fix potential race conditions in ModelObjectPool

[mksergiy]

This PR fixes #123

---

Rebus is MIT-licensed. The code submitted in this pull request needs to carry the MIT license too. By leaving this text in, I hereby acknowledge that the code submitted in the pull request has the MIT license and can be merged with the Rebus codebase.

[CLAassistant]

All committers have signed the CLA.

[mookid8000]

Hi @mksergiy , thanks for contributing!

It looks like a couple of tests are failing - did you try to run the tests locally? and did they work?

[mksergiy]

Hi @mksergiy , thanks for contributing!

It looks like a couple of tests are failing - did you try to run the tests locally? and did they work?

Yes, I ran the tests locally and didn’t find any issues. Today, I decided to double-check, and while the tests do run successfully, they fail in debug mode in my environment. After some investigation, I discovered that the main problem is the slow publishing of messages in parallel to the RabbitMQ queue when PublisherConfirms = true.

Since the failing test (CheckMaximumDegreeOfParallelism) is designed to test parallel message processing rather than message publishing, I recommend setting SetPublisherConfirms(false) for the RabbitMQ transport. This resolved the test failures for me in debug mode.

[mookid8000]

Excellent work 🙂 👍

It's out as Rebus.RabbitMq.9.4.1. on NuGet.org now

Thanks for contributing!