Kernel OOPS reconfiguring physical network device [message #34209] |
Sun, 14 December 2008 20:31 |
laotse
Messages: 35 Registered: December 2008
|
Member |
|
|
I mapped eth3 to a container. I see it and apparently everything is well. However, if I change the configuration of eth3 inside the container and then restart the container, I get kernel crashes:
container:/# nano /etc/network/interfaces.template
container:/# exit
logout
exited from VE 1007
node:~# vzctl stop 1007
Stopping VE ...
VE was stopped
VE is unmounted
node:~# vzctl start 1007
Starting VE ...
VE is mounted
Adding IP address(es): 172.16.6.7
Message from syslogd@node at Sun Dec 14 19:48:19 2008 ...
node kernel: [ 4541.666647] Oops: 0000 [1] SMP
Message from syslogd@node at Sun Dec 14 19:48:19 2008 ...
node kernel: [ 4541.666647] Code: c3 41 54 49 83 c8 ff 31 d2 fc 55 48 89 fd 88 d0 4c 89 c1 49 89 f4 53 31 db 48 83 ec 10 48 8b 3e f2 ae 48 89 ef 48 89 ce 4c 89 c1 <f2> ae 48 f7 d6 f7 d9 8d 4c 31 ff be d0 00 00 00 48 63 f9 e8 97
Message from syslogd@node at Sun Dec 14 19:48:19 2008 ...
node kernel: [ 4541.666647] CR2: 0000000500000001
Getötet
The node is running current Debian/Lenny on an Intel 9500 Quad-Core, i.e:
Linux node 2.6.26-1-openvz-amd64 #1 SMP Wed Nov 26 19:23:13 UTC 2008 x86_64 GNU/Linux
vzctl version 3.0.22
Following this note the node becomes unstable. When trying to shutdown the node, closing down the containers leads to infinite retries. Issuing sync before halt usually works and minimizes the effects from pressing the reset button.
I guess that no misuse - if it was - should crash the kernel. However, if someone has an idea for a work around until this is fixed, I'd gladly appreciate it.
Regards,
- lars.
|
|
|