[ops] RabbitMQ Config suggestions for > 100 nodes?

Sean Mooney smooney at redhat.com
Fri Apr 17 12:41:26 UTC 2020


On Fri, 2020-04-17 at 14:14 +0200, Thomas Goirand wrote:
> On 4/17/20 9:02 AM, Fabian Zimmermann wrote:
> > As next try I would add some haproxy-lb and change the transport_url to
> > the VIP, this should afaik reduce the amount of connections to 1/3, but
> > I dont know if there are any drawbacks jet.
> 
> To my experience, you really don't need to do this, oslo.messaging takes
> care of reconnecting to another node if one fails. Though you may want
> to randomize the order in transport_url. If you use puppet, you can use
> the fqdn_rotate() function to achieve it.
i think that that behavior is consider deprecated.
at least i recall converstaion in nova irc channel before about not advising operators to
rely on that for loadblancing and instead suggesting they take the ha proxy approch.
> 
> Cheers,
> 
> Thomas Goirand (zigo)
> 




More information about the openstack-discuss mailing list