Re: [Qemu-devel] [PATCH] sockets: ensure we can bind to both ipv4 & ipv6 separately

[Gerd Hoffmann <kraxel@redhat.com>]

On Di, 2017-05-16 at 14:35 +0100, Daniel P. Berrange wrote:
> On Tue, May 16, 2017 at 03:29:51PM +0200, Gerd Hoffmann wrote:
> >   Hi,
> > 
> > > Separately from this patch, I noticed one further possible
> > > problem, 
> > > 
> > > Currently, the ipv4=on|off/ipv6=on|off settings are only
> > > used to determine what getaddrinfo results we request/use.
> > > 
> > > This leads to the somewhat odd situation where if you set
> > > ipv4=off,ipv6=on, QEMU won't be listening on an IPv4
> > > socket, but *will still* accept IPv4 clients over the IPv6
> > > socket due to our use of IPV6_V6ONLY=off.
> > 
> > Hmm, maybe we should just use IPV6_V6ONLY=on unconditionally?
> 
> This would certainly be my long term desire...
> 
> > Now that qemu finally supports multiple listening sockets we should be
> > able to do the switch without regressions, and it also should remove all
> > those nasty corner cases ...
> 
> ...but we can't do it yet - only the VNC server code has been reworked
> to support multiple listening sockets. We'd still have work todo on the
> NBD server, chardevs, migration, network socket backend and guest agent
> to support multiple listeners.

Ah, ok.

I guess then I would not worry too much that we break stuff in case
ipv4=off actually does what it is supposed to do ...

cheers,
  Gerd