[
https://issues.apache.org/jira/browse/FLINK-3274?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15112273#comment-15112273
]
ASF GitHub Bot commented on FLINK-3274:
---------------------------------------
GitHub user rmetzger opened a pull request:
https://github.com/apache/flink/pull/1541
[FLINK-3274] Make accumulator names of Kafka connector unique
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/rmetzger/flink flink3274
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/1541.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #1541
----
commit 9779a06cb5b16aedd84a04147b5eb24a593314a9
Author: Robert Metzger <[email protected]>
Date: 2016-01-22T10:10:08Z
[FLINK-3274] Make accumulator names of Kafka connector unique
----
> Prefix Kafka connector accumulators with unique id
> --------------------------------------------------
>
> Key: FLINK-3274
> URL: https://issues.apache.org/jira/browse/FLINK-3274
> Project: Flink
> Issue Type: Bug
> Components: Kafka Connector
> Affects Versions: 1.0.0
> Reporter: Robert Metzger
> Assignee: Robert Metzger
>
> Flink might chain together two data sinks under certain conditions.
> In that case, the metrics accumulators will have a name clash (each task can
> have an accumulator only once, with chaining multiple sinks are sharing a
> task).
> Also, when merging accumulators from different tasks for a per-job result,
> having multiple kafka producers or consumers will lead to wrong results as
> the merging happens only on a per-name basis.
> With this issue, I'd like to introduce a unique id for consumers and
> producers when registering accumulators
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
