[
https://issues.apache.org/jira/browse/BEAM-11779?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17549199#comment-17549199
]
Danny McCormick commented on BEAM-11779:
----------------------------------------
This issue has been migrated to https://github.com/apache/beam/issues/20806
> Dataflow runner v2 tests occasionally fails with misconfigured sharding logic
> -----------------------------------------------------------------------------
>
> Key: BEAM-11779
> URL: https://issues.apache.org/jira/browse/BEAM-11779
> Project: Beam
> Issue Type: Bug
> Components: runner-dataflow
> Reporter: Yichi Zhang
> Priority: P3
> Time Spent: 9h
> Remaining Estimate: 0h
>
> Dataflow runner v2 test sometimes mysteriously fails data loss error message.
> This is due to misconfigured sharding logic from backend with combination of
> shuffle_mode=service and use_runner_v2 logic.
> Example failing pipeline:
> https://pantheon.corp.google.com/dataflow/jobs/us-central1/2021-02-08_10_20_51-4474446822883759937?project=apache-beam-testing
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
