[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] Live migration results in non-working virtio-net device
From: |
Neil Skrypuch |
Subject: |
Re: [Qemu-devel] Live migration results in non-working virtio-net device (sometimes) |
Date: |
Wed, 05 Mar 2014 13:32:51 -0500 |
User-agent: |
KMail/4.10.2 (Linux/3.9.1-gentoo-r1; KDE/4.10.2; x86_64; ; ) |
On Wednesday 05 March 2014 16:59:24 Andreas Färber wrote:
> Am 30.01.2014 19:23, schrieb Neil Skrypuch:
> > First, let me briefly outline the way we use live migration, as it is
> > probably not typical. We use live migration (with block migration) to
> > make backups of VMs with zero downtime. The basic process goes like this:
> >
> > 1) migrate src VM -> dest VM
> > 2) migration completes
> > 3) cont src VM
> > 4) gracefully shut down dest VM
> > 5) dest VM's disk image is now a valid backup
> >
> > In general, this works very well.
> >
> > Up until now we have been using qemu-kvm 1.1.2 and have not had any
issues
> > with the above process. I am now attempting to upgrade us to a newer
> > version of qemu, but all newer versions I've tried occasionally result in
> > the virtio- net device ceasing to function on the src VM after step 3.
>
> While I don't know this particular symptom, I can definitely tell you
> that migrating from qemu-kvm to qemu is bound to fail unless you enable
> at least a version_id change in piix4.c, possibly also in kvmvapic.c.
> Such errors would lead to migration not successfully completing though,
> with cryptic error on the dest side.
>
> Regards,
> Andreas
I should clarify, all of these migrations happen from same version to same
version (and same host to same host). So 1.7.0 -> 1.7.0, 1.6.0 -> 1.6.0, etc.
What we're looking for out of this is a clean copy of the disk image
(consistent and from a graceful shutdown) for backup purposes, without having
to shut down the VM.
I expected cross-version migration to be dicey and made a point of avoiding
it.
- Neil