[one-users] Interact with running VM

Fri Jan 14 01:14:11 PST 2011

Dear Tino,

the MAC address is correctly assigned from leases, but probably the 
vmcontext.sh is not executed at boot (maybe rc2.d is not the correct 
runlevel for Ubuntu Server).

With regard to the snapshot problem, I can confirm that when I execute 
the command virsh -c gsx://<host-name>?no_verify=1 snapshot-create 
$ONE_LOCATION/lib/remotes/vmm/vmware/checkpoint, executed also in the 
"save" vmware script, it gives me the error

	"Domain snapshot not found: Could not find snapshot with name 'checkpoint'"

In my opinion this is because it creates the snapshot (in fact if I use 
virsh snapshot-list there is a snapshot) but it gives the snapshot a 
totally different name, in my case "VMware Server Undopoint", and not 
the name that I decide. If I comment out the lines with virsh 
snapshot-create in the "save" script I'm able to stop the machine 
correctly, so the problem is precisely snapshot-create.

Regards,

Filippo Dalla Gassa

Il 13/01/2011 17:23, Tino Vazquez ha scritto:
> Dear Filippo,
>
> comments inline,
>
>
> On Thu, Jan 13, 2011 at 11:35 AM, Filippo Dalla Gassa
> <pippo.dallagassa at gmail.com>  wrote:
>> Dear Tino,
>> thanks for your answer, but problems remain.
>>
>> 1) Network problem: I followed what
>> "http://opennebula.org/documentation:rel2.0:cong" says.
>>
>> copied the vmcontext.sh file in the /etc/init.d directory of the disk used
>> by the VM
>> created the link to vmcontext.sh in /etc/rc2.d
>>
>> However, when I start vm with opennebula, it seems opennebula assigned an IP
>> from leases of the VN, but ping fails and if I enter the virtual machine via
>> VI Client I see that the real assigned IP is totally different.
> Please note that different linux flavors have different startup
> schemas, maybe this is what is happening. Does the MAC address get
> correctly set?
>
>> 2) Snapshot problem: I made some tests and found this: if I try to make a
>> snapshot directly with virsh snapshot-create it gives to me this error
>> "Domain snapshot not found: Could not find snapshot with name '1294913536'"
>> (I let him assign a name). The strange thing is that VI Client gives me no
>> error message, like all goes well.  Moreover if I go into
>> $ONE_LOCATION/var/<VMID>/images/disk.0/ directory I see two files,
>> one-<VMID>-Snapshot1.vmem and one-<VMID>-Snapshot1.vmsn, and if try to
>> revert to snapshot the VM through VI Client it detects correctly the
>> snapshot.
> So the create functionality with virsh  doesn't work (this is what
> opennebula vmware drivers uses in the save state)? Or is it the virsh
> snapshot-revert?
>
> Regards,
>
> -Tino
>
>
>> Regards,
>>
>> Filippo Dalla Gassa
>>
>> Il 11/01/2011 18:58, Tino Vazquez ha scritto:
>>
>> Dear Filippo,
>>
>> Let's see ...
>>
>> 1) The network problem is probably related to a bad contextualization
>> of the VM. have ou set the vmcontext.sh script [see then end of 1] ?
>>
>> 2) VM monitoring in VMware is quite minimal, it just detects if the VM
>> is still running or not
>>
>> 3) This may be due to an undeleted checkpoint file. Could you please
>> use the VI Client to find out the exact error message?
>>
>> Regards,
>>
>> -Tino
>>
>>
>> [1] http://opennebula.org/documentation:rel2.0:vgg
>> --
>> Constantino Vázquez Blanco | dsa-research.org/tinova
>> Virtualization Technology Engineer / Researcher
>> OpenNebula Toolkit | opennebula.org
>>
>>
>>
>> On Mon, Jan 10, 2011 at 3:35 PM, Filippo Dalla Gassa
>> <pippo.dallagassa at gmail.com>  wrote:
>>
>> Hi,
>> I create a vm with this template
>>
>> NAME = "Ubuntu-Server"
>>
>> MEMORY = 512
>>
>> CPU = 1
>>
>> VCPU = 1
>>
>> OS = [
>>    ARCH = i686
>> ]
>>
>> DISK = [
>>    IMAGE = "Ubuntu Server"
>> ]
>>
>> NIC = [NETWORK = "Public"]
>>
>> NIC = [NETWORK = "Other"]
>>
>> After the vm goes running I try to ping the IP address assigned to the vm
>> but the result is "Destination host is unreachable". Also if I use onevm
>> show I see all monitored values (memory, cpu, net_tx, net_rx) are 0
>> (possible there is no activity?) and in log file they result as -1, so for
>> some reason it can't monitor the vm (but it can monitor the host). Finally I
>> can't stop the virtual machine, if I try the state remains "running" and log
>> file says
>>
>> Mon Jan 10 10:34:51 2011 [LCM][I]: New VM state is SAVE_STOP
>> Mon Jan 10 10:35:22 2011 [VMM][E]: Error saving VM state, -
>> Mon Jan 10 10:35:22 2011 [LCM][I]: Fail to save VM state. Assuming that the
>> VM is still RUNNING (will poll VM).
>>
>> I use vmware server as hypervisor, what can I do?
>>
>> Regardly,
>>
>> --
>> Filippo Dalla Gassa
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.opennebula.org
>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>
>>
>>
>
>
> --
> Constantino Vázquez Blanco | dsa-research.org/tinova
> Virtualization Technology Engineer / Researcher
> OpenNebula Toolkit | opennebula.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20110114/8fc5e486/attachment-0003.htm>