[rabbitmq-discuss] RabbitMQ hanging during 'rabbitmqctl stop' - please help

[Casey Marshall]

On 07/30/2013 06:12 AM, Simon MacMullen wrote:
> On 26/07/13 17:45, Casey Marshall wrote:
>> On 07/26/2013 05:30 AM, Simon MacMullen wrote:
>>> Are you restarting the server immediately after reconfiguring
>>> federation, or some time later?
>>>
>> I'm restarting the server immediately after, on each federated node, in
>> parallel with Ansible.
> 
> Ah - you do know you don't need to restart for the changes to federation
> to take effect?
> 

Thanks for clarifying. I was the impression that this was necessary from
another source.

Today I confirmed that a service restart is in fact not required. Well,
now I feel a bit silly, but also relieved.

>> I suspect something about the other servers' upstream exchange
>> connection attempts, *during* the restart of a server, might have
>> something to do with it.
> 
> Yes - the deadlock we've found requires a (client) connection or channel
> to be opening at just the wrong point during shutdown.
> 
>> Is there a workaround I might try to avoid the deadlock?
> 
> You are much less likely to hit it if you wait a few seconds before
> restarting.
> 
> Hopefully we should have a fixed version available for test soon.
> 

That's great, I'll look out for it. I certainly have a repeatable test
environment :)

> Cheers, Simon
>