Hello,
Maybe you could specify the MAC address – are You sure the autoamtically
assigned mac adress is not used by some other virtual machine? I would prefere
to assign the mac addrss explicitely in the „vif“ cofiguration line
to avoid the possible conflict... especially if Yo have more servers with Dom0
then it was the problem by us...
With best regards, Archie
P.S.
One note to your configuration:
You
have there
vncdisplay=1
vncunused=1
What
is not correct I think – as far as I now, vncdisplay (port) is used only in
the case if vncunused=0, otherwise this option will be ignored.
From: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
[mailto:xen-users-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Marco
Strullato
Sent: Thursday, December 06, 2007 5:31 PM
To: xen-users
Subject: [Xen-users] Re: linux hvm network problem
I've noticed this into log files:
Dec 6 17:23:05 HYP02 kernel: device tap0 entered promiscuous mode
Dec 6 17:23:05 HYP02 kernel: audit(1196958185.809:38): dev=tap0 prom=256
old_prom=0 auid=4294967295
Dec 6 17:23:05 HYP02 kernel: xenbr0: port 3(tap0) entering learning state
Dec 6 17:23:05 HYP02 kernel: xenbr0: topology change detected,
propagating
Dec 6 17:23:05 HYP02 kernel: xenbr0: port 3(tap0) entering forwarding
state
Dec 6 17:23:06 HYP02 kernel: device vif18.0 entered promiscuous mode
Dec 6 17:23:06 HYP02 kernel: audit(1196958186.493:39): dev=vif18.0
prom=256 old_prom=0 auid=4294967295
Dec 6 17:23:06 HYP02 kernel: ADDRCONF(NETDEV_UP): vif18.0: link is not
read
About DomU, network packets does'n pass through vif18.0 but only though tap0
and xenbr0...
into the qemu log, I have
domid: 18
qemu: the number of cpus is 2
shared page at pfn:5ffff, mfn: 24fca
buffered io page at pfn:5fffd, mfn: 24fcc
char device redirected to /dev/pts/1
False I/O request ... in-service already: 0, pvalid: 0, port: 0, data: 0,
count: 0, size: 0
False I/O request ... in-service already: 0, pvalid: 0, port: 0, data: 0,
count: 0, size: 0
Thanks,
Marco
2007/12/6,
Marco Strullato <marco.strullato@xxxxxxxxx>:
Hi all,
I'm having network problems with hvm.
My conf is centos5 with xen 3.0.3. I'm booting a RedHat 9 with this conf file
kernel = "/usr/lib/xen/boot/hvmloader"
builder='hvm'
memory = 1536
shadow_memory = 8
name = "SERVER"
vcpus=2
pae=0
acpi=0
apic=0
disk = [ 'phy:/dev/HYP02VM/LUMINA,ioemu:hda,w', ]
device_model = '/usr/lib/xen/bin/qemu-dm'
sdl=1
vnc=1
vnclisten="0.0.0.0"
vncdisplay=1
vncunused=1
vncconsole=1
vncpasswd='abc123'
nographic=0
stdvga=0
serial='pty'
vif = [ 'type=ioemu, bridge=xenbr0, model=ne2k_pci']
After the 1st boot, the guest recognize the new configuration and reconfigure
the network correctly with kudzu: the guest can reach the dhcp server and get a
valid ip address. After the first connection to the dhcp server the guest
became "isolated". I mean I can not connet to the default gateway. I
tried to assign manually the ip address but nothing changes.
I took a tcpdump on the dom0 listenting to the bridge and to the hvm domU. I
have the same result: only arp requests
10.0.1.29 is the ip address of
the hvm guest... the other is the mine ip.
15:53:30.472523 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:31.492565 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:32.492586 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:33.492619 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:34.492697 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:35.492687 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:36.492717 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:37.492786 arp who-has 10.0.0.34
tell 10.0.1.29
15:53:38.492787 arp who-has 10.0.0.34
tell 10.0.1.29
Do you have some suggestions? Dom0's network is running well, I have
other linux paravirtualized vm running...
Regards,
Marco
