openstack team mailing list archive

[Balamurugan V G <balamuruganvg@xxxxxxxxx>]

"The doc cited explains how this works with Quantum, but not with
Openstack."
-As of Folsom, you have the option of using Quantum for networking or Nova
Network and I believe Nova network will eventually be deprecated.

"VM1 is shown as being accessible via two VNICs, that implies that their is
a Hypervisor that is available by two NICs"
-There can be any number of vNICs while using just one physical NIC. Each
port created in quantum is a vNIC. However the illustration showed 3
physical NICs all of which could have been on a single hypervisor or each
on it sown hypervisor.

Looks like we are deviating from the original question which was posed by
Umar.

Regards,
Balu

On Thu, Jan 31, 2013 at 11:18 PM, Caitlin Bestler <
caitlin.bestler@xxxxxxxxxxx> wrote:

> The doc cited explains how this works with Quantum, but not with Openstack.
>
> VM1 is shown as being accessible via two VNICs, that implies that their is
> a Hypervisor that is available by two NICs. How does Nova know that this is
> in fact one Hypervisor and not two?
>
> Similar question must be addressed for Cinder and Swift. I am most
> familiar with Swift, and it really does not address this situation.
>
>
> Balamurugan V G <balamuruganvg@xxxxxxxxx> wrote:
>
> Perhaps you can use the multiple flat network model shown here:
>
>
> http://docs.openstack.org/trunk/openstack-network/admin/content/use_cases_multi_flat.html
>
> The TenantC VM1 gets two IPs, one from each net. But in this case each net
> is in a different subnet.
>
> May be you can create the nets for the same subet but with diff ranges or
> even just one IP(/32) and disable DHCP.
>
> Basically the idea is to use multiple vNICs.
>
> Just throwing at you, I am not entirely sure if it will work, I havent
> tried it.
>
> Regards,
> Balu
>
> On Thu, Jan 31, 2013 at 1:51 PM, Umar Draz <unix.co@xxxxxxxxx> wrote:
>
>> Hi Vish,
>>
>> Assign direct/static IP address to Instance something like bridge rather
>> than nating is possible? Same like vSphare
>>
>> Best Regards,
>>
>> Umar
>> On Thu, Jan 31, 2013 at 8:37 AM, Vishvananda Ishaya <
>> vishvananda@xxxxxxxxx> wrote:
>>
>>>
>>> On Jan 30, 2013, at 11:35 AM, Umar Draz <unix.co@xxxxxxxxx> wrote:
>>>
>>> > Hi Caitlin,
>>> >
>>> > I need multiple ip address for my Haproxy server.
>>> >
>>> > Here is my senario
>>> >
>>> > I have already running Haproxy Server virtual machine for web load
>>> balancing on vSphare with 45 public ip address. We are running 45 diffrent
>>> websites and we required diffrent ips for each webiste.
>>> >
>>> > We have 6 webservers with only local ip address. These 6 webservers
>>> holding all 45 websites. Haproxy just forward the request to local
>>> webservers accorindg to DNS.
>>>
>>> You can associate multiple floating ips to a single instance. Floating
>>> ips just create natting entries. Keep in mind that traffic that originates
>>> from the server will only come from one of these ips.  Also there is no way
>>> for the server to determine which ip address the traffic came in on so you
>>> will need to do the forwarding at the HTTP layer which has access to the
>>> domain name used for the request.
>>>
>>> Vish
>>>
>>>
>>
>>
>> --
>> Umar Draz
>> Network Architect
>> _______________________________________________
>> Mailing list: https://launchpad.net/~openstack
>> Post to     : openstack@xxxxxxxxxxxxxxxxxxx
>> Unsubscribe : https://launchpad.net/~openstack
>> More help   : https://help.launchpad.net/ListHelp
>>
>>
>