[one-users] VM sticks in 'pend'

[Tino Vazquez]

Hi Dan,

Another thing I just noticed in your email is that the free memory is
quite scarce. If the machine has more RAM than the one reported
(around 400 Mb), it is quite possible that the Dom0 is missconfigured.
This can also cause the Vms to be stuck in 'pend' state, since the
scheduler won't find any gap to place the VM (this depends also on how
many memory the VM is asking for, if you send us the VM template we
will look at its requirements and see if it fits in your
infrastructure). Take a look at [1] for hints on how to correctly set
up the Dom0.

Best regards,

-Tino

[1] http://lists.opennebula.org/htdig.cgi/users-opennebula.org/2009-May/000436.html
--
Constantino Vázquez, Grid & Virtualization Technology
Engineer/Researcher: http://www.dsa-research.org/tinova
DSA Research Group: http://dsa-research.org
Globus GridWay Metascheduler: http://www.GridWay.org
OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org



On Thu, Jun 17, 2010 at 4:53 PM, Tino Vazquez <tinova at fdi.ucm.es> wrote:
> Hi Dan,
>
> From the steps you sent, it seems like the scheduler is not starting.
> This is due to the way in which OpenNebula is started (oned -f), only
> starts the core, but the scheduler (mm_sched, another process) is not
> started.
>
> You can check if the scheduler is started by issuing an "pgrep mm_sched".
>
> To correctly start opennebula, do a:
>
> $ one start
>
> and to stop it (both processes, oned and mm_sched), do a:
>
> $ one stop
>
> Please, do not hesitate to ask further doubts.
>
> Best regards,
>
> -Tino
>
> --
> Constantino Vázquez, Grid & Virtualization Technology
> Engineer/Researcher: http://www.dsa-research.org/tinova
> DSA Research Group: http://dsa-research.org
> Globus GridWay Metascheduler: http://www.GridWay.org
> OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org
>
>
>
> On Wed, Jun 16, 2010 at 11:17 PM, Dan Yocum <yocum at fnal.gov> wrote:
>> Hi,
>>
>> I'm trying to get a VM to run, but it sticks in 'pend' state and never
>> actually starts.  Here's what I've done so far:
>>
>> 2 systems, both installed with Scientific Linux v5.4, running xen kernel
>> installed opennebula express on one system
>> generated the correct node-install.sh and run it on the worker node
>> editted oned.conf and uncommented all xen related sections, and
>>  commented out all kvm related sections
>> 'oned -f &' as user oneadmin
>> onehost create worker_node im_xen vmm_xen tm_ssh
>> acquired the small network and ttylinux tarballs and edited the ttylinux.one
>> and small_network.net
>> onenet create small_network.net
>> onevm create ttylinux.one
>>
>> Here's what I see:
>>
>> $ onehost list
>>  ID NAME                      RVM   TCPU   FCPU   ACPU    TMEM    FMEM STAT
>>   8 fgitb318                    0    800    799    799 1677619  392192   on
>>
>> $ onevm list
>>  ID     USER     NAME STAT CPU     MEM        HOSTNAME        TIME
>>   8 oneadmin fgitb318 pend   0       0                 00 00:22:42
>>
>> $ tail oned.log:
>> Wed Jun 16 15:52:57 2010 [ReM][D]: VirtualMachineAllocate invoked
>> Wed Jun 16 15:52:57 2010 [DiM][D]: Allocating a new VM
>> Wed Jun 16 15:53:01 2010 [ReM][D]: VirtualMachinePoolInfo method invoked
>> ...
>>
>> And that's it.
>>
>> Any suggestions?
>>
>> Thanks,
>> Dan
>>
>>
>>
>> --
>> Dan Yocum
>> Fermilab  630.840.6509
>> yocum at fnal.gov, http://fermigrid.fnal.gov
>> "I fly because it releases my mind from the tyranny of petty things."
>> _______________________________________________
>> Users mailing list
>> Users at lists.opennebula.org
>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>
>