[
https://issues.apache.org/jira/browse/FLINK-31358?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17697433#comment-17697433
]
Martijn Visser commented on FLINK-31358:
----------------------------------------
I'm actually not sure that this information is even captured in metrics, which
would be a first requirement before being able to display it in the UI
> Display consumed split/partition/queue info on Web UI
> -----------------------------------------------------
>
> Key: FLINK-31358
> URL: https://issues.apache.org/jira/browse/FLINK-31358
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Web Frontend
> Reporter: Zhanghao Chen
> Priority: Minor
>
> Many data sources have the concept of "split", which is a partition of the
> whole data (e.g. partition in Kafka, queue in RocketMQ), and each Flink
> source task is allocated with a subset of splits to consume. When a job is
> lagging on only a few splits, it would be useful for determining whether it
> is a data source issue or a Flink issue if users can view which source task
> consumes which split on Web UI.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
