[lxc-users] Question on bridging in nested containers

[Anjali Kulkarni]

I mean “bypass the networking (and bridge) on *base* LXC"

On 3/12/15, 12:35 PM, "Anjali Kulkarni" <anjali at juniper.net> wrote:

>Thanks Fajar and Guido.
>The use case for this is that I don’t want to have double the no of IP and
>MACs for each interface I want on the nested container. I want to be able
>to bypass the networking (and bridge) on LXC so that I can use only one IP
>and MAC for any eth interface (I don’t need connectivity on the base
>container)
>
>Anjali
>
>On 3/11/15, 11:45 PM, "Guido Jäkel" <G.Jaekel at DNB.DE> wrote:
>
>>Dear Fajar,
>>
>>Theoretical, for the root host it should be possible to "pull out" the
>>"fist container level end" of the vet's of for the nested containers to
>>the host by changing it's namespace, isn't it?
>>
>>But because this will need the "cooperation" of the host environment to
>>start a nested container from the first level container, which will
>>undermine the nesting szenario.
>>
>>
>>Dear Anjali,
>>
>>what's your usecase or aim to try such a way?
>>
>>Guido
>>
>>
>>On 11.03.2015 22:21, Anjali Kulkarni wrote:
>>> Sorry, I didn¹t phrase this correctly.
>>> What I meant is I ok to bypass the base container (don¹t need bridging
>>> connectivity to it), but just want to expose the host bridge to the
>>>nested
>>> container.
>>> 
>>> Anjali
>>> 
>>> On 3/11/15, 2:14 PM, "Fajar A. Nugraha" <list at fajar.net> wrote:
>>> 
>>>> On Thu, Mar 12, 2015 at 3:10 AM, Anjali Kulkarni <anjali at juniper.net>
>>>> wrote:
>>>>
>>>>> interested in knowing if one bridge on the host can be used to bridge
>>>>> the
>>>>> base container and the nested container within the base container?
>>>>
>>>> No. A network interface can't be both in the host and in the container
>>>> at the same time.
>>
>>_______________________________________________
>>lxc-users mailing list
>>lxc-users at lists.linuxcontainers.org
>>http://lists.linuxcontainers.org/listinfo/lxc-users
>