[
https://issues.apache.org/jira/browse/BEAM-5299?focusedWorklogId=165648&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-165648
]
ASF GitHub Bot logged work on BEAM-5299:
----------------------------------------
Author: ASF GitHub Bot
Created on: 13/Nov/18 21:27
Start Date: 13/Nov/18 21:27
Worklog Time Spent: 10m
Work Description: aaltay commented on issue #7012: Revert "Revert "Revert
"[BEAM-5299] Define max timestamp for global w…
URL: https://github.com/apache/beam/pull/7012#issuecomment-438443510
The internal issue could have been better phrased as dataflow runner fails
to run with this change. I agree on that.
Post commit policies have content about how to handle post commit failures.
It does not explain on all conditions about how and why a PR revert should
happen.
If you think that discussing reverts on dev@ list would make sense, I would
suggest adding that to the documents. It is not clear to me, it is unlikely to
be clear to new comer.
> That doesn't make sense. Anyone could find a problem in any runner outside
of the project and cite that as reason to revert properly reviewed PRs that
passed post-commit. I wonder how that would be received if it wasn't Dataflow,
but any other runner..
If there is a user reported issue about a runner not working, I think it
makes sense to look at it. This is true for any runner. And in the past we did
these investigations at release time for runners other dataflow runner.
(example 2.7 release and investigations related to the spark runner.)
The question is, if we know that there is an issue, do we want to wait until
a later time (release time) to address it?
One clear improvement here would be to, with any such errors to augment Beam
tests to provide better coverage.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 165648)
Time Spent: 15h 10m (was: 15h)
> Define max global window as a shared value in protos like URN enums.
> --------------------------------------------------------------------
>
> Key: BEAM-5299
> URL: https://issues.apache.org/jira/browse/BEAM-5299
> Project: Beam
> Issue Type: Improvement
> Components: beam-model, sdk-go, sdk-java-core, sdk-py-core
> Reporter: Luke Cwik
> Assignee: Maximilian Michels
> Priority: Minor
> Labels: portability
> Fix For: 2.9.0
>
> Time Spent: 15h 10m
> Remaining Estimate: 0h
>
> Instead of having each language define a max timestamp themselves, define the
> max timestamps within proto to be shared across different languages.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
