(kvm) no network after upgrade

N

natxo

Member
May 17, 2008
30
0
6
hi,

this morning I upgraded the virutal environmente following the instructions on http://proxmox.com/forum/showthread.php?t=533. It all went fine and most vm's came up online all right. Except some (not all) the kvm vm's. This is beta software, so this is expected ;-)

this is the conf file of one of those vm's:

ostype: l26
vlan0: rtl8139=6A:0F:03:A5:E5:70
memory: 256
onboot: 1
ide2: none,media=cdrom
name: dhcp
ide0: vm-105-default.qcow2
bootdisk: ide0

this is a kvm with debian etch (stable). What is funny is that I have another kvm with debian testing (lenny) and it works fine.

Where can I look for the problem?

tia.
 
natxo said:
hi,

this morning I upgraded the virutal environmente following the instructions on http://proxmox.com/forum/showthread.php?t=533. It all went fine and most vm's came up online all right. Except some (not all) the kvm vm's. This is beta software, so this is expected ;-)

this is the conf file of one of those vm's:

ostype: l26
vlan0: rtl8139=6A:0F:03:A5:E5:70
memory: 256
onboot: 1
ide2: none,media=cdrom
name: dhcp
ide0: vm-105-default.qcow2
bootdisk: ide0

this is a kvm with debian etch (stable). What is funny is that I have another kvm with debian testing (lenny) and it works fine.

Where can I look for the problem?

tia.
Click to expand...

hi,

what is not working? only the network? so check if you go unique MAC addresses - if not correct it manually in the config file.

also see the log files: /var/log/syslog (this is also on the web interface).
 
mac addresses are unique:

hpxen:/etc/qemu-server# grep vlan *.conf
103.conf:vlan0: rtl8139=EA:FE:09:DA:AA:97
104.conf:vlan0: rtl8139=62:C8:82:F2:40:EE
105.conf:vlan0: rtl8139=6A:0F:03:A5:E5:70
106.conf:vlan0: rtl8139=66:11:90:7A:38:25
107.conf:vlan0: rtl8139=DE:61:EF:21:4E:CC
109.conf:vlan0: virtio=82:89:74:C6:6D:6C

the only 2 vm's that can network right now are 103 and 109 (103 is debian lenny and 109 is winxp)


and in syslog of the server all I can see is:
Jul 29 17:24:27 hpxen pvedaemon[7321]: VM 105 stopped
Jul 29 17:24:28 hpxen qemuserver[3100]: VM 105 stopped
Jul 29 17:24:39 hpxen pvedaemon[7332]: starting VM 105 on node 0 (localhost)
Jul 29 17:24:39 hpxen qemuserver[3100]: Starting VM 105
Jul 29 17:24:39 hpxen kernel: device tap1 entered promiscuous mode
Jul 29 17:24:39 hpxen kernel: audit(1217345079.897:19): dev=tap1 prom=256 old_prom=
0 auid=4294967295
Jul 29 17:24:39 hpxen kernel: vmbr0: port 3(tap1) entering learning state
Jul 29 17:24:40 hpxen pvedaemon[7332]: VM 105 started
Jul 29 17:24:50 hpxen kernel: tap1: no IPv6 routers present
Jul 29 17:24:54 hpxen kernel: vmbr0: topology change detected, propagating
Jul 29 17:24:54 hpxen kernel: vmbr0: port 3(tap1) entering forwarding state

my /etc/network/interfaces for this kvm vm:

auto eth0
iface eth0 inet static
address 192.168.0.120
netmask 255.255.255.0
gateway 192.168.0.2

# /etc/init.d/networking restart
Reconfiguring network interfaces...SIOCSIFADDR: no such device
eth0: ERROR while getting interface flags: No such device
SIOCSIFNETMASK: no such device
eth0: ERROR while getting interface flags: No such device
Failed to bring up eth0.
done.
 
natxo said:
mac addresses are unique:

hpxen:/etc/qemu-server# grep vlan *.conf
103.conf:vlan0: rtl8139=EA:FE:09:DA:AA:97
104.conf:vlan0: rtl8139=62:C8:82:F2:40:EE
105.conf:vlan0: rtl8139=6A:0F:03:A5:E5:70
106.conf:vlan0: rtl8139=66:11:90:7A:38:25
107.conf:vlan0: rtl8139=DE:61:EF:21:4E:CC
109.conf:vlan0: virtio=82:89:74:C6:6D:6C

the only 2 vm's that can network right now are 103 and 109 (103 is debian lenny and 109 is winxp)


and in syslog of the server all I can see is:
Jul 29 17:24:27 hpxen pvedaemon[7321]: VM 105 stopped
Jul 29 17:24:28 hpxen qemuserver[3100]: VM 105 stopped
Jul 29 17:24:39 hpxen pvedaemon[7332]: starting VM 105 on node 0 (localhost)
Jul 29 17:24:39 hpxen qemuserver[3100]: Starting VM 105
Jul 29 17:24:39 hpxen kernel: device tap1 entered promiscuous mode
Jul 29 17:24:39 hpxen kernel: audit(1217345079.897:19): dev=tap1 prom=256 old_prom=
0 auid=4294967295
Jul 29 17:24:39 hpxen kernel: vmbr0: port 3(tap1) entering learning state
Jul 29 17:24:40 hpxen pvedaemon[7332]: VM 105 started
Jul 29 17:24:50 hpxen kernel: tap1: no IPv6 routers present
Jul 29 17:24:54 hpxen kernel: vmbr0: topology change detected, propagating
Jul 29 17:24:54 hpxen kernel: vmbr0: port 3(tap1) entering forwarding state

my /etc/network/interfaces for this kvm vm:

auto eth0
iface eth0 inet static
address 192.168.0.120
netmask 255.255.255.0
gateway 192.168.0.2

# /etc/init.d/networking restart
Reconfiguring network interfaces...SIOCSIFADDR: no such device
eth0: ERROR while getting interface flags: No such device
SIOCSIFNETMASK: no such device
eth0: ERROR while getting interface flags: No such device
Failed to bring up eth0.
done.
Click to expand...

so I assume the MAC´s changed during update, therefore some guest has problems - eth0 is configured for the old MAC. check the guests OS.
 
natxo said:
you are right, and the solution for future reference is here: http://vmfaq.com/entry/24/

Just delete or rename the file /etc/udev/rules.d/rules.d/z25_persistent-net.rules en reboot, all wil work again.

tricky.
Click to expand...

perfect, thanks for this link!

but we hope there is not need to change MAC´s again during Proxmox VE updates but for all migrations this is important.
 
You must log in or register to reply here.
Top Bottom