On 03/24/17 14:23, Peter Krempa wrote: > On Fri, Mar 24, 2017 at 14:04:08 +0100, Laszlo Ersek wrote: >> On 03/24/17 13:46, Peter Krempa wrote: >>> On Thu, Mar 23, 2017 at 17:49:56 +0100, Laszlo Ersek wrote: >>>> On 03/23/17 15:07, Peter Krempa wrote: > > [...] > >>> >>> I'll post a v3 with the option to override it, if users insist that they >>> don't care about the state of their varstore. >> >> Thank you, this all sounds good to me. > > I've already posted v3, but forgot to CC you: > > https://www.redhat.com/archives/libvir-list/2017-March/msg01151.html > > >> >> Could you please test v3 with both live (online) and offline OVMF VMs, >> with and without --unsafe? (That is, four cases in total.) > > Sigh. I've thought that I tested this properly, but libvirt code is not > able to detect the failure from taking the snapshot and thus thinks it > created the snapshot properly. > > I've got: > > {"return": "Device 'pflash1' is writable but does not support snapshots.\r\n", "id": "libvirt-37"} > > This is due to the fact that 'savevm' uses HMP passthrough and the code > to detect errors sucks. > > Sorry for the noise. > > v3 is not necessary with all the fancy code. I'll push v2 with a better > explanation, since it fixes an actual bug :/ Thanks! It's just as well that we've been thorough enough to track this down, so that's a plus at least. Laszlo -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list
