[nova][neutron][cyborg] Bandwidth (and accel) providers are broken if CONF.host is set

Balázs Gibizer balazs.gibizer at est.tech
Fri Nov 22 11:18:54 UTC 2019



On Fri, Nov 22, 2019 at 02:11, "Nadathur, Sundar" 
<sundar.nadathur at intel.com> wrote:
>>  -----Original Message-----
>>  From: Eric Fried <openstack at fried.cc>
>>  Sent: Thursday, November 21, 2019 3:28 PM
> 
>>  Action Summary
>>  ==============
>>  If the above sounds reasonable, it would entail the following 
>> actions:
>>  - Neutron(/Cyborg?): backportable patch to
>>  s/CONF.host/socket.gethostname()/
>>  - Nova: GET /os-hypervisors*?service_host=X in a new microversion.
>>  - Neutron/Cyborg: master-only patch to do the logic described in 
>> `Upgrade
>>  Concerns`_ (though for now without the `elif` branch).
> 
> Cyborg does use CONF.host today. We can use socket.gethostname() 
> instead.
> 
> On a related node, the patch for ARQ binding [1] uses instance.host, 
> which comes from CONF.host. I'll change it to instance.node, which 
> comes from [2], which in turn comes from get_hostname() [3].

I'm not sure that for binding you need to use the instance.node. See my 
neutron port binding discussion in my response to Eric's original mail.

> 
> [1] 
> https://review.opendev.org/#/c/631244/46/nova/compute/manager.py@2634
> [2] 
> https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L9137
> [3] 
> https://github.com/openstack/nova/blob/1cd5563f2dd2b218db2422397c8aab394d484626/nova/virt/libvirt/driver.py#L9614
> 
>>  Thanks,
>>  efried
> 
> 
> Thanks & Regards,
> Sundar
> 





More information about the openstack-discuss mailing list