allthingssecurity opened a new pull request, #28196:
URL: https://github.com/apache/camel/pull/28196

   # Description
   
   [CAMEL-25790](https://issues.apache.org/jira/browse/CAMEL-25790)
   
   `GooglePubsubConsumer` runs one loop per concurrent consumer: the 
asynchronous loop starts a `Subscriber` and waits until it terminates, the 
synchronous loop pulls, routes the messages and pulls again. Both loops 
continued while `isRunAllowed() && !isSuspendingOrSuspended()`, which misses 
two cases:
   
   - Graceful shutdown. `deferShutdown` (CAMEL-23291) stops the subscribers and 
cancels the pending pulls, but the consumer stays started while the shutdown 
strategy waits for its pending exchanges. The loop saw a running consumer and 
at once started a new subscriber, or issued a new pull, so new messages kept 
being routed during the wait. With a steady flow of messages the shutdown 
waited for its full timeout (45 s by default) and was then forced, with 
exchanges in flight, whose messages are delivered again.
   - Route restart. `doStop` only calls `stopAsync`; the subscriber terminates 
later. When the consumer was started again before that, the old loop woke up in 
a started consumer and subscribed again, next to the loops of the new start. 
Every restart added `concurrentConsumers` subscriber loops, and the executors 
of the previous start stayed until `doShutdown`.
   
   This change gives each loop a generation. The consumer increments it when it 
starts, in `deferShutdown` and in `doStop`; a loop of an older generation ends 
instead of subscribing or pulling again. The checks right after a subscriber 
start and after a pull is issued (CAMEL-24315) use the same condition. A start 
after a stop also shuts down, not gracefully, the two executors of the previous 
start; their loops end on their own. `doShutdown` is unchanged. No message was 
lost before this change and no option or default changes, so there is no 
upgrade note.
   
   Tests:
   - New `GooglePubsubConsumerStopPullingTest` (3 tests, mocked `Subscriber` 
and `SubscriberStub` as in `GooglePubsubConsumerStopRaceTest`): no new 
subscriber after `deferShutdown` (asynchronous pull), no new pull after 
`deferShutdown` (synchronous pull), and one running subscriber after a stop and 
start while the first subscriber is still terminating. Each test checks with 
Awaitility `during` that the count stays the same for a second.
   - Without the change all three fail: `expected: <1> but was: <2>` 
(subscribers created after the consumer stopped pulling), `expected: <1> but 
was: <2>` (pulls issued after the consumer stopped pulling), `expected: <2> but 
was: <3>` (subscribers created by a start and a restart with one concurrent 
consumer).
   - With the change the camel-google-pubsub unit tests pass: 20 run, 0 
failures, 10 skipped (they need credentials or the emulator). The integration 
tests need the Pub/Sub emulator and were not run.
   
   How it was found: mutation testing showed the stop and shutdown paths of the 
consumer and the asynchronous pull untested; reading the subscriber loop 
against the shutdown strategy (a deferred consumer stays started until its 
pending exchanges are done) and against the intent of CAMEL-23291, then the 
JUnit tests.
   
   # Target
   
   - [x] I checked that the commit is targeting the correct branch (Camel 4 
uses the `main` branch)
   
   # Tracking
   - [x] If this is a large change, bug fix, or code improvement, I checked 
there is a [JIRA issue](https://issues.apache.org/jira/browse/CAMEL) filed for 
the change (usually before you start working on it).
   
   # Apache Camel coding standards and style
   
   - [x] I checked that each commit in the pull request has a meaningful 
subject line and body.
   - [ ] I have run `mvn clean install -DskipTests` locally from root folder 
and I have committed all auto-generated changes.
     (I built and tested camel-google-pubsub, including the formatter and 
import-sort plugins. I did not run the full root build.)
   
   # AI-assisted contributions
   
   - [x] If this PR includes AI-generated code, commits have proper 
co-authorship attribution (e.g., `Co-authored-by` trailers) and the PR 
description identifies the AI tool used.
     This PR was prepared with Claude Code (Claude Opus 5.5). The commit 
carries a `Co-Authored-By` trailer.
   
   _Claude Code on behalf of allthingssecurity_
   
   🤖 Generated with [Claude Code](https://claude.com/claude-code)
   


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to