Do not join @thread if that is the current Thread #590
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
If you're running multiple Bunny connections in a pooled manner, for example using the gem `connection_pool`, the threading model can be altered by the pooling gem to disable handling of exceptions.
For example, the gem `connection_pool` disables the handling when it creates an instance, and enables the handling when you're using a connection:
```ruby
def with(options = {})
Thread.handle_interrupt(Exception => :never) do
conn = checkout(options)
begin
Thread.handle_interrupt(Exception => :immediate) do
yield conn
end
ensure
checkin
end
end
end
```
When you've setup a connection and released it to the pool, it slumbers in a Thread of which the handling of exceptions is disabled. If you receive a (forced) disconnect or a NetworkError occurs, it triggers the recovery (when enabled) of Bunny. This recovery relies on a ShutdownSignal to be raised on the `reader_loop` thread which needs to be caught by the `clean_up_on_shutdown` in `Session`. Because the exception is not handled anymore, the code in `maybe_shutdown_reader_loop` in `Session` tries to join the `reader_loop` thread to force a re-raise of a thrown unhandled exception. That join is done on the `reader_loop` thread which is the same thread that started the recovery so we'll receive a ThreadError telling us "Target thread must not be current thread (ThreadError)".
To prevent the above from happening we can check if the thread we're trying to join is the same thread that is currently running and not join if it is the same thread.
michaelklishin
approved these changes
Feb 5, 2020
|
Backported to |
|
@michaelklishin: just as an update, we've encountered no irregularities during testing. We did see correctly recovering connections. |
|
OK, I'll cut a new release now. Thanks for following up. |
michaelklishin
added a commit
that referenced
this pull request
Feb 11, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
If you're running multiple Bunny connections in a pooled manner, for example using the gem
connection_pool, the threading model can be altered by the pooling gem to disable handling of exceptions.For example, the gem
connection_pooldisables the handling when it creates an instance, and enables the handling when you're using a connection:When you've setup a connection and released it to the pool, it slumbers in a Thread of which the handling of exceptions is disabled. If you receive a (forced) disconnect or a NetworkError occurs, it triggers the recovery (when enabled) of Bunny. This recovery relies on a ShutdownSignal to be raised on the
reader_loopthread which needs to be caught by theclean_up_on_shutdowninSession. Because the exception is not handled anymore, the code inmaybe_shutdown_reader_loopinSessiontries to join thereader_loopthread to force a re-raise of a thrown unhandled exception. That join is done on thereader_loopthread which is the same thread that started the recovery so we'll receive a ThreadError telling us "Target thread must not be current thread (ThreadError)".To prevent the above from happening we can check if the thread we're trying to join is the same thread that is currently running and not join if it is the same thread.
This PR fixes #589.