Re: [libvirt] [PATCH v2] snapshot: Store both config and live XML in the snapshot domain

On Wed, Jul 31, 2019 at 14:52:17 -0500, Eric Blake wrote: > On 7/31/19 2:45 PM, Daniel Henrique Barboza wrote: >> Max, >> >> Code looks ok. Two tests (virsh-checkpoint and virsh-snapshot) are >> failing, but they are also failing on master, thus I say this patch get a >> pass because it didn't break anything else. > > What failures are you seeing? Those were just recently added, and if > they are failing for you, it's worth getting them fixed. But I'm not > seeing them fail on my end. > >> On 7/23/19 4:47 PM, Maxiwell S. Garcia wrote: >>> The snapshot-create operation of running guests saves the live >>> XML and uses it to replace the active and inactive domain in >>> case of revert. So, the config XML is ignored by the snapshot >>> process. This commit changes it and adds the config XML in the >>> snapshot XML as the <inactiveDomain> entry. > > Since checkpoints are brand new, and also created always on a running > image, should they also gain an <inactiveDomain> entry? And if we are > fast enough, would it be worth mandating that entry on a checkpoint > REDEFINE (even though we can't do it for a snapshot REDEFINE)? Can you actually revert to a checkpoint? I don't think so, which means there's no reason for storing the inactive XML for it.