Open Stack

You want:
resume_guests_state_on_host_boot=true
start_guests_on_host_boot=false

The start_ option will essentially force a reboot every time nova is started, but the resume_ option will only attempt to reboot them if they are supposed to be running and the driver says they are not.

Vish

On Aug 30, 2012, at 10:38 AM, Alejandro Comisario <alejandro.comisario at mercadolibre.com> wrote:

> but what if still ... i want that if the compute node to autostart de vms if for any reason, it gets rebooted ? i dont want to start them by hand ( or issue an API reboot )
> Thats why i think it was great that this feature is managed ( at least in KVM ) from the libvirt side.
> 
> Alejandro
> 
> 
> On Thu, Aug 30, 2012 at 1:59 PM, Alessandro Tagliapietra <tagliapietra.alessandro at gmail.com> wrote:
> Try to disable
> 
>         • # start_guests_on_host_boot=false
> 
> in nova.conf, if nova-compute dies it doesn't means that the hypervisor is death.
> 
> Il giorno 28/ago/2012, alle ore 04:42, Sam Su <susltd.su at gmail.com> ha scritto:
> 
> > Thanks for your guys help!
> > I guess this problem may be caused by nova packages auto upgrade.
> >
> > I just found these two lines in the file /var/log/kern.log:
> > Aug 23 06:34:33 cnode-01 kernel: [4955691.256036] init: nova-network main process (9191) terminated with status 143
> > Aug 23 06:34:35 cnode-01 kernel: [4955693.402082] init: nova-compute main process (9275) terminated with status 143
> >
> > Here is the link for detail of /var/log/kern.log
> > http://pastebin.com/GqH1ju1R
> >
> > Also found these info in /var/log/dpkg.log:
> > 2012-08-23 06:34:33 upgrade nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> > 2012-08-23 06:34:33 status half-configured nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:33 status unpacked nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:33 status half-installed nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:34 status triggers-pending ureadahead 0.100.0-12
> > 2012-08-23 06:34:34 status half-installed nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:34 status triggers-pending ureadahead 0.100.0-12
> > 2012-08-23 06:34:34 status triggers-pending man-db 2.6.1-2
> > 2012-08-23 06:34:34 status half-installed nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:34 status half-installed nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:34 status unpacked nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> > 2012-08-23 06:34:34 status unpacked nova-network 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> > 2012-08-23 06:34:34 upgrade nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.2 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> > 2012-08-23 06:34:34 status half-configured nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:35 status unpacked nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:35 status half-installed nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:35 status half-installed nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.2
> > 2012-08-23 06:34:35 status unpacked nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> > 2012-08-23 06:34:35 status unpacked nova-compute-kvm 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> > 2012-08-23 06:34:35 upgrade nova-compute 2012.1+stable~20120612-3ee026e-0ubuntu1.2 2012.1+stable~20120612-3ee026e-0ubuntu1.3
> >
> > Here is detail:
> > http://pastebin.com/juiSxCue
> >
> > But I am not 100% sure of this. Does anyone knows what it means that a process was terminated with status 143 in Ubuntu 12.04?
> >
> > Thanks ahead,
> > Sam
> >
> >
> > On Mon, Aug 27, 2012 at 5:56 PM, Alejandro Comisario <alejandro.comisario at mercadolibre.com> wrote:
> > One of the things i dont like in essex.
> > That the "autostart" flag in nova.conf with KVM doesnt work with the autostart feature of libvirt/kvm, so if, for some reason you need to restart nova-compute to apply some kind of modification, the instances get soft/hard rebooted because now nova-compute handles the "autostart" flag from nova.conf
> >
> > Why is that if some one can explain ?
> >
> > On Mon, Aug 27, 2012 at 9:21 PM, Gabe Westmaas <gabe.westmaas at rackspace.com> wrote:
> > Hey Sam,
> >
> >
> >
> > Is it possible your hypervisors restarted?  I see this entry in the logs:
> >
> >
> >
> > 2012-08-23 06:35:02 INFO nova.compute.manager [req-f1598257-3f35-40e6-b5aa-d47a0e93bfba None None] [instance: ce00ff1d-cf46-44de-9557-c5a0f91c8d67] Rebooting instance after nova-compute restart.
> >
> >
> >
> > Gabe
> >
> >
> >
> > From: openstack-bounces+gabe.westmaas=rackspace.com at lists.launchpad.net [mailto:openstack-bounces+gabe.westmaas=rackspace.com at lists.launchpad.net] On Behalf Of Sam Su
> > Sent: Monday, August 27, 2012 8:10 PM
> > To: openstack
> > Subject: [Openstack] HELP: All instances automatically rebooted.
> >
> >
> >
> > Hi,
> >
> >
> >
> > I have an Essex cluster with 6 compute nodes and one control nodes. All compute nodes are working not any interrupted, for some reason all instances in my cluster automatically rebooted. I am trying to but not figured out why this happened in these couple of days.
> >
> >
> >
> > It's much appreciated if someone can give me some hints about how to deal with this situation.
> >
> >
> >
> > Logs in my /var/log/upstart/nova-compute.log:
> >
> > http://pastebin.com/WYJtS5a5
> >
> >
> >
> > Let me know if need more info.
> >
> >
> > TIA,
> >
> > Sam
> >
> >
> >
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~openstack
> > Post to     : openstack at lists.launchpad.net
> > Unsubscribe : https://launchpad.net/~openstack
> > More help   : https://help.launchpad.net/ListHelp
> >
> >
> >
> >
> > --
> > Alejandro
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~openstack
> > Post to     : openstack at lists.launchpad.net
> > Unsubscribe : https://launchpad.net/~openstack
> > More help   : https://help.launchpad.net/ListHelp
> 
> 
> _______________________________________________
> Mailing list: https://launchpad.net/~openstack
> Post to     : openstack at lists.launchpad.net
> Unsubscribe : https://launchpad.net/~openstack
> More help   : https://help.launchpad.net/ListHelp

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20120830/ff8feb6e/attachment.html>