You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Zhu Zhu (Jira)" <ji...@apache.org> on 2022/07/05 09:47:00 UTC

[jira] [Created] (FLINK-28402) Create FailureHandlingResultSnapshot with the truly failed execution

Zhu Zhu created FLINK-28402:
-------------------------------

             Summary: Create FailureHandlingResultSnapshot with the truly failed execution
                 Key: FLINK-28402
                 URL: https://issues.apache.org/jira/browse/FLINK-28402
             Project: Flink
          Issue Type: Sub-task
          Components: Runtime / Coordination
            Reporter: Zhu Zhu
             Fix For: 1.16.0


Previously, FailureHandlingResultSnapshot was always created to treat the only current attempt of an execution vertex as the failed execution. This is no longer right in speculative execution cases, in which an execution vertex can have multiple current executions, and any of them may fail.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)