[one-users] ttylinux fails to start

Fri Apr 15 07:27:21 PDT 2011

Don't do that, the disk.0 that is written in the deployment file is
the actual disk copied by the TM. It seems that you are using tm_nfs
but the shared storage is not mounted in the correct path in the
remote node. Try creating a file in /srv/cloud/one/var and check if
you can see it in the remote host in the same path.

On Fri, Apr 15, 2011 at 4:05 PM, Mike Putter <mp at wirelesswebsolutions.nl> wrote:
> Got some progress here:
>
>> qemu: could not open disk image /srv/cloud/one/var//8/images/disk.0: No
>> such file or directory
>
> Indeed: the XML file is called deployment.0
> Don't know why it created the file as disk.0 ...
>
> Moreover, I could get the image running directly in the node by invoking:
>
> virsh --connect qemu:///system create deployment.0
> Domain one-8 created from deployment.0
>
> (off course after editting the deployment.0 file changing disk.0 to
> deployment.0)
>
> So, it seems it's possible to get it going. Now I have to figure out how to
> teach OpenNebula to create a proper deployment file !
> Where is this file generated ?
>
>
> On 04/15/2011 01:13 PM, Mike Putter wrote:
>>
>> Thanks for your swift reply: the output of /var/log/libvirt/qemu/one-8.log
>> reads:
>>
>> LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin
>> /usr/bin/kvm -S -M pc
>> -0.12 -cpu qemu32 -enable-kvm -m 64 -smp 1 -name one-8 -uuid
>> a0a30e17-c21d-6e28-39bb-aed8df76c2b
>> 4 -nographic -chardev
>> socket,id=monitor,path=/var/lib/libvirt/qemu/one-8.monitor,server,nowait -
>> monitor chardev:monitor -no-acpi -boot c -drive
>> file=/srv/cloud/one/var//8/images/disk.0,if=ide,
>> index=0,boot=on,format=raw -net
>> nic,macaddr=02:00:c0:a8:01:0b,vlan=0,name=nic.0 -net tap,fd=38,v
>> lan=0,name=tap.0 -serial none -parallel none -usb
>> open /dev/kvm: No such file or directory
>> Could not initialize KVM, will disable KVM support
>> qemu: could not open disk image /srv/cloud/one/var//8/images/disk.0: No
>> such file or directory
>>
>>
>> Is it possible to have a fall back to qemu if the VT-x instruction set is
>> absent ?
>> It looks like qemu is trying ...
>>
>>
>> On 04/15/2011 01:01 PM, Javier Fontan wrote:
>>>
>>> Probably the problem is the support for virtualization. To be sure you
>>> can check in the node /var/log/libvirt/qemu/one-<vmid>.log.
>>>
>>> On Tue, Apr 12, 2011 at 3:31 PM, Mike Putter<mp at wirelesswebsolutions.nl>
>>>  wrote:
>>>>
>>>> Hi,
>>>>
>>>> I followed the example of creating a ttylinux VM.
>>>>
>>>> The VM starts, boots and eventually fails.
>>>>
>>>> The error log reads:
>>>>
>>>> Tue Apr 12 15:02:23 2011 [DiM][I]: New VM state is ACTIVE.
>>>> Tue Apr 12 15:02:23 2011 [LCM][I]: New VM state is PROLOG.
>>>> Tue Apr 12 15:02:23 2011 [VM][I]: Virtual Machine has no context
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh:
>>>> voyager:/srv/cloud/one/one-templates/ttylinux.img
>>>> wws-vmhost01:/srv/cloud/one/var//1/ima
>>>> ges/disk.0
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: DST:
>>>> /srv/cloud/one/var//1/images/disk.0
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: Creating directory
>>>> /srv/cloud/one/var//1/images
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: Executed "mkdir -p
>>>> /srv/cloud/one/var//1/images".
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: Executed "chmod a+w
>>>> /srv/cloud/one/var//1/images".
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: Cloning
>>>> /srv/cloud/one/one-templates/ttylinux.img
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: Executed "cp -r
>>>> /srv/cloud/one/one-templates/ttylinux.img /srv/cloud/one/var//1/images/d
>>>> isk.0".
>>>> Tue Apr 12 15:02:24 2011 [TM][I]: tm_clone.sh: Executed "chmod a+rw
>>>> /srv/cloud/one/var//1/images/disk.0".
>>>> Tue Apr 12 15:02:26 2011 [LCM][I]: New VM state is BOOT
>>>> Tue Apr 12 15:02:26 2011 [VMM][I]: Generating deployment file:
>>>> /srv/cloud/one/var/1/deployment.0
>>>> Tue Apr 12 15:02:58 2011 [VMM][I]: Command execution fail: 'if [ -x
>>>> "/var/tmp/one/vmm/kvm/deploy" ]; then /var/tmp/one/vmm/kvm/deploy /
>>>> srv/cloud/one/var//1/images/deployment.0; else
>>>>  exit 42; fi'
>>>> Tue Apr 12 15:02:58 2011 [VMM][I]: STDERR follows.
>>>> Tue Apr 12 15:02:58 2011 [VMM][I]: error: Failed to create domain from
>>>> /srv/cloud/one/var//1/images/deployment.0
>>>> Tue Apr 12 15:02:58 2011 [VMM][I]: error: monitor socket did not show
>>>> up.:
>>>> Connection refused
>>>> Tue Apr 12 15:02:58 2011 [VMM][I]: ExitCode: 255
>>>> Tue Apr 12 15:02:58 2011 [VMM][E]: Error deploying virtual machine:
>>>> error:
>>>> Failed to create domain from /srv/cloud/one/var//1/images/de
>>>> ployment.0
>>>> Tue Apr 12 15:02:58 2011 [DiM][I]: New VM state is FAILED
>>>> Tue Apr 12 15:02:58 2011 [TM][W]: Ignored: LOG - 1 tm_delete.sh:
>>>> Deleting
>>>> /srv/cloud/one/var//1/images
>>>>
>>>> Tue Apr 12 15:02:58 2011 [TM][W]: Ignored: LOG - 1 tm_delete.sh:
>>>> Executed
>>>> "rm -rf /srv/cloud/one/var//1/images".
>>>>
>>>> Tue Apr 12 15:02:58 2011 [TM][W]: Ignored: TRANSFER SUCCESS 1 -
>>>>
>>>> Tue Apr 12 15:16:36 2011 [DiM][I]: New VM state is DONE.
>>>>
>>>>
>>>>
>>>> I restarted the host machine before deployment of the VM, so VM services
>>>> should be initiated.
>>>>
>>>> BTW: I found out that the host CPU has no support for the virtualization
>>>> instruction set required by KVM. Can it be the cause of the above
>>>> failure ?
>>>> Or should KVM revert automatically to another state to continue
>>>> deployment ?
>>>>
>>>> I'm glad to provide more output from logs if necessary.
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at lists.opennebula.org
>>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>>>
>>>
>>>
>

-- 
Javier Fontan, Grid & Virtualization Technology Engineer/Researcher
DSA Research Group: http://dsa-research.org
Globus GridWay Metascheduler: http://www.GridWay.org
OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org