I wonder if the limitation of forced durable queues will make things too slow? Right now I believe it’s recommended to not use queues in durable mode with openstack (if I recall correctly)?

Also, it does add to the setup side of things, as it would require explicitly creating each vhost in Pulsar ahead of deployment, but that might not matter if people don’t use a ton of vhosts (we use one per service, but that’s not actually too many I suppose, and certainly isn’t required). I guess a lot of deployments may only use a single vhost?

On Mon, Feb 14, 2022 at 16:21 liyang <ly2008080808@yeah.net> wrote:
I think if we deploy Apache pulsar with AOP in cluster & HA, the effect will be ok. I am going to ready practise this schema.

在 2022-02-14 15:18:25,"liyang" <ly2008080808@yeah.net> 写道:

Hi, Thanks to rely. I found this link should be very helpful to us. https://streamnative.io/en/blog/tech/2020-06-15-announcing-aop-on-pulsar/ .

在 2022-02-14 14:55:55,"Adam Harwell" <flux.adam@gmail.com> 写道:

Hey! It’s cool to see someone else talking about this. My team considered this, and personally I think it would work very well, we just didn’t have the development resources to throw at the problem on our own. I think it’s likely that we could participate in a larger discussion about it though, and possibly help here and there if someone else was driving it.

   —Adam

On Mon, Feb 14, 2022 at 11:59 liyang <ly2008080808@yeah.net> wrote:
Hi  everyone,
    I find Apache Pulsar is very popular in MQ area, especially in clound native area. Is there any body integrate this MQ to oslo.messaging?
    Said to the truth. RabbitMQ developed in Erlang has very little source code level material, Qpid has very little operation practice in the internet. Kafka only support notification in oslo.messaging.
    I think the Apache Pulsar  is clound native based. It may enforece the big scale in extension about Openstack deployment.


 

--
Thanks,
    --Adam


 



 

--
Thanks,
    --Adam