You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Hang Ruan (Jira)" <ji...@apache.org> on 2022/11/17 11:03:00 UTC
[jira] [Comment Edited] (FLINK-29801) OperatorCoordinator need open the way to operate metricGroup interface
[ https://issues.apache.org/jira/browse/FLINK-29801?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17635287#comment-17635287 ]
Hang Ruan edited comment on FLINK-29801 at 11/17/22 11:02 AM:
--------------------------------------------------------------
Hi, [~MengYue] ,
I may rely on this feature to accomplish https://issues.apache.org/jira/browse/FLINK-21000. In the feature https://issues.apache.org/jira/browse/FLINK-21000, I provide a POC which contains some parts about the metric group in OperatorCoordinator.
With these research, I have some question about this PR:
# How should we use the operator coordinator metric group when implement the OperatorCoordinator? For example, we always get the metric group from one context in the SourceCoordinator. Why not pass the metric group in its context?
# Is there some problems to use the JobManagerJobMetricGroup? There may be serval operator coordinators in the same job. It seems that every operator coordinator should make sure that its metric group does not conflict with others.
ps: It is my POC(https://github.com/ruanhang1993/flink/commit/dc41d3ab465d03e2829e5f1d10137ea34621ad87). I am very glad to help if you need.
Best, Hang
was (Author: ruanhang1993):
Hi, [~MengYue] ,
I may rely on this feature to accomplish https://issues.apache.org/jira/browse/FLINK-21000. In the feature https://issues.apache.org/jira/browse/FLINK-21000, I provide a POC which contains some parts about the metric group in OperatorCoordinator.
With these research, I have some question about this PR:
# How should we use the operator coordinator metric group when implement the OperatorCoordinator? For example, we always get the metric group from one context in the SourceCoordinator. Why not pass the metric group in its context?
# Is there some problems to use the JobManagerJobMetricGroup? There may be serval operator coordinators in the same job. It seems that every operator coordinator should make sure that its metric group does not conflict with others.
Thanks, Hang
> OperatorCoordinator need open the way to operate metricGroup interface
> ----------------------------------------------------------------------
>
> Key: FLINK-29801
> URL: https://issues.apache.org/jira/browse/FLINK-29801
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Metrics
> Reporter: yuemeng
> Assignee: yuemeng
> Priority: Major
> Labels: pull-request-available
>
> Currently, We have no way to get metric group instances in OperatorCoordinator
> In some cases, we may report some metric in OperatorCoordinator such as Flink hudi integrate scene， some meta will send to operator coordinator to commit to hdfs or hms
> but we also need to report some metrics in operator coordinator for monitor purpose
--
This message was sent by Atlassian Jira
(v8.20.10#820010)