Fix perpetual auto-recovery problem

[mot256]

Proposed Changes

We have found an issue where if a consumer of an auto-delete queue is cancelled, the auto-recovery will fail perpetually.

After debugging we found that, when a consumer of an auto-delete queue is cancelled, the recording of the queue is removed from the auto-recovery list, but the associated bindings were not removed from the recording. Leading to a perpetual error stating that the binding cannot be done because the queue does not exist.

Types of Changes

What types of changes does your code introduce to this project?
Put an x in the boxes that apply

• Bug fix (non-breaking change which fixes issue #NNNN)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause an observable behavior change in existing systems)
• Documentation improvements (corrections, new content, etc)
• Cosmetic change (whitespace, formatting, etc)

Checklist

Put an x in the boxes that apply. You can also fill these out after creating
the PR. If you're unsure about any of them, don't hesitate to ask on the
mailing list. We're here to help! This is simply a reminder of what we are
going to look for before merging your code.

• I have read the CONTRIBUTING.md document
• I have signed the CA (see https://cla.pivotal.io/sign/rabbitmq)
• All tests pass locally with my changes
• I have added tests that prove my fix is effective or that my feature works
• I have added necessary documentation (if appropriate)
• Any dependent changes have been merged and published in related repositories

Further Comments

You will also note that I fixed inconsistent locking that could lead to further auto-recovery issues due to _recordedBindings variable not being locked with the same lock while reading and writing.

[michaelklishin]

Your description of the problem and proposed solution make sense to me.

[michaelklishin]

@mot256 thank you. Can you please submit the same change against main?

[michaelklishin]

I cherry-picked this to 7.x but main uses file-scoped namespaces, so cherry-picking cannot apply cleanly. It may be more practical to submit the same PR for main.

[mot256]

@mot256 thank you. Can you please submit the same change against main?

Will do. Thanks.

[mot256]

@michaelklishin when are you planning to release 3.6.1 that will include this fix? We are having issues in production that are directly linked to this issue.

[michaelklishin]

I cannot make any ETA promises but we can try to ship a patch release this week.

[lukebakken]

Sorry it didn't seem that urgent. I'll get a release done within the next few hours.

[lukebakken]

@mot256 - https://www.nuget.org/packages/RabbitMQ.Client/6.3.1

I'm getting the release details sorted out here right now. Thanks again for your contribution.