[
https://issues.apache.org/jira/browse/KAFKA-1481?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14093822#comment-14093822
]
Vladimir Tretyakov commented on KAFKA-1481:
-------------------------------------------
Hi Jun Rao, I understood what you suggest, thx, but I am afraid it will not
work everywhere or I've missed something. Look at example from my local env:
1)
"kafka.producer":type="ProducerRequestMetrics",name="af-servers-AllBrokersProducerRequestRateAndTimeMs"
2)
"kafka.consumer":type="ZookeeperConsumerConnector",name="af-servers-af-servers-spm-new-cluster-topic-af-servers-wawanawna-Dell-1401353748289-fcaaea29-0-FetchQueueSize"
>From 1 I can parse "af-servers" as clientId, ok.
>From 2 I have to parse: topic ,groupId, consumer host, timestamp, uuid,
>num_streams.
If I remove "af-servers" from 2 I will get
"spm-new-cluster-topic-af-servers-wawanawna-Dell-1401353748289-fcaaea29-0-FetchQueueSize".
I can extract "timestamp, uuid, num_streams" without problems, now let me talk
about this part "spm-new-cluster-topic-af-servers-wawanawna-Dell" I see
"af-servers" in the middle but can I be sure that "clientId" will be always
here?
Why initial bean name contains "af-servers" 3 times? I have a feeling sometimes
all these 3 parts may be different and I will not able to parse everything in
right way.
It is painful in general as you mentioned above because we have to remember
"context" for parse each line, I am pretty sure nobody like it.
Even if somebody parse names in way you have described I think they will glad
to change its parse mechanism with more easiest.
> Stop using dashes AND underscores as separators in MBean names
> --------------------------------------------------------------
>
> Key: KAFKA-1481
> URL: https://issues.apache.org/jira/browse/KAFKA-1481
> Project: Kafka
> Issue Type: Bug
> Components: core
> Affects Versions: 0.8.1.1
> Reporter: Otis Gospodnetic
> Labels: patch
> Fix For: 0.8.2
>
> Attachments: KAFKA-1481_2014-06-06_13-06-35.patch
>
>
> MBeans should not use dashes or underscores as separators because these
> characters are allowed in hostnames, topics, group and consumer IDs, etc.,
> and these are embedded in MBeans names making it impossible to parse out
> individual bits from MBeans.
> Perhaps a pipe character should be used to avoid the conflict.
> This looks like a major blocker because it means nobody can write Kafka 0.8.x
> monitoring tools unless they are doing it for themselves AND do not use
> dashes AND do not use underscores.
> See: http://search-hadoop.com/m/4TaT4lonIW
--
This message was sent by Atlassian JIRA
(v6.2#6252)
