Re: [Qemu-devel] Re: [PATCH 04/22] savevm: do_loadvm(): Always resume the VM

[Jamie Lokier <jamie@shareable.org>]

Juan Quintela wrote:
> Luiz Capitulino <lcapitulino@redhat.com> wrote:
> > On Wed, 21 Apr 2010 15:28:16 +0200
> > Kevin Wolf <kwolf@redhat.com> wrote:
> I tried a variation of this in the past, and was not a clear agreement.
> 
> Basically, after a working migration to other host, you don't want to
> allow "cont" on the source node (it target has ever changed anything, it
> would give disk corruption).

This is not true if the target is using a copy of the disk.

Making copies is cheap on some hosts (Linux btrfs with it's COW features).

Forking a guest can be handy for testing things, starting from a known
run state.  The only thing to get confused is networking because of
duplicate addresses, and there are host-side ways around that (Linux
network namespaces).

If I understand correctly, we can already do this by migrating to a
file and copying the files.  There's no reason to block the live
equivalent, provided there is a way to copy the disk image when it's
quiesced.

So it's wrong to block "cont" on the source, but 
"cont --I_know_what_I_am_doing" might be good advice :-)

> But my suggestion to disable "cont" after that got complains that people
> wanted a "I_know_what_I_am_doing_cont". (not the real syntax).  Perhaps
> it is time to revise this issue?

-- Jamie