[
https://issues.apache.org/jira/browse/FLINK-3669?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15262485#comment-15262485
]
Konstantin Knauf commented on FLINK-3669:
-----------------------------------------
[[~aljoscha] Thanks, will do the next time.
I tried to reproduce the issue locally with a small example [1] and got some
unexpected behaviour. In the example I would not expect any output, as the
processing timers are deleted way before they would trigger, but still (at
least in my IDE) many windows are fired. As this only happens in "high
throughput" scenarios I guess it is some scheduling problem, i.e. the thread
which deletes the timers is not scheduled before the timer is triggered. What
do you think?
[1] https://gist.github.com/knaufk/8870f4e7f59ed9586666c579d1d2344c
> WindowOperator registers a lot of timers at StreamTask
> ------------------------------------------------------
>
> Key: FLINK-3669
> URL: https://issues.apache.org/jira/browse/FLINK-3669
> Project: Flink
> Issue Type: Bug
> Components: Streaming
> Affects Versions: 1.0.1
> Reporter: Aljoscha Krettek
> Assignee: Konstantin Knauf
> Priority: Blocker
>
> Right now, the WindowOperator registers a timer at the StreamTask for every
> processing-time timer that a Trigger registers. We should combine several
> registered trigger timers to only register one low-level timer (timer
> coalescing).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
