[one-users] "Inappropriate ioctl for device"

[Claude Noshpitz]

Hi Javier,

I appreciate your following up on this.  The problem went away after a
reboot and my current theory is that the host actually ran out of fds,
leading to "catastrophic" failure (i.e. no proper error message was
generated).  

It's just a theory but it is making me feel better for the moment :)



Thanks!

--Claude


On 3/24/10 4:17 AM, "Javier Fontan" <jfontan at gmail.com> wrote:

> Hello,
> 
> I've been searching but I cannot find any useful information on that
> problem. We've tested libvirt/kvm with more than then 8 VMs per host
> and everything worked fine.
> 
> Bye
> 
> On Tue, Mar 16, 2010 at 12:00 AM, Claude Noshpitz
> <cnoshpitz at attinteractive.com> wrote:
>> Hello,
>> 
>> I've started to see launch failures recently, as my VM hosts are getting more
>> loaded, that generate oned.log messages like this:
>> 
>> Mon Mar 15 15:34:13 2010 [VMM][D]: Message received: LOG - 249 error:
>> internal error Failed to add tap interface 'vnet%d' to bridge 'br0' :
>> Inappropriate ioctl for device
>> 
>> Likely I am missing some documented limitation on the KVM/libvirt or
>> networking side, but it doesn't seem like a hard limit
>> on VMs per host (at least, the failures seem to happen with e.g. 5, 6, 7, or
>> 8 VMs on a host).
>> 
>> Ideas?  RTFMs?
>> 
>> Thanks!
>> 
>> --Claude
>> _______________________________________________
>> 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