard
code the group. Instead we should be passing in the group and tags in the
constructor.
> make the metrics name in new producer more standard
> ---------------------------------------------------
>
> Key: KAFKA-1723
> URL: https://issues.apache.org/jira/browse/KAFKA-1723
> Project: Kafka
> Issue Type: Sub-task
> Components: clients
> Affects Versions: 0.8.2
> Reporter: Jun Rao
> Assignee: Manikumar Reddy
> Priority: Blocker
> Fix For: 0.8.2
>
> Attachments: KAFKA-1723.patch
>
>
> The jmx name in the new producer looks like the following:
> kafka.producer.myclientid:type=mytopic
> However, this can be ambiguous since we allow "." in client id and topic.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
