[rabbitmq-discuss] RabbitMQ second-time message delivery after service restart

Thilo-Alexander Ginkel thilo at ginkel.com
Mon Nov 25 11:46:58 GMT 2013


On Mon, Nov 25, 2013 at 12:05 PM, Michael Klishin <
michael.s.klishin at gmail.com> wrote:

> 2013/11/25 Thilo-Alexander Ginkel <thilo at ginkel.com>
>
>> AFAIK, all messages had been properly ACKed before
>
>
> Does management UI or rabbitmqctl confirm this?
>

I don't know the exact situation during the weekend, but when restoring a
(one-week-old) backup [1] into a different VM (and tweaking the hostnames
so that RabbitMQ starts up), I see zero unACKed messages (and that the
queue actually contained > 500.000 re-surrected messages, not 250.000 as I
originally stated):

# rabbitmqctl list_queues name messages_ready messages_unacknowledged
Listing queues ...
[...]
psm_subscriber_api_production   520111  0
[...]

BTW, the same issue also happens with RabbitMQ 3.2.1 when using the backed
up data (just to rule out we are hitting a bug that has been fixed a long
time ago).

Thanks,
Thilo

[1] restoring a newer version is underway
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20131125/829487cc/attachment.htm>


More information about the rabbitmq-discuss mailing list