Xen users - Sep 2007 - Server moved - networking stopped

This is all on SuSE enterprise 10

Our server was moved to a new location last night. After powerup there
was no sign of networking so our ISP reconfigured the box. Outbound
connectivity had previously been through eth1 but is now through eth0

Since restarting Xen I now have no network connectivity through the hosts.

I have tried to provide all the relevant details below. If anyone has
a moment to take a look and offer a suggestion as to why my networking
may be down, I''d love to hear it. Thanks.

Our configuration is as follows.

eth0      Link encap:Ethernet  HWaddr 00:1A:A0:05:7E:F9
          inet addr:111.222.82.76  Bcast:111.222.82.255  Mask:255.255.255.0

We have 111.222.93.104/29 routed through that IP

At the command prompt I did
# modprobe dummy
# ifconfig dummy0 111.222.93.105 netmask 255.255.255.248 up

That gives me:

dummy0    Link encap:Ethernet  HWaddr 26:65:C6:05:99:96
          inet addr:111.222.93.105  Bcast:111.222.93.111  Mask:255.255.255.248

Which looks okay.

I restart xen:

# rcxend restart

And my bridge appears:

xenbr0    Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF
          inet addr:111.222.93.105  Bcast:0.0.0.0  Mask:255.255.255.248

starting a machine makes a vif appear too:

vif3.0    Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF
          inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link

The networking part of xend-config.sxp looks like this:

(network-script ''network-virtual bridgeip="111.222.93.105/29"
brnet="111.222.93.104/29" netdev="eth0"'')

(vif-script vif-bridge)

The variables section in network virtual are:

vifnum=${vifnum:-0}
bridgeip=${bridgeip:-111.222.93.104/29}
brnet=${brnet:-111.222.93.105/29}
netmask=${netmask:-255.255.255.248}
bridge=${bridge:-xenbr${vifnum}}

I''m not aware of any changes to the vif-bridge script

I am unable to ping the vm from dom0
>From an external host I can ping the bridge IP
Traceroutes from an external host to 111.222.93.107 do not receive a
reply from the bridge IP. I''m unsure if they should.

Traceroutes to 111.222.93.105 (the bridge ip itself) do receive a reply.

The only vm config for networking is:

vif = [ ''mac=00:16:3e:9a:d4:33'' ]

Where the MAC address is changed in each vm script.

As I say, this config worked until yesterday,  My only change today
was in xend-config.sxp changing eth1 to eth0

Thanks again,

Russell

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

hi,

Russell Horn schrieb:
> This is all on SuSE enterprise 10
> As I say, this config worked until yesterday,  My only change today
> was in xend-config.sxp changing eth1 to eth0
when installing Xen on a SLES Server, yast setup up everything for eth0,
so i assume, that you have to change more then only the xend-config.sxp
file. Use grep or similar to find the other necessary files.

i had similar problems, but i don''t know anymore, which files i have
changed.

cu denny
-- 
Stoppt den Überwachungswahn - Stoppt den Schäuble Katalog:
http://www.nopsis.de



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

On 9/5/07, Denny Schierz <linuxmail@4lin.net> wrote:
> when installing Xen on a SLES Server, yast setup up everything for eth0,
> so i assume, that you have to change more then only the xend-config.sxp
> file. Use grep or similar to find the other necessary files.
I did grep the directory but this was the only instance of eth1

The setup was not done through yast but from the command line. I am,
however, using the SuSE xen 3.0 RPMs

Anyone else see any reason why my network traffic isn''t flowing?

Russell

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