You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Dawid Wysakowicz (Jira)" <ji...@apache.org> on 2020/01/10 12:19:00 UTC

[jira] [Commented] (FLINK-15431) Add numLateRecordsDropped/lateRecordsDroppedRate/watermarkLatency in CepOperator

    [ https://issues.apache.org/jira/browse/FLINK-15431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17012783#comment-17012783 ] 

Dawid Wysakowicz commented on FLINK-15431:
------------------------------------------

Hi [~libenchao],

I think it makes sense to add {{numLateRecordsDropped}}.
I would not add {{lateRecordsDroppedRate}} this is a derivative of the upper metric. Imho, it is a good practice to calculate such metrics on the receiver side rather than inside of an operator.

I think this could be helpful as a general metric for all operators. I would open a separate issue to add such metric to a {{StreamOperator}} rather than duplicate that logic in multiple operators. It has no CEP specific logic.

> Add numLateRecordsDropped/lateRecordsDroppedRate/watermarkLatency in CepOperator
> --------------------------------------------------------------------------------
>
>                 Key: FLINK-15431
>                 URL: https://issues.apache.org/jira/browse/FLINK-15431
>             Project: Flink
>          Issue Type: Improvement
>          Components: Library / CEP
>            Reporter: Benchao Li
>            Assignee: Benchao Li
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> I find that current CepOperator has no metric indicates numLateRecordsDropped/lateRecordsDroppedRate/watermarkLatency which is available in WindowOperator.
> It is inconvenient  for users to debugging and monitoring. So I propose to add these metrics in CepOperator too.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)