[Starlingx-discuss] Does rabbitmq listener work in NFV module

[Liu, ZhipengS]

Hi Bart,

Thanks!  It exactly works in nfv.

I still have a question about Rabbitmq access.

Currently, I’m working on pci-interrupt-affinity service which will run on every worker node.
It need to listen to rabbitmq to get notifications from nova
Before containerized version, I can use configurations in /etc/sysinv/sysinv.conf to get connect to rabbitmq.
Now, if I use below configuration, it can work.
/opt/platform/puppet/19.01/hieradata/system.yaml
nfv::nfvi::platform_username: admin
nfv::nfvi::rabbit_host: rabbitmq.openstack.svc.cluster.local
nfv::nfvi::rabbit_password: 28a5834cf803Ti0*
nfv::nfvi::rabbit_port: 5672
nfv::nfvi::rabbit_userid: nova-rabbitmq-user
nfv::nfvi::rabbit_virtual_host: nova

Then, I tried to add below code to get above configuration. But it didn’t work.
“Utils.is_openstack_installed” this check failed! And also cannot get “helm_data”

Who can help?  Any comment or proposal on it?
BTW, can I get these configuration from
/opt/platform/puppet/19.01/hieradata/system.yaml directly? Is it reasonable?


===================================================================
from sysinv.helm import helm
from sysinv.common import utils
from sysinv.db import api as db_api
...

dbapi = db_api.get_instance()
if dbapi and utils.is_openstack_installed(dbapi):
    helm_data = helm.HelmOperatorData(dbapi)
    nova_oslo_messaging_data = helm_data.get_nova_oslo_messaging_data()
    rabbit_cfg['rabbit_host'] = nova_oslo_messaging_data['host']
    rabbit_cfg['rabbit_userid'] = nova_oslo_messaging_data['username']
    rabbit_cfg['rabbit_password'] = nova_oslo_messaging_data['password']
rabbit_cfg['rabbit_virtual_host'] = nova_oslo_messaging_data['virt_host']


Thanks!
Zhipeng

From: Wensley, Barton [mailto:Barton.Wensley at windriver.com]
Sent: 2019年4月17日 20:24
To: Liu, ZhipengS <zhipengs.liu at intel.com>; starlingx-discuss at lists.starlingx.io
Subject: RE: [Starlingx-discuss] Does rabbitmq listener work in NFV module

Zhipeng,

These logs are normal in cases where the containerized rabbitmq servers are temporarily unavailable (i.e. the osh-openstack-rabbitmq-rabbitmq-0/1 pods). This can happen when a controller is locked or rebooted. The VIM should automatically re-connect to the rabbitmq servers - this could take anywhere from a few seconds to as long as a minute or two depending on the reason the rabbitmq server(s) were not available.

Bart

From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com]
Sent: April 17, 2019 6:46 AM
To: starlingx-discuss at lists.starlingx.io<mailto:starlingx-discuss at lists.starlingx.io>
Subject: [Starlingx-discuss] Does rabbitmq listener work in NFV module

Hi all,

>From nfv-vim.log I see below
2019-04-12T08:38:11.409 controller-0                           VIM_Thread[31835]     INFO                            rpc_listener.py.127   RPC-Listener not connected to exchange nova, queue=notifications.nfvi_nova_listener_queue.

It seems NFV could not get notifications from nova now.
Have we enabled rabbitmq listener in NFV after containerized version?

Thanks!
zhipeng

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20190418/4a722722/attachment.html>