* 802.1Q VLAN support
@ 2007-12-05 13:48 Luis Vinay
2007-12-07 0:10 ` Maximilian Wilhelm
0 siblings, 1 reply; 5+ messages in thread
From: Luis Vinay @ 2007-12-05 13:48 UTC (permalink / raw)
To: xen-devel
I'm working on adding the 802.1Q vlan suport to Xen. Half of the work
is done. The idea is that we can specify something like this on the domU
configuration file:
vlan = [ 'pdev=eth1, vlanid=123, bridge=xenbr1123',
'pdev=eth1, vlanid=100 ' ]
perhaps even specify an IP address (I founded this usefull in some
implementations), and then, as allways:
vif = [ 'bridge=xenbr1123, mac=00:16:3e:00:00:00',
'bridge=xenbr1101, mac=00:16:3e:00:00:01' ]
and perhaps also specify a list in xend-config.spx.
I'll check if the bridge is already created before connecting the VIF,
at domU creation, if not bring it up, and when destroying the domU, also
check if other domU is using the same bridge before shuting it down.
All this raises me some questions:
- Suggestions/advices regarding on where/how put my code?
- Could it bring problems if I add this code in 'xen/xm/create.py'?
gopts.var('vlan', val="iface=IFACE,vlanid=VLANID,vlanbridge=VLANBRIDGE",
fn=append_value, default=[],
[...]
- Other suggestions/advices?
best regards
--
/* ---
Luis Vinay
XTech - Soluciones linux para Empresas
lvinay@xtech.com.ar
--- */
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 802.1Q VLAN support
2007-12-05 13:48 802.1Q VLAN support Luis Vinay
@ 2007-12-07 0:10 ` Maximilian Wilhelm
2007-12-07 0:13 ` Stefan de Konink
2007-12-11 13:28 ` Luis Vinay
0 siblings, 2 replies; 5+ messages in thread
From: Maximilian Wilhelm @ 2007-12-07 0:10 UTC (permalink / raw)
To: xen-devel
Am Wednesday, den 5 December hub Luis Vinay folgendes in die Tasten:
Hi!
> I'm working on adding the 802.1Q vlan suport to Xen. Half of the work
> is done. The idea is that we can specify something like this on the domU
> configuration file:
> vlan = [ 'pdev=eth1, vlanid=123, bridge=xenbr1123',
> 'pdev=eth1, vlanid=100 ' ]
What should that actualy do?
I guess you want to connect vlan 123 from a trunk on eth1 to
xenbr1123. And the next line?
Why not just use the linux way and create vlan interfaces?
e.g.
# vconfig add eth1 123
# vconfig add eth1 100
and then put eth1.123 to xenbr123 and use this?
> perhaps even specify an IP address (I founded this usefull in some
> implementations), and then, as allways:
What's the purpose of the IP here?
[...]
> I'll check if the bridge is already created before connecting the VIF,
> at domU creation, if not bring it up, and when destroying the domU, also
> check if other domU is using the same bridge before shuting it down.
I just create all the vlan interfaces, bondings and bridges via the
networking setup of my OS on bootup so everything is just there.
No need to think about networking stuff after setup once.
> All this raises me some questions:
> - Suggestions/advices regarding on where/how put my code?
> - Could it bring problems if I add this code in 'xen/xm/create.py'?
> gopts.var('vlan', val="iface=IFACE,vlanid=VLANID,vlanbridge=VLANBRIDGE",
> fn=append_value, default=[],
> [...]
I don't speak parseltongue :)
Ciao
Max
--
Follow the white penguin.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 802.1Q VLAN support
2007-12-07 0:10 ` Maximilian Wilhelm
@ 2007-12-07 0:13 ` Stefan de Konink
2007-12-07 0:17 ` Maximilian Wilhelm
2007-12-11 13:28 ` Luis Vinay
1 sibling, 1 reply; 5+ messages in thread
From: Stefan de Konink @ 2007-12-07 0:13 UTC (permalink / raw)
To: xen-devel
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
Maximilian Wilhelm schreef:
> I just create all the vlan interfaces, bondings and bridges via the
> networking setup of my OS on bootup so everything is just there.
> No need to think about networking stuff after setup once.
