Message retried indefinitely due to unknown delivery tag error in RabbitMQ

[simongullberg]

Hi,

A few days ago, we encountered an issue that we haven't experienced before. A specific message was processed multiple times, and the exceptions we observed were as follows:

RabbitMQ.Client.Exceptions.AlreadyClosedException: Already closed: The AMQP operation was interrupted: AMQP close-reason, initiated by Peer, code=406, text='PRECONDITION_FAILED - unknown delivery tag 5', classId=60, methodId=80
   at RabbitMQ.Client.Impl.SessionBase.Transmit(OutgoingCommand& cmd)
   at RabbitMQ.Client.Framing.Impl.Model.BasicAck(UInt64 deliveryTag, Boolean multiple)
   at Rebus.RabbitMq.RabbitMqTransport.<>c__DisplayClass55_0.<<Receive>b__0>d.MoveNext()
--- End of stack trace from previous location ---
   at Rebus.Transport.TransactionContext.InvokeAsync(Func`2 actions)
   at Rebus.Transport.TransactionContext.Complete()
   at Rebus.Workers.ThreadPoolBased.ThreadPoolWorker.ProcessMessage(TransactionContext context, TransportMessage transportMessage)

and

An error occurred when attempting to complete the transaction context

System.NullReferenceException: Object reference not set to an instance of an object.
   at Rebus.RabbitMq.RabbitMqTransport.<>c__DisplayClass55_0.<<Receive>b__0>d.MoveNext()
--- End of stack trace from previous location ---
   at Rebus.Transport.TransactionContext.InvokeAsync(Func`2 actions)
   at Rebus.Transport.TransactionContext.Complete()
   at Rebus.Workers.ThreadPoolBased.ThreadPoolWorker.ProcessMessage(TransactionContext context, TransportMessage transportMessage)

With each message redelivery, a new MessageId was generated, which leads me to believe the error tracker was unable to accurately track the error count. As a result, the message was retried indefinitely.

Is this a known issue, or do you have any insights into what might be causing this behavior?

Thank you!