[kolla][rabbitmq] Trouble with RMQ 3.8 in Train

Mark Goddard mark at stackhpc.com
Wed Feb 23 09:09:59 UTC 2022


You will need changes in kolla and kolla-ansible. You might try
comparing with ussuri, since we use 3.8 in ussuri.

Upgrading to ussuri and beyond would be (far) better still.

Mark

On Tue, 22 Feb 2022 at 21:17, Albert Braden <ozzzo at yahoo.com> wrote:
>
> We are running kolla-ansible Train and I successfully followed the recommendations in [1] to change our RMQ settings, and our RMQ works a lot better now, but one of the recommendations is to run RMQ 3.8 and so far I haven’t been able to make it work. Our working RMQ container has these settings in the Dockerfile to force 3.7 and the appropriate erlang version:
>
>             'erlang-21.3*',
>             'rabbitmq-server-3.7.23*'
>
> I changed those settings to allow 3.8:
>
>             'rabbitmq-server-3.*.*el8ost*'
>
> Now my RMQ container fails to start with:
>
> {"log":"20:29:28.440 [error] Supervisor rabbit_prelaunch_sup had child prelaunch started with rabbit_prelaunch:run_prelaunch_first_phase() at undefined exit with reason no_configuration_schema_found in context start_error\r\n","stream":"stdout","time":"2022-02-22T20:29:28.441064637Z"}
>
> What am I missing?
>
> [1] https://wiki.openstack.org/wiki/Large_Scale_Configuration_Rabbit
>



More information about the openstack-discuss mailing list