On Wed, Nov 04, 2009 at 09:46:01PM +0900, Ryota Ozaki wrote:
On Wed, Nov 4, 2009 at 6:02 AM, Daniel P. Berrange
<berrange(a)redhat.com> wrote:
> On Tue, Nov 03, 2009 at 12:23:26AM +0900, Ryota Ozaki wrote:
>> Note that this patch is not mature yet and still proof-
>> of-concept prototype, although it actually works.
>>
>> Current implementation of lxc driver creates vethN named
>> interface(s) in the host and passes as it is to a container.
>> The reason why it doesn't use ethN is due to the limitation
>> that one namespace cannot have multiple iterfaces that have
>> an identical name so that we give up creating ethN named
>> interface in the host for the container.
>>
>> However, we should be able to allow the container to have
>> ethN by changing the name after clone(CLONE_NEWNET) in
>> controller.
>>
>> To this end, this patch extends --veth argument of
>> controller to be able to have two names; one is the name
>> of a created veth, and the other is a new name. The format
>> is:
>>
>> --veth veth1,eth0 --veth veth3,eth1 ...
>
> Do we need to pass the names around on command line args ? Could
> we simply make libvirt_lxc auto-assign the names starting from
> eth0 instead of passing the names from libvirtd
I'm worrying two things; Is it ok that libvirt_lxc has such a decision? and
it seems a bit inflexible, is it acceptable?
I think it is ok, because the administrator inside the container can easily
rename interfaces to whatever they wish later. So a simple fixed naming
policy in libvirt_lxc is sufficient
Daniel
--
|: Red Hat, Engineering, London -o-
http://people.redhat.com/berrange/ :|
|:
http://libvirt.org -o-
http://virt-manager.org -o-
http://ovirt.org :|
|:
http://autobuild.org -o-
http://search.cpan.org/~danberr/ :|
|: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|