Hello, I used qemu-img-rhev-0.12.1.2-2.355.el6 successfully with -s parameter, but now qemu-img-rhev-0.12.1.2-2.415.el6 (latest atm) errors out with: convert: invalid option -- 's' What am I missing? Do I need any special rpmbuild switches or something? I see the spec already says "rhev 1" which should be enough. Regards, Lucian

On 03.01.2014 13:45, Dafna Ron wrote: > yes, there are bugs reported in openstack for the same issue on > qemu-kvm: > > https://bugzilla.redhat.com/show_bug.cgi?id=1027074 > https://bugzilla.redhat.com/show_bug.cgi?id=1016896 > Hi and thanks, I cannot access https://bugzilla.redhat.com/show_bug.cgi?id=1016896 with my user. Care to sum it up, is there a resolution in sight? Regards, Lucian

On 03.01.2014 15:17, Dafna Ron wrote: > On 01/03/2014 02:43 PM, Nux! wrote: >> On 03.01.2014 13:45, Dafna Ron wrote: >>> yes, there are bugs reported in openstack for the same issue on >>> qemu-kvm: >>> >>> https://bugzilla.redhat.com/show_bug.cgi?id=1027074 >>> https://bugzilla.redhat.com/show_bug.cgi?id=1016896 >>> >> >> Hi and thanks, >> >> I cannot access https://bugzilla.redhat.com/show_bug.cgi?id=1016896 >> with my user. >> Care to sum it up, is there a resolution in sight? >> >> Regards, >> Lucian > > well, there was a decision in virt to no longer skip the -s and > prompt an error message which was decided to be the correct way to go. > I am adding some people to this, maybe they can help with some > workaround for ovirt. Thanks, I don't want to appear cheeky, but I did not need the feature for Ovirt. Currently KVM snapshots in Cloudstack are broken as well because the "-s" went away. I have found that qemu-img-rhev-0.12.1.2-2.355 still has the capability so that's what I'm stuck with on my hypervisor; thought you guys know more. Lucian

On Fri, Jan 03, 2014 at 01:34:39PM +0000, Nux! wrote: > Hello, > > I used qemu-img-rhev-0.12.1.2-2.355.el6 successfully with -s > parameter, but now qemu-img-rhev-0.12.1.2-2.415.el6 (latest atm) > errors out with: > convert: invalid option -- 's' Could you provide more context, preferably with the traceback from vdsm.log?