[one-users] Openvswitch db corruption
Hello everyone, I am facing some weird behaviours in a full openvswitch based environment (3.6 and KVM based). We faced some power outages in the last month that caused VMs to fail. The main reason is that the interfaces are not available for the instances after we restart them and being unable to bind them to the instances we had several failing images (testing env). I don't know what could be related to that. In 3.2 we were able to develop some scripts to solve those problems tho. I would like to know if anyone else is facing this same issue after an unclean shutdown. Cheers -- Have you tried turning it off and on again? ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Openvswitch db corruption
Hello, This happened to me yesterday, elecritic tension failure, all servers rebooted, VMs unknown, everything failed... Don't know... CyD Blog : http://blog.cduverne.com Twitter : @CydsWorld Imagination is more important than Knowledge Albert Einstein Mercredi 03/10/2012 à 13:31 Rodolfo Conte Brufatto a écrit: Hello everyone, I am facing some weird behaviours in a full openvswitch based environment (3.6 and KVM based). We faced some power outages in the last month that caused VMs to fail. The main reason is that the interfaces are not available for the instances after we restart them and being unable to bind them to the instances we had several failing images (testing env). I don't know what could be related to that. In 3.2 we were able to develop some scripts to solve those problems tho. I would like to know if anyone else is facing this same issue after an unclean shutdown. Cheers -- Have you tried turning it off and on again? ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Openvswitch db corruption
Hi CyD. I am developing a faul tolerant script when these events happen. By checkin the uptime of the vmhost, state of the instances and the openvswitch db. My recommendation is a basic check list which involves primarily removing the vnetX associated in the bridges, with this ou will be good to go. (using ovs-vsctl to delete these ports) However, I am concerned with the resonse time to not lose the image contents when it fails and the re-deploy time. If you guys know any other solution... Cheers On Wed, Oct 3, 2012 at 9:14 AM, Duverne, Cyrille cyrille.duve...@euranova.eu wrote: Hello, This happened to me yesterday, elecritic tension failure, all servers rebooted, VMs unknown, everything failed... Don't know... CyD Blog : http://blog.cduverne.com Twitter : @CydsWorld Imagination is more important than Knowledge Albert Einstein Mercredi 03/10/2012 à 13:31 Rodolfo Conte Brufatto a écrit: Hello everyone, I am facing some weird behaviours in a full openvswitch based environment (3.6 and KVM based). We faced some power outages in the last month that caused VMs to fail. The main reason is that the interfaces are not available for the instances after we restart them and being unable to bind them to the instances we had several failing images (testing env). I don't know what could be related to that. In 3.2 we were able to develop some scripts to solve those problems tho. I would like to know if anyone else is facing this same issue after an unclean shutdown. Cheers -- Have you tried turning it off and on again? -- Have you tried turning it off and on again? ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
