[Openstack] How to commucation vms in multi nodes using quantum?

Robert Kukura rkukura at redhat.com
Wed Oct 24 19:26:32 UTC 2012


On 10/24/2012 12:42 PM, Dan Wendlandt wrote:
> On Wed, Oct 24, 2012 at 3:22 AM, Gary Kotton <gkotton at redhat.com> wrote:
>> Hi,
>> In addition to Dan's comments you can also take a look at the following link
>> http://wiki.openstack.org/ConfigureOpenvswitch.
> 
> Is there any content on that wiki page that is not yet in the quantum
> admin guide: http://docs.openstack.org/trunk/openstack-network/admin/content/?
>    If so, we should file a bug to make sure it ends up in the admin
> guide and that the wiki page is deleted so there is exactly one place
> where we direct people and we avoid stale content.
> 
> Bob is probably best to answer that question.

I've already filed a docs bug to update the admin guide with the current
configuration details for linuxbridge and openvswitch, and its assigned
to me. I hope to get to this in the next few days. I'll remove the wiki
page, which is also out-if-date, when its complete.

-Bob

> 
> Dan
> 
> 
>> Thanks
>> Gary
>>
>>
>> On 10/24/2012 08:21 AM, livemoon wrote:
>>
>> Thanks Dan
>>
>> On Wed, Oct 24, 2012 at 2:15 PM, Dan Wendlandt <dan at nicira.com> wrote:
>>>
>>> On Tue, Oct 23, 2012 at 10:56 PM, livemoon <mwjpiero at gmail.com> wrote:
>>>> Dan:
>>>> Thank you for your help.
>>>> If the server have three nics, which one will be used as port of
>>>> "br-int". I
>>>> must know how "br-int" work between two machines, and then I can make
>>>> the
>>>> physical interface which "br-int" use to one switch
>>>
>>> If you are using tunneling, the traffic will exit out the NIC based on
>>> your physical server's routing table and the destination IP of the
>>> tunnel.  For example, if your physical server is tunneling a packet to
>>> a VM on a physical server with IP W.X.Y.Z, the packet will leave
>>> whatever NIC has the route to reach W.X.Y.Z .
>>>
>>> Dan
>>>
>>>
>>>
>>>
>>>>
>>>> On Wed, Oct 24, 2012 at 11:52 AM, Dan Wendlandt <dan at nicira.com> wrote:
>>>>>
>>>>> all you need to do is create a bridge named "br-int", which is what
>>>>> the linux devices representing the vm nics will be plugged into.
>>>>>
>>>>> since you are using tunneling, there is no need to create a br-ethX
>>>>> and add a physical interface to it.
>>>>>
>>>>> dan
>>>>>
>>>>> p.s. btw, your config looks like its using database polling, which is
>>>>> not preferred.  I'd suggest you use the default config, which uses RPC
>>>>> communication between agents and the main quantum-server process
>>>>>
>>>>>
>>>>> On Tue, Oct 23, 2012 at 8:44 PM, livemoon <mwjpiero at gmail.com> wrote:
>>>>>> I know in one node,vm can work well.
>>>>>> I want to know in multi nodes, do I need to create a br-ethX, and
>>>>>> port
>>>>>> the
>>>>>> physical interface to it? how to do that in configuration?
>>>>>>
>>>>>> On Wed, Oct 24, 2012 at 11:36 AM, 刘家军 <iamljj at gmail.com> wrote:
>>>>>>>
>>>>>>> you just need to create one or more networks and specify which
>>>>>>> network
>>>>>>> to
>>>>>>> use when booting vm.
>>>>>>>
>>>>>>> 2012/10/24 livemoon <mwjpiero at gmail.com>
>>>>>>>>
>>>>>>>> Hi, I use quantum as network. A question is if there are multi
>>>>>>>> nodes,
>>>>>>>> how
>>>>>>>> to config to make vms communicate with each other in the same
>>>>>>>> subnet.
>>>>>>>>
>>>>>>>> I use openvswitch as my plugin. And my setting is blow:
>>>>>>>>
>>>>>>>> [DATABASE]
>>>>>>>> sql_connection = mysql://quantum:openstack@172.16.1.1:3306/quantum
>>>>>>>> reconnect_interval = 2
>>>>>>>>
>>>>>>>> [OVS]
>>>>>>>>
>>>>>>>> tenant_network_type = gre
>>>>>>>> tunnel_id_ranges = 1:1000
>>>>>>>> integration_bridge = br-int
>>>>>>>> tunnel_bridge = br-tun
>>>>>>>> local_ip = 172.16.1.2
>>>>>>>>
>>>>>>>> enable_tunneling = True
>>>>>>>>
>>>>>>>>
>>>>>>>> [AGENT]
>>>>>>>> polling_interval = 2
>>>>>>>> root_helper = sudo /usr/bin/quantum-rootwrap
>>>>>>>> /etc/quantum/rootwrap.conf
>>>>>>>>
>>>>>>>> --
>>>>>>>> 非淡薄无以明志,非宁静无以致远
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Mailing list: https://launchpad.net/~openstack
>>>>>>>> Post to     : openstack at lists.launchpad.net
>>>>>>>> Unsubscribe : https://launchpad.net/~openstack
>>>>>>>> More help   : https://help.launchpad.net/ListHelp
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> 刘家军@ljjjustin
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> 非淡薄无以明志,非宁静无以致远
>>>>>>
>>>>>> _______________________________________________
>>>>>> Mailing list: https://launchpad.net/~openstack
>>>>>> Post to     : openstack at lists.launchpad.net
>>>>>> Unsubscribe : https://launchpad.net/~openstack
>>>>>> More help   : https://help.launchpad.net/ListHelp
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>> Dan Wendlandt
>>>>> Nicira, Inc: www.nicira.com
>>>>> twitter: danwendlandt
>>>>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> 非淡薄无以明志,非宁静无以致远
>>>
>>>
>>>
>>> --
>>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> Dan Wendlandt
>>> Nicira, Inc: www.nicira.com
>>> twitter: danwendlandt
>>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>
>>
>>
>>
>> --
>> Blog Site: livemoon.org
>> Twitter: mwjpiero
>> 非淡薄无以明志,非宁静无以致远
>>
>>
>>
>> _______________________________________________
>> Mailing list: https://launchpad.net/~openstack
>> Post to     : openstack at lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~openstack
>> More help   : https://help.launchpad.net/ListHelp
>>
>>
>>
>> _______________________________________________
>> Mailing list: https://launchpad.net/~openstack
>> Post to     : openstack at lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~openstack
>> More help   : https://help.launchpad.net/ListHelp
>>
> 
> 
> 





More information about the Openstack mailing list