[
https://issues.apache.org/jira/browse/KAFKA-585?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13491895#comment-13491895
]
Joel Koshy commented on KAFKA-585:
----------------------------------
The CSV reporter basically does not work if the file already exists - i.e., it
doesn't try to open it in append mode. I haven't heard back from metrics-user
list on that, but I think it is sufficient for now to just delete the CSV dir
when starting up.
> Remove custom metrics jar and replace with latest from metrics HEAD
> -------------------------------------------------------------------
>
> Key: KAFKA-585
> URL: https://issues.apache.org/jira/browse/KAFKA-585
> Project: Kafka
> Issue Type: Bug
> Affects Versions: 0.8
> Reporter: Joel Koshy
> Attachments: KAFKA-585-v1.patch,
> metrics-annotation-3.0.0-c0c8be71.jar, metrics-core-3.0.0-c0c8be71.jar
>
>
> This is for at least until metrics 3.x is mavenized.
> Also:
> The KafkaCSVMetricsReporter object may be better named as
> KafkaMetricsReporter since startCSVMetricsReporter
> potentially starts up other (non-CSV) reporters (if any) as well - in which
> case KafkaMetricsReporter.scala would be a
> better place for it. Or, you can just filter out non-CSV reporters.
> Also, the top-level/config/server.properties need not enable the csv
> reporter. I thought the system test replication
> suite's server.properties would need to be patched, but it isn't. Should look
> into whether the test suite picks up the top-level
> config as a template.
--
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