[
https://issues.apache.org/jira/browse/BEAM-7785?focusedWorklogId=282754&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-282754
]
ASF GitHub Bot logged work on BEAM-7785:
----------------------------------------
Author: ASF GitHub Bot
Created on: 25/Jul/19 16:33
Start Date: 25/Jul/19 16:33
Worklog Time Spent: 10m
Work Description: je-ik commented on issue #9108: [BEAM-7785] synchronize
watermark update with bundle processing
URL: https://github.com/apache/beam/pull/9108#issuecomment-515115042
I think I can imagine that, but it would require linking each KeyedWorkItem
with respective watermark snapshot. It can be done, but I fear it might be a
major change. I have also successfully written test case which exploits the
edge case of the integration test and fails. The issue is related to the fact
that the naive locking limited parallelism of processing to 1, which was not
intended and was a clear bug. It resulted in deadlock, when one bundle was
blocking all the others. I don't have working code yet, but I think it would be
possible to overcome this with more clever locking. WDYT?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to 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: 282754)
Time Spent: 2h 50m (was: 2h 40m)
> DirectRunner: watermarks are updated asynchronously from bundle processing
> --------------------------------------------------------------------------
>
> Key: BEAM-7785
> URL: https://issues.apache.org/jira/browse/BEAM-7785
> Project: Beam
> Issue Type: Bug
> Components: runner-direct
> Affects Versions: 2.13.0
> Reporter: Jan Lukavský
> Assignee: Jan Lukavský
> Priority: Major
> Fix For: 2.15.0
>
> Time Spent: 2h 50m
> Remaining Estimate: 0h
>
> Watermarks are updated in QuiescenceDriver (by calling fireTimers, which
> calls forceRefresh()) on WatermarkManager. This results in creating timer
> bundles, that are then processed asynchronously as DirectTransformExecutor.
> Because of that, watermarks (input watermarks mostly) might be updated while
> bundle is being processed. That violates assumption, that bundle processing
> should be atomical (with identical external conditions during processing of
> whole bundle).
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
