[
https://issues.apache.org/jira/browse/FLINK-40668?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Rui Fan resolved FLINK-40668.
-----------------------------
Fix Version/s: 2.4.0
Resolution: Fixed
Merged to master(2.4.0) via: 345f99c094ca87317a73311ddeb2f4e825caeb98
> Document why the recovery drain cannot stall on zero-credit needsRecovery
> channels
> ----------------------------------------------------------------------------------
>
> Key: FLINK-40668
> URL: https://issues.apache.org/jira/browse/FLINK-40668
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Checkpointing
> Reporter: Rui Fan
> Assignee: Rui Fan
> Priority: Major
> Labels: pull-request-available
> Fix For: 2.4.0
>
>
> The drain awaits `upstreamReady`, which opens only on `onBuffer()`; a
> needsRecovery channel starts at zero server credit, so only an upstream event
> can open it. This is safe because recovery init always sends
> `EndOfOutputChannelStateEvent` downstream, but the reasoning is non-obvious —
> add a comment making it explicit. Low priority.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
