On Wed, Sep 15, 2021 at 10:57:23 +0100, Daniel P. Berrangé wrote: > On Wed, Sep 15, 2021 at 11:47:04AM +0200, Christian Borntraeger wrote: > > > > > > On 15.09.21 11:45, Christian Borntraeger wrote: > > > On 24.08.21 16:44, Peter Krempa wrote: > > > > The qemu driver didn't ever implement any meaningful handling for the > > > > 'preserve' action. > > > > > > > > Forbid the flag in the qemu def validator and update the documentation > > > > to be factual. > > > > > > > > Signed-off-by: Peter Krempa <pkrempa@xxxxxxxxxx> > > > > > > NACK. It is a perfectly sane usecase to have "preserve" on a crash. The qemu > > > will simply sit in that state and the user can then do virsh dump or similar. > > > > > > > To make this statement stronger. I use this A LOT in real life to force libvirt > > to leave the QEMU in that state to get dump and debug info. > > We probably just need to add a commentin libvirt source somewhere > relevant to the effect that "preserve" is the default behaviour > from QEMU so libvirt doesn't need todo anything extra. Hmm, yeah for 'onCrash' it actually might make sense. Unfortunately it was totally broken for onReboot and onShutdown. I'll remove the check for onCrash, but for the other two I'll leave it in as it didn't work at all and I don't really care enough to implement it.
