[Xen-users] Xen, multiple NICs, VLANs and a headache!

["matt wanless" <matt.wanless@xxxxxxxxxxxxxxxx>]

Hi All.


I'm currently building a set of Xen servers to host our dev and test
environments. I've got them all up and running fine, but I've hit a snag
with assigning them VLANS. OS is FC6, Xen is 2.6.19-1.2911.fc6xen. I've
IPTables and SELinux turned off to make life simple.

The config I'm aiming for is to have one physical NIC on each physical
server for management (eth1), and the other to be a VLAN trunk with all the
environment VLANs on it (each env lives in it's own VLAN). I've got the
VLANs working fine on Domain0 (config below...)

[root@andromeda network-scripts]# cat ifcfg-eth0
# Intel Corporation 82541GI Gigabit Ethernet Controller
DEVICE=eth0
ONBOOT=yes
TYPE=Ethernet
#VLAN=yes
#IPADDR=172.20.10.10
#NETMASK=255.255.255.240
#GATEWAY=172.20.10.14
[root@andromeda network-scripts]#

[root@andromeda network-scripts]# cat ifcfg-eth0.100
# Intel Corporation 82541GI Gigabit Ethernet Controller
DEVICE=eth0.100
ONBOOT=yes
TYPE=Ethernet
VLAN=yes
NETWORK=172.20.10.0
IPADDR=172.20.10.5
NETMASK=255.255.255.240
#GATEWAY=172.20.10.14 - disabled as the default GW is on Eth1
[root@andromeda network-scripts]#

All that works fine. I can ping the gw, access the network etc.

On any DomainU machine, I've no connectivity at all, even to Domain0. 

Config of Domain1 below.

[root@cms1 network-scripts]# cat ifcfg-eth0
# Xen Virtual Ethernet
DEVICE=eth0
ONBOOT=yes
TYPE=Ethernet
#IPADDR=172.20.10.12
#NETMASK=255.255.255.240
#GATEWAY=172.20.10.14
#VLAN=yes
[root@cms1 network-scripts]#


[root@cms1 network-scripts]# cat ifcfg-eth0.100
# Xen Virtual Ethernet
DEVICE=eth0.100
ONBOOT=yes
VLAN=yes
TYPE=Ethernet
IPADDR=172.20.10.12
NETMASK=255.255.255.240
GATEWAY=172.20.10.14
[root@cms1 network-scripts]#


I can ping it's own interface, but not the ip on Domain0 or the gw, and the
arp cache is empty.

Looking at http://wiki.xensource.com/xenwiki/XenNetworking#line-53 it seems
the issue is either VIF1.0 if the bridge itself.

The instructions in the above link refer to a network script that doesn't
exist, so I had a poke around the ones there.

I've tried restarting the bridge with...

/etc/xen/scripts/network-bridge stop
/etc/xen/scripts/network-bridge start

...before the DomainU's start and that loads a xenbr100. I still don't get
any connectivity.

xenbr100  Link encap:Ethernet  HWaddr 00:00:00:00:00:00
          inet addr:172.20.10.0  Bcast:172.20.10.15  Mask:255.255.255.240
          inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:210 errors:0 dropped:0 overruns:0 frame:0
          TX packets:58 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:22216 (21.6 KiB)  TX bytes:9960 (9.7 KiB)


But from Domain1, pinging the gateway fails

[root@cms1 ~]# ping 172.20.10.14
PING 172.20.10.14 (172.20.10.14) 56(84) bytes of data.
>From 172.20.10.12 icmp_seq=1 Destination Host Unreachable


And the arp cache is empty.

[root@cms1 ~]# arp -a
? (172.20.10.14) at <incomplete> on eth0.100
[root@cms1 ~]#



Anyone got any ideas? 


 
Stumped of London. (aka Matt)


 






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