Re: [Qemu-block] qemu exit with pending I/O

[Peter Lieven]

Am 27.04.2015 um 12:47 schrieb Paolo Bonzini:
> On 16/04/2015 10:46, Peter Lieven wrote:
> > I just run tests with the new libiscsi branch that allows async
> > reconnects. I found that
> > I cannot quit qemu while qemu is waiting for a reconnect to the storage.
> > E.g.
> > I start qemu and then shut down the storage. Qemu will then loop forever in
> > bdrv_flush() and/or bdrv_drain_all().
> >
> > It would be nice to be able to shut down qemu via normal quit and not have
> > to kill it in such a case.
> The reason to do this, is that while I/O is pending you do not know if
> the I/O has been completed or not.

But if I consider a raw image, quitting without a clean shutdown can
also leave an inconsistent state?

Peter

> It's then possible that the I/O is completed after some writes from
> another invocation of the VM, causing data corruption.  For the same
> reason the NBD driver does not do timeouts.
>
> Basically, bdrv_drain_all() is there to emulate nfs=hard.  You always
> have SIGKILL, so it's not exactly the same, but leaving it out would
> have the same problems as nfs=soft.
>
> Paolo


--

Mit freundlichen Grüßen

Peter Lieven

...........................................................

  KAMP Netzwerkdienste GmbH
  Vestische Str. 89-91 | 46117 Oberhausen
  Tel: +49 (0) 208.89 402-50 | Fax: +49 (0) 208.89 402-40
  address@hidden | http://www.kamp.de

  Geschäftsführer: Heiner Lante | Michael Lante
  Amtsgericht Duisburg | HRB Nr. 12154
  USt-Id-Nr.: DE 120607556

...........................................................