On Wed, Feb 18, 2009 at 4:32 PM, Martin Fuxa <fuxa_kos@xxxxxxxxxx> wrote:
> Hi guys,
> there must be something wrong, maybe with xen vif.
> I have two test servers
> 1/ Supermicro 6015C-NT (this story isn't Supermicro problem!)
It probably IS
> 2/ HP ML 110 G5
> System CentOS 5 (x86_64 for dom0, i86 for domU)
what version of xen and kernel-xen (try rpm -q) ?
> I can perform dom0 installation on both servers, without problem.
> But I can't install domU at Supermicro (I can at HP).
> Unable to retrieve http://192.168.6.100/centos/images/stage2.img
> virt-install -d --nographics -p -r 2048 --vcpus 1 -f /dev/lvg1/lv1 -l
> http://192.168.6.100/centos/ -n test
> Ok, I can login to domU at Supermicro from dom0 console. But there is still
> problem with network. I can ping, dig elsewhere, but I can't do wget, elinks,
> run network services ...
Sounds like an old problem
If you're still using an old version (from Centos 5.0, for example),
you should try updating it to the latest available version first (RHEL
5 is currently at xen-3.0.3-80.el5 and kernel-xen-2.6.18-128.1.1.el5).
> I can't see any errors log/message.
> Supermicro have eth0 + eth1 module igb, Intel 82575EB.
> [root@vali2 xen]# brctl show
> bridge name bridge id STP enabled interfaces
> virbr0 8000.000000000000 yes
> xenbr0 8000.feffffffffff no vif1.0
> Maybe, why two vif?
It's normal. vif1.0 is your domU's eth0, vif0.0 is the backend for
your dom0's eth0, peth0 is your real eth0.
Xen-users mailing list