Re: [PATCH RFC] migration: warn about non-migratable configurations unless '--no-migration' was specified

From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Eduardo Habkost <ehabkost@redhat.com>
Cc: Peter Maydell <peter.maydell@linaro.org>,
	Juan Quintela <quintela@redhat.com>,
	QEMU Developers <qemu-devel@nongnu.org>,
	Markus Armbruster <armbru@redhat.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Vitaly Kuznetsov <vkuznets@redhat.com>,
	"Dr . David Alan Gilbert" <dgilbert@redhat.com>
Subject: Re: [PATCH RFC] migration: warn about non-migratable configurations unless '--no-migration' was specified
Date: Mon, 19 Apr 2021 17:37:32 +0100	[thread overview]
Message-ID: <YH2xzIlCfjwdVF2R@redhat.com> (raw)
In-Reply-To: <CAOpTY_oCNcPjB6aHn6kp6Pb=TJSYKLuTOWyLD+1LP24w3F_c=A@mail.gmail.com>

On Mon, Apr 19, 2021 at 12:28:04PM -0400, Eduardo Habkost wrote:
> On Sun, Apr 18, 2021 at 11:54 AM Peter Maydell <peter.maydell@linaro.org> wrote:
> >
> > On Thu, 15 Apr 2021 at 16:46, Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
> > > When a migration blocker is added nothing is reported to the user,
> > > inability to migrate such guest may come as a late surprise. As a bare
> > > minimum, we can print a warning. To not pollute the output for those, who
> > > have no intention to migrate their guests, introduce '--no-migration'
> > > option which both block the migration and eliminates warning from
> >
> > I'm not a fan. For a lot of people and configurations this
> > is going to be "add an extra complaint from QEMU to a previously
> > working configuration". We add too many of those already.
> 
> I agree that warning with machine types that never supported live
> migration would be useless noise, but warning if using an explicit
> versioned machine type sounds like a reasonable default (as long as
> the warnings includes clear instructions on how to silence them).

Libvirt will always expand a machine type alias into a versioned
machine type, because stable ABI is useful even if never migrating,
because it ensures the guest OS doesn't see hardware changes that
may trigger license re-activation

At the same time a large portion of users will never care about
live migration/save/restore, or they do care but will hotunplug
the problems devices before attempting a migration.

IMHO tieing messages to versioned machine types is not desirable
as a strategy by default.

Warning about migration compatibility should be an explicit opt-in

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|