Repository navigation
Feature: AMQP event bus - buffer events while the broker is unreachable #14676
Description
The required feature described as a wish
I would like the RabbitMQ event bus to keep events that are published while the broker is unreachable and deliver them, in order, once the connection is back, within configurable limits (number of events and maximum
age).
As we tested during the network outage, every event that is published while the broker is unreachable is lost.
At the moment it works like that events are sent with without publisher confirms, so when the connection is down, the event is discarded.
The reconnect logic restores the connection only and nothing is replayed:
- when
publish()catchesAlreadyClosedExceptionit callscloseConnection().
That shutdown is application initiated, soDisconnectHandlerignores it and no
ReconnectionTaskis started. The nextpublish()then connects on the calling thread, which can block for up to the client's connection timeout if the host does not answer.
Another observation - the RabbitMQ client reports any connection close it did not initiate as "shutdown by AMQP server", including closes by a load balancer idle timeout.
Also it seems, the current code has no confirmation mechanism, so we never know the messages really reached the AMQP.
Proposal
- Keep events that cannot be sent in a bounded in-memory FIFO buffer and send
them in order after the reconnect. - Two optional properties on the existing event bus bean: maximum number of
buffered events (default 1000, oldest dropped first) and maximum age of a
buffered event (default 5 minutes). - Buffer only after the broker was reachable at least once; before that keep
today's behavior.
Activity
rabbitmq event bus: buffer events while the broker is unreachable
github-actions commentedon Nov 12, 2026
Warning
threat detection engine error
The threat detection engine encountered an error and could not complete analysis. This is a tooling failure, not a security finding.
What happened
The threat detection engine failed to produce results.
Review the workflow run logs for details.
🎯 Triage report
The request is for the RabbitMQ event bus to buffer events published while the broker is unreachable, in a bounded in-memory FIFO (default 1000 events, 5 minute maximum age), and to send them in order after reconnect. The reporter says events are currently dropped, because publishing is not confirmed and
publish()closes the connection without starting the reconnect task. A pull request (#14677) implementing a buffer is already open.📊 Assessment
🔗 Similar issues
💡 Notes and suggestions
AlreadyClosedExceptionpath that closes the connection without a reconnect, and publisher confirms. Those may need their own items if they are not covered by Rabbitmq eventbus buffer implementation #14677.Add this agentic workflow to your repo
To install this agentic workflow, run