On 5/7/21 3:46 PM, Daniel P. Berrangé wrote: > On Fri, May 07, 2021 at 03:43:32PM +0200, Michal Prívozník wrote: >> On 5/7/21 2:33 PM, Daniel P. Berrangé wrote: >>> I'm not going to apply this yet as I realized that I have broken >>> internal secrets with monolithic libvirtd instead :-( >> >> You did? I've tested it (by starting a pool with a secret) and it worked. > > Secrets aren't used when starting a pool with merely encrypted vols present. > > They're only needed to perform I/O, so would have to test starting a guest, > or cloning a encrypted vol. Alternatively starting a RBD/iSCSI pool that > requires auth. Yeah, it's an iSCSI pool that requires auth. But okay, I'll try > > I'm testing with starting a QEMU guest with an encrypted vol this for v2. Michal
