GitHub user uce opened a pull request:
https://github.com/apache/flink/pull/1578
[FLINK-3310] [runtime, runtime-web] Add back pressure statistics
This PR introduces back pressure statistics to the web frontend.
The idea is that tasks, which are back pressured by the network
Github user fhueske commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-179130238
Will this also be displayed / collected for batch jobs? Might this raise
questions / confuse users?
---
If your project is set up for it, you can reply to this email an
Github user uce commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-179138699
Yes, it's independent and relevant for both streaming and batch.
I think you are raising a valid point with your last question. In contrast
to the other information
Github user StephanEwen commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-179758751
Is it actually a bad thing that this is available for batch jobs? Batch
jobs have multiple pipelined tasks as well, and this would figure out where in
the pipeline t
Github user uce commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-179764192
It's definitely a good thing in my opinion. I was just asking whether we
should "hide" it behind an Advanced tab (with further advanced monitoring
features in the future).
Github user StephanEwen commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-179863296
We can still move the tab in the UI later.
I don't think there is a thing like "breaking changes" in the UI ;-)
---
If your project is set up for it, you can rep
Github user uce commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-180286078
Did anyone have time to check this out? Would appreciate it :-) Or do we
want to postpone this until after the 1.0 release?
---
If your project is set up for it, you can re
Github user StephanEwen commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-180365233
I would very much like to have this in the 1.0 release.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as wel
Github user StephanEwen commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-180374301
This looks pretty good all in all.
One thing I do not like that much is that the sampling process needs a
coordinator on the ExecutionGraph. A functionality
Github user uce commented on the pull request:
https://github.com/apache/flink/pull/1578#issuecomment-181127247
I've addressed the comments and moved the code to the runtime monitor,
which now starts the sample coordinator with an actor to receive the responses
(instead of going thoug
Github user asfgit closed the pull request at:
https://github.com/apache/flink/pull/1578
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enab
11 matches
Mail list logo