GeezerVison

@cincuranet Is there a chance to integrate this into upstream? Is there anything else I need to do?

I need to review the PR first. Currently busy.

@Howaner do you have a reproductible test case which shows the problem?

Maybe adding a test case to this PR?

@fdcastel It is a race condition and therefore quite difficult to trigger. I tried to reproduce it in a test project today but I failed.
The race condition occurs when using the Firebird library in an application where many things happen simultaneously.
Our application crashed every few weeks due to this bug and since we changed this few lines, the deadlock has no longer occurred.

I'm really surprised that the bug doesn't already occur much more often with other people. Probably only a few people make many simultaneous requests with Firebird.

I understand. Surely they are a hell to reproduce.

Thank you for you feedback. Much appreciated.

I too have started hitting this issue in our test projects after updating to version 9.1.1.0. I'll try and see if I can write a minimal reproducible test case.

@willibrandon, @Howaner Have you had any success reproducing the issue?

@Howaner, if I understand correctly, you've been running a fork of this project in production with this PR applied. Have you never encountered any further errors since applying it?

@willibrandon, would you be able to try the same approach?

While I understand the changes @Howaner made in this PR, it's challenging to approve them based solely on verbal explanations (no offense intended, please). As engineers, I trust you both will understand.

We saw this issue pop up in our test runs after updating the data provider version for the first time in a decade. I was not successful reproducing this issue however, and have not seen the it reoccur since.

@PaladinMattt - Do you recall the last time you saw this one?

It has been some time since I've noticed it. If my memory serves me correctly, it mainly happened during debugging, when calling FbConnection.ClearAllPools();, but I don't recalling seeing this issue for a few months now.

I'm looking at the PR now, with the help of various AI models, and I'm not convinced the fix fixes what it claims to fix. Moreover it also slightly changes the ordering of concurrent operations and that brings other new questions.

I'm not saying the bug is not there, far from it, but a dump when the deadlock happens would help to pinpoint the exact constellation of threads, etc.