ASB transport infinitely retries the incoming messages and logs large amount of exceptions

Customers using the new Azure Service Bus might encounter the following issue: when an endpoint is operating in SendsAtomicWithReceive transport transaction mode and is losing connectivity, a received message might get stuck on retries while the transport is trying to send the messages from the handler. We have investigated the issue and identified a problem with the underlying Azure Service Bus client. We’ve raised a GitHub issue with the messaging team and working with Microsoft on a resolution.

Meanwhile, if you are affected by this issue, we recommend one of the following workarounds:

  1. Restart the affected endpoints.
  2. Lower the transport transaction mode if your system allows.

Once the client issue is resolved, we’ll be updating and releasing the transport.

The cause of the issue has been identified (a bug in the underlying AMQP library).
Microsoft will be patching AMQP library, which will unblock ASB client and allow us to release a patch as well.

A new minor of the transport is out and it contains the fix for this issue.

Release announcement here.