Re: [Xen-users] How to get eth0 & eth1 fox XEN domU's?

[Thomas Halinka <lists@xxxxxxxxx>]

Am Freitag, den 28.11.2008, 00:53 +0200 schrieb Rudi Ahlers:
> On Thu, Nov 27, 2008 at 5:48 PM, Thomas Halinka <lists@xxxxxxxxx> wrote:
> > Hi Rudi,
> >
> > Am Donnerstag, den 27.11.2008, 14:24 +0200 schrieb Rudi Ahlers:
> >> Hi all,
> >>
> >> How is it possible to get a eth0 & eth1 for a XEN domU?
> >
> > jap,
> >
> >>  The reason I
> >> ask, is that we need to monitor the bandwidth every XEN VPS uses, but
> >> I don't want to count local bandwidth between VPS's, and between the
> >> VPS & backup server.
> >
> > ok
> >
> >>
> >> So, if all internet traffic gets routed on eth0, and IP rather
> >> 196.34.x.x & internal traffic on eth1 & 192.168.10.x - how would I do
> >> that?
> >
> > fire up 2 bridges:
> > - xen-internal   (192.168.10.0)
> > - xen-external   (public IPs)
> >
> 
> Can you please explain this to me in lay-man's terms? How do I do this
> on the dom0?

Supposing you have 2 Nics: eth0 and eth1

eth0 = internet/router/outer-world
eth1 = lan /private-net

in dom0:
cat /etc/network/interfaces 
# /etc/network/interfaces - network interfaces configuration

# loopback interface

auto lo
iface lo inet loopback

# ethernet interface

auto eth0
auto eth1

auto external
iface external inet static
        address 192.34.x.x
        netmask 255.255.255.0
        bridge_ports eth0
        bridge_fd 0
        bridge_hello 2
        bridge_maxage 12
        bridge_stp off

auto internal
iface internal inet static
        address 192.168.10.x
        netmask 255.255.255.0
        bridge_ports eth1
        bridge_fd 0
        bridge_hello 2
        bridge_maxage 12
        bridge_stp off


domu.cfg:
vif         =
[ 
'ip=196.34.x.x,mac=00:16:3E:1B:D8:8D,bridge=external','ip=192.168.10.3,mac=00:16:3E:1B:D8:84,bridge=internal'
 ]


eth1 is not really needed, you also can use dummy-devices too.

> 
> > With this you can do separated traffic-accounting for internal and
> > external traffic.
> >
> >>
> >> Currently when I look @ a domU, I have eth0, eth0:1 & eth0:2
> >> (192.168.10.63) setup, but they all pass through the same interface
> >> vifwise0 on the dom0 side,
> >>
> >>
> >> vifwise0  Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF
> >>           inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
> >>           UP BROADCAST RUNNING NOARP  MTU:1500  Metric:1
> >>           RX packets:1961043 errors:0 dropped:0 overruns:0 frame:0
> >>           TX packets:2521537 errors:0 dropped:4218 overruns:0 carrier:0
> >>           collisions:0 txqueuelen:32
> >>           RX bytes:4195421597 (3.9 GiB)  TX bytes:1075982597 (1.0 GiB)
> >>
> >>
> >> So, from the 3.9BG traffice transmitted, we don't know what is local &
> >> what is internet traffic. I'd like to completely segment the traffic,
> >
> > because theres no separation and every traffic is transmitted over this
> > bridge.
> >
> >> how can I do that?
> >
> > Im using traffic-accounting with iptables....
> 
> How exactly do you do this?

counting the packets which are transferred over the external-bridge ;)

since my external bridge is the gateway for the hosts this was not a
huge problem....

> >
> >
> > hth,
> >
> > Thomas


Thomas


_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users