Because some people actually manage a system that has dynamics, can
migrate between hosts and want to have extra functionality not created
by hand everytime.
Stefan
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFHWJAsYH1+F2Rqwn0RCmRrAKCQVb7nrqrDw0qFsN6i5QvZ0ktoXgCfTa5f
oj6VgnHxsiwijx+J96vmsmQ=
=eHQp
-----END PGP SIGNATURE-----
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 802.1Q VLAN support
2007-12-07 0:13 ` Stefan de Konink
@ 2007-12-07 0:17 ` Maximilian Wilhelm
0 siblings, 0 replies; 5+ messages in thread
From: Maximilian Wilhelm @ 2007-12-07 0:17 UTC (permalink / raw)
To: xen-devel
Am Friday, den 7 December hub Stefan de Konink folgendes in die Tasten:
> Maximilian Wilhelm schreef:
> > I just create all the vlan interfaces, bondings and bridges via the
> > networking setup of my OS on bootup so everything is just there.
> > No need to think about networking stuff after setup once.
> Because some people actually manage a system that has dynamics, can
> migrate between hosts and want to have extra functionality not created
> by hand everytime.
Sorry, I don't take that point.
If I have systems I move host between, I put the same network config
configuration on them by design.
Especially on such machines I do not want any script/tool to fiddle
with my network configuration, but YMMV.
Ciao
Max
--
Follow the white penguin.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 802.1Q VLAN support
2007-12-07 0:10 ` Maximilian Wilhelm
2007-12-07 0:13 ` Stefan de Konink
@ 2007-12-11 13:28 ` Luis Vinay
1 sibling, 0 replies; 5+ messages in thread
From: Luis Vinay @ 2007-12-11 13:28 UTC (permalink / raw)
Cc: xen-devel
In my particulary case I'm developing a solution to an ISP that provides
the VM with some other services like IP PBX, file servers, etc., that
integrates to the client LAN, and instead of using firewalls and routers
to separate clients, they use VLANS, the customers can buy their VMs via
a web page, and all the process must be automatic all that plus some
other are the reasons why I'm doing such effort as add this kind of
behavior to Xen.
On Fri, 2007-12-07 at 01:10 +0100, Maximilian Wilhelm wrote:
> Am Wednesday, den 5 December hub Luis Vinay folgendes in die Tasten:
>
> Hi!
>
> > I'm working on adding the 802.1Q vlan suport to Xen. Half of the work
> > is done. The idea is that we can specify something like this on the domU
> > configuration file:
>
> > vlan = [ 'pdev=eth1, vlanid=123, bridge=xenbr1123',
> > 'pdev=eth1, vlanid=100 ' ]
>
> What should that actualy do?
> I guess you want to connect vlan 123 from a trunk on eth1 to
> xenbr1123. And the next line?
>
> Why not just use the linux way and create vlan interfaces?
> e.g.
> # vconfig add eth1 123
> # vconfig add eth1 100
>
> and then put eth1.123 to xenbr123 and use this?
>
> > perhaps even specify an IP address (I founded this usefull in some
> > implementations), and then, as allways:
>
> What's the purpose of the IP here?
>
> [...]
> > I'll check if the bridge is already created before connecting the VIF,
> > at domU creation, if not bring it up, and when destroying the domU, also
> > check if other domU is using the same bridge before shuting it down.
>
> I just create all the vlan interfaces, bondings and bridges via the
> networking setup of my OS on bootup so everything is just there.
> No need to think about networking stuff after setup once.
>
> > All this raises me some questions:
>
> > - Suggestions/advices regarding on where/how put my code?
> > - Could it bring problems if I add this code in 'xen/xm/create.py'?
>
> > gopts.var('vlan', val="iface=IFACE,vlanid=VLANID,vlanbridge=VLANBRIDGE",
> > fn=append_value, default=[],
> > [...]
>
> I don't speak parseltongue :)
>
> Ciao
> Max
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2007-12-11 13:28 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-12-05 13:48 802.1Q VLAN support Luis Vinay
2007-12-07 0:10 ` Maximilian Wilhelm
2007-12-07 0:13 ` Stefan de Konink
2007-12-07 0:17 ` Maximilian Wilhelm
2007-12-11 13:28 ` Luis Vinay
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.