problems migrating Ubuntu 9.10 guest from Proxmox 1.5 to 1.7

[zordio]

I'm trying to migrate all of the guests from a Proxmox 1.5 machine so it can be upgraded. Debian guests have had no problems, but there are a couple of Ubuntu guests which have a problem. After migration, nothing is started except for init. Services can manually be started, but still do not work correctly.

A notable symptom is that if I say "ifup eth0", it will hang indefinitely. ctrl-c will return to command line, and networking will appear to work. Most network dependent services will work, except when I got to mysql (which listens only on localhost), it would not start. The guest uses bridged ethernet since its IP is not in the same netblock as the host.

The guest works correctly after migrating it back to the 1.5 server. Migration is being done with vzmigrate, not using live migration.

Since the Ubuntu 9.10 guests do not start correctly on the 1.7 server, I'm hesitant to upgrade the 1.5 server while those guests are still on it. I'm also hesitant to upgrade the guests themselves, for fear of making things worse.

server 1 info: pveversion -v
pve-manager: 1.5-8 (pve-manager/1.5/4674)
running kernel: 2.6.24-10-pve
pve-kernel-2.6.24-10-pve: 2.6.24-21
pve-kernel-2.6.24-8-pve: 2.6.24-16
qemu-server: 1.1-11
pve-firmware: 1.0-3
libpve-storage-perl: 1.0-10
vncterm: 0.9-2
vzctl: 3.0.23-1pve8
vzdump: 1.2-5
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1

server 2 info: pveversion -v
pve-manager: 1.7-10 (pve-manager/1.7/5323)
running kernel: 2.6.24-11-pve
pve-kernel-2.6.24-11-pve: 2.6.24-23
pve-kernel-2.6.24-8-pve: 2.6.24-16
qemu-server: 1.1-25
pve-firmware: 1.0-10
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-10
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1

 

[tom]

do you talk about Ubuntu container or KVM guest?

in any case, you need to follow the update instructions (which you did not).
see http://pve.proxmox.com/wiki/Downloads#Update_a_running_Proxmox_Virtual_Environment_to_1.7

 

[zordio]

This is Ubuntu OpenVZ containers.

I can double check later, but I'm pretty sure the server was set up with a 1.7 install disc, roughly a month ago. Just about everything on there came from the 1.5 machine, so I can move stuff off and upgrade it to the 2.6.32 kernel if you think that will help.

 

[zordio]

1.7 server has been updated to a 2.6.32 kernel:

# pveversion -v
pve-manager: 1.7-10 (pve-manager/1.7/5323)
running kernel: 2.6.32-4-pve
proxmox-ve-2.6.32: 1.7-30
pve-kernel-2.6.32-4-pve: 2.6.32-30
pve-kernel-2.6.24-11-pve: 2.6.24-23
pve-kernel-2.6.24-8-pve: 2.6.24-16
qemu-server: 1.1-28
pve-firmware: 1.0-10
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-10
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.13.0-3
ksm-control-daemon: 1.0-4

I tried migrating the Ubuntu container, and it still fails to fully start. The web interface shows memory usage of 2MB, and VNC won't connect. VNC does work with a Debian container on the same server.

 

[dietmar]

What is the outpout when you start manually:

# vzctl start <VMID>

 

[zordio]

# vzctl start 113
Starting container ...
Container is mounted
Setting CPU units: 1000
Setting CPUs: 1
Set hostname: ****
File resolv.conf was modified
Setting quota ugidlimit: 0
Configure veth devices: veth113.0
Adding interface veth113.0 to bridge vmbr0 on CT0 for CT113
Container start in progress...

 

[dietmar]

Adm what is in the init.log?

 

[zordio]

init.log:
starting init logger
init: hwclock main process (24) terminated with status 77
init: openvz pre-start process (22) terminated with status 32
mountall: root filesystem isn't mounted
init: mountall main process (26) terminated with status 1
General error mounting filesystems.
A maintenance shell will now be started.
CONTROL-D will terminate this shell and re-try.
Give root password for maintenance
(or type Control-D to continue): init: mountall-shell main process (29) killed by TERM signal
init: hwclock-save main process (34) terminated with status 1
* Asking all remaining processes to terminate...

init logger finished


I found reference to udev causing problems, but udev does not appear to be installed in this container. There is no /etc/init.d/udev, and when it is running on the 1.5 machine, "dpkg -l '*udev*'" lists only libudev0 installed. I also found reference to https://bugs.launchpad.net/ubuntu/+source/mountall/+bug/436130 but there was no change using the newer openvz.conf shown near the end.

 

[dietmar]

Obviously a problem with upstart - But I have no idea how to fix that.

 

[zordio]

I found a working openvz.conf on http://blog.bodhizazen.net/linux/ubuntu-10-04-openvz-templates/

This says it works on 10.04, and I am using it successfully on 9.10. I wouldn't assume that nothing else has broken in 10.10 though.