[GH] (pekko-persistence-r2dbc/persistence-journal-batching): Workflow run "Headers" is working again!

[GitBox <gi...@apache.org>]

The GitHub Actions job "Headers" on pekko-persistence-r2dbc.git/persistence-journal-batching has succeeded.
Run started by GitHub user minjibir (triggered by minjibir).

Head commit for run:
2ba2161f48df8afef738ddfddd01beebe3529f63 / Jabir Minjibir <ja...@gmail.com>
fix: address review nits on batched write journal (#489)

Motivation:
A stale FlushDone from a previous incarnation matched no case and was logged as an unhandled dead letter. The publish warning logged only the exception message and lost the stack trace. The docs called the timestamp cursor per-actor when it is JVM-wide.

Modification:
Move the FlushDone generation check into the case body so stale messages are ignored silently. Log the publish exception as the warning cause. Reword the scaladoc and journal.md to say the stamps are strictly increasing across all journal actor instances in the JVM.

Result:
Stale FlushDone produces no dead-letter noise, publish failures keep their stack trace, and the docs match the actual timestamp scope.

Tests:
- sbt "core / Test / testOnly org.apache.pekko.persistence.r2dbc.journal.R2dbcBatchJournal*" - 79 passed, 0 failed, PostgreSQL 18.4
- sbt docs/paradox
- scalafmt --test on R2dbcBatchJournal.scala (3.11.5)
- git diff --check

References:
Refs #471, Refs #489

Report URL: https://github.com/apache/pekko-persistence-r2dbc/actions/runs/36621791082

With regards,
GitHub Actions via GitBox


---------------------------------------------------------------------
To unsubscribe, e-mail: notifications-unsubscribe@pekko.apache.org
For additional commands, e-mail: notifications-help@pekko.apache.org