[one-users] unable to start a VM

Sun Feb 1 04:22:42 PST 2009

OK,
	Yes, I agree that there should be an error message regarding the 
misconfiguration with sudo. 

	Regarding the pending VM: your hosts report 128MB of  Free Memory, an the 
TestVM needs 256MB. This is probably because of you are using ballooning with 
Xen. In this case, you can easily solve this, see:

http://trac.opennebula.org/ticket/54

This bug affects ballooning configurations in 1.0. Unfortunately, the patch is 
not in 1.2.

You can get more info about scheduling in the sched.log file (that one is C++ 
;)).

Cheers

Ruben

On Sunday 01 February 2009 07:08:15 you wrote:
> Thanks Ruben,
>
> I eventually debugged that problem, but the VM is still not running
>
> What was the problem with "onehost list"?
> I found out that my sudo on destination machines (obelix41 and obelix43)
> required tty. So I disabled it (visudo and comment out requiretty) and now
> the result is as shown below onehost list
>  HID NAME                      RVM   TCPU   FCPU   ACPU    TMEM    FMEM
> STAT 0 obelix43                    0    800    799    800 4192256  130048  
> on 1 obelix41                    0    800    799    800 4192256  130048  
> on The Surprising thing is that although sudo was not working remotely
> there was no error reported in oned.log. Anyways, that seems to be clean
> now .... but the problem still remains.
>
> My VM is always on the STATUS of pending.
>
>  onevm list
>   ID     NAME STAT CPU     MEM        HOSTNAME        TIME
>    3   TestVM pend   0       0                 00 00:10:13
>
> I do not know what is happening now. None of the logs report any error. Is
> there is quick fix which you know  (the fact is that I do not know ruby).
> do you think that since this is a beta version so I have unearthed some
> untested portion of code and I should switch to the stable version, i.e.
> 1.0?
>
> thanks,
> -upendra
>
> > Date: Sun, 1 Feb 2009 00:43:51 +0100
> > From: rubensm at dacya.ucm.es
> > Subject: Re: [one-users] unable to start a VM
> > To: users at lists.opennebula.org
> > CC: upendras at hotmail.com
> >
> > Hi
> > 	It seems that the VMs are not deployed because the obelix nodes are not
> > reporting their status. From the onehost output, you can see that they do
> > not have any Available CPU (ACPU)
> >
> > 	This is usually a problem with ssh. You should have access two the nodes
> > without being prompt for a password. So at the front-end:
> > $ssh obelix41
> > should give you a shell at obelix41 (without asking nothing).
> >
> > 	You can get more details in oned.log. There should be something related
> > to ERROR Monitoring obelix...
> >
> > Cheers
> >
> > Ruben
> >
> > On Sunday 01 February 2009 00:05:11 Upendra Sharma wrote:
> > > Hi,
> > >
> > > I have downloaded OpenNebul -1.1.85
> > > I have installed it on CentOS 5.2
> > >
> > > I am trying to create a VM (using "onevm create myvm.tmp") but it
> > > returns the result is that the VM creation is always pending
> > >
> > > onevm list:
> > >   ID     NAME STAT CPU     MEM        HOSTNAME        TIME
> > >    2   TestVM pend   0       0                 00 00:00:20
> > >
> > >  onevm show TestVM
> > > VID            : 2
> > > UID            : 0
> > > STATE          : PENDING
> > > LCM STATE      : LCM_INIT
> > > DEPLOY ID      :
> > > MEMORY         : 0
> > > CPU            : 0
> > > LAST POLL      : 0
> > > START TIME     : 01/31 17:59:48
> > > STOP TIME      : 12/31 19:00:00
> > > NET TX         : 0
> > > NET RX         : 0
> > >
> > > ....: Template :....
> > >     CPU             : 1
> > >     DISK            :
> > > READONLY=no,SOURCE=/home/upendra/nebula/images/CentOS5.img,TARGET=sda
> > > MEMORY          : 256
> > >     NAME            : TestVM
> > >     NIC             :
> > > BRIDGE=eth0,IP=192.168.245.80,MAC=00:03:c0:a8:f5:50,NETWORK=Public
> > > VLAN,VNID=0 OS              : BOOTLOADER=/usr/bin/pygrub
> > >
> > >
> > > I do not know what is the problem. I frankly do not know where to start
> > > the debugging from? Any help is appreciated.
> > >
> > > My cluster configuration is as follows: my cluster front-end is again a
> > > CentOS 5.2 machine with xen on it. I have added two more nodes on it
> > > (obelix41, obelix43) onehost list
> > >  HID NAME                      RVM   TCPU   FCPU   ACPU    TMEM    FMEM
> > > STAT 0 obelix41                    0             0    100
> > > on 1 obelix43                    0             0    100
> > > on
> > >
> > > Each of these machines are a quadcore dual-processor machine.
> > >
> > > thanks,
> > > -upendra
> > >
> > > tha
> > >
> > >
> > > _________________________________________________________________
> > > Chose your Life Partner! Join MSN Matrimony FREE
> > > http://www.in.msn.com/matrimony
> >
> > --
> > +---------------------------------------------------------------+
> >  Dr. Ruben Santiago Montero
> >  Associate Professor
> >  Distributed System Architecture Group (http://dsa-research.org)
> >
> >  URL:    http://dsa-research.org/doku.php?id=people:ruben
> >  Weblog: http://blog.dsa-research.org/?author=7
> >
> >  GridWay, http://www.gridway.org
> >  OpenNEbula, http://www.opennebula.org
> > +---------------------------------------------------------------+
>
> _________________________________________________________________
> Plug in to the MSN Tech channel for a full update on the latest gizmos that
> made an impact. http://computing.in.msn.com/

-- 
+---------------------------------------------------------------+
 Dr. Ruben Santiago Montero
 Associate Professor
 Distributed System Architecture Group (http://dsa-research.org)

 URL:    http://dsa-research.org/doku.php?id=people:ruben
 Weblog: http://blog.dsa-research.org/?author=7

 GridWay, http://www.gridway.org
 OpenNEbula, http://www.opennebula.org
+---------------------------------------------------------------+