Problem after Xen moved to other Host

I have a problem with xens moved to another host system.

I user the "xen-squeeze-base (Debian Squeeze Xen instance with base OS)" image to create the xen. Now i move this system to another host (live migration). This works fine... But during this move, cloudmin add a kernel and ramdisk parameter to the config file... After rebooting the VM i get the following message:

Starting up webserver01.cloud.xxx.de .. .. failed : Error: (2, 'Invalid kernel', 'elf_init: not an ELF binary\n') Using config file "/xenstore/webserver01.cfg".

I tried to remove the kernel and ramdisk parameter manually... Now the startup works again...

Status: 
Closed (fixed)

Comments

Which parameters did Cloudmin add (and you remove) exactly?

This is the config after i moved the server.

root@node01:/xenstore# cat webserver01.cfg memory = 1024 name = 'webserver01' vif = [ 'ip=80.69.x.x,mac=00:16:3e:05:01:6A' ] address = '80.69.xx.xx' netmask = '255.255.0.0' disk = ['file:/xenstore/webserver01.img,hda,w','file:/xenstore/webserver01.swap,hdb,w'] builder = "hvm" device_model = "/usr/lib64/xen/bin/qemu-dm" acpi = 1 pae = 1 vnc = 1 vnclisten = "0.0.0.0" vncunused = 1 vncpasswd = "netlabadmin" kernel = "/xenstore/vmlinuz-vm2-xenU" ramdisk = "/xenstore/initrd.vm2.xenU.img"

The last two line causes the error.

Ok, I see the cause of this now .. will fix in the next Cloudmin release (version 5.9).

Thx. Do you know a release date of cloudmin 5.9?

It will be a couple of weeks at least..

This is a long time... At the moment the migration feature is not usable in an productive environment. How can we fix it manually?

I can send you a new build with a fix if you like - just let me know which Linux distribution you are running the Cloudmin master on.

This would be nice.

We use Debian Squeeze on our master server.

Ok, I have sent the update via email ..

Automatically closed -- issue fixed for 2 weeks with no activity.