All of lore.kernel.org
 help / color / mirror / Atom feed
* vif0.0/veth0 or eth0
@ 2006-02-20  2:21 Nicholas Lee
  2006-02-20  9:51 ` Robbie Dinn
  0 siblings, 1 reply; 4+ messages in thread
From: Nicholas Lee @ 2006-02-20  2:21 UTC (permalink / raw)
  To: Xen development list

When does vif0.0 get created?

Is there some shell magic?

I'm trying to figure out how to do a similar peth1/veth1 split for eth1.

--
Nicholas Lee
http://stateless.geek.nz
gpg 8072 4F86 EDCD 4FC1 18EF  5BDD 07B0 9597 6D58 D70C

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: vif0.0/veth0 or eth0
  2006-02-20  2:21 vif0.0/veth0 or eth0 Nicholas Lee
@ 2006-02-20  9:51 ` Robbie Dinn
  2006-02-20 15:19   ` Ewan Mellor
  0 siblings, 1 reply; 4+ messages in thread
From: Robbie Dinn @ 2006-02-20  9:51 UTC (permalink / raw)
  To: Nicholas Lee; +Cc: Xen development list

Nicholas Lee wrote:
> When does vif0.0 get created?
don't know
> 
> Is there some shell magic?
Are you using bridging?
If so, I think I know this!
look for shell function 'op_start' in file /etc/xen/scripts/network-bridge

> 
> I'm trying to figure out how to do a similar peth1/veth1 split for eth1.

Maybe it is as simple as replacing /etc/xen/scripts/network-bridge
with a wrapper that invokes the original /etc/xen/scripts/network-bridge
twice, each time with different values for environment variables
vifnum, bridge, and netdev. Those environment variables are documented
at the top of /etc/xen/scripts/network-bridge.

quoting from /etc/xen/xend-config.sxp:
> # It is possible to use the network-bridge script in more complicated
> # scenarios, such as having two outgoing interfaces, with two bridges, and
> # two fake interfaces per guest domain.  To do things like this, write
> # yourself a wrapper script, and call network-bridge from it, as appropriate.
> #

HTH

Robbie Dinn

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: vif0.0/veth0 or eth0
  2006-02-20  9:51 ` Robbie Dinn
@ 2006-02-20 15:19   ` Ewan Mellor
  2006-02-20 20:13     ` Nicholas Lee
  0 siblings, 1 reply; 4+ messages in thread
From: Ewan Mellor @ 2006-02-20 15:19 UTC (permalink / raw)
  To: Robbie Dinn; +Cc: Xen development list, Nicholas Lee

On Mon, Feb 20, 2006 at 09:51:41AM +0000, Robbie Dinn wrote:

> Nicholas Lee wrote:
> >When does vif0.0 get created?
> don't know
> >
> >Is there some shell magic?
> Are you using bridging?
> If so, I think I know this!
> look for shell function 'op_start' in file /etc/xen/scripts/network-bridge
> 
> >
> >I'm trying to figure out how to do a similar peth1/veth1 split for eth1.
> 
> Maybe it is as simple as replacing /etc/xen/scripts/network-bridge
> with a wrapper that invokes the original /etc/xen/scripts/network-bridge
> twice, each time with different values for environment variables
> vifnum, bridge, and netdev. Those environment variables are documented
> at the top of /etc/xen/scripts/network-bridge.

Indeed.  See also the final comments on bug #332.

Ewan.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: vif0.0/veth0 or eth0
  2006-02-20 15:19   ` Ewan Mellor
@ 2006-02-20 20:13     ` Nicholas Lee
  0 siblings, 0 replies; 4+ messages in thread
From: Nicholas Lee @ 2006-02-20 20:13 UTC (permalink / raw)
  To: Ewan Mellor; +Cc: Xen development list

On 21/02/06, Ewan Mellor <ewan@xensource.com> wrote:
> Indeed.  See also the final comments on bug #332.

Brillant!!! That solved by bridging problem completely.

Seems that xen 3.0 doesn't like networking set up at all like it was in 2.0.

ie. You can't just create your own bridge and add eth1 in host0 to
that bridge, then all the 2nd vifX.1 interfaces to it.

So the method that works it to configure eth0 and eth1 via the
standard methods. Then run the script as per
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=332.

This is something that should be written about on XenNetworking and
probably included in the manual.

Only issue seems to be an interface alias eth1:0 didn't get
transfered. I'll explore that later, but this saves me trying
unstable. :D

Furthermore, I understand now that veth0 .. veth7 along with vif0.0 to
vif0.7 get created automatically by xen based on nloopbacks=8, which
is the default.

So its easy enough to follow the same process for xenbr2 to xenbr7, if
you so wish.

--
Nicholas Lee
http://stateless.geek.nz
gpg 8072 4F86 EDCD 4FC1 18EF  5BDD 07B0 9597 6D58 D70C

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2006-02-20 20:13 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-02-20  2:21 vif0.0/veth0 or eth0 Nicholas Lee
2006-02-20  9:51 ` Robbie Dinn
2006-02-20 15:19   ` Ewan Mellor
2006-02-20 20:13     ` Nicholas Lee

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.