Re: [gtk-vnc-devel] [PATCH 0/7] Optimizations for high-latency X connections

[Anthony Liguori <anthony codemonkey ws>]

Daniel P. Berrange wrote:
> On Wed, Dec 17, 2008 at 07:46:37PM -0600, Anthony Liguori wrote:
>   
> > Federico Mena Quintero wrote:
> >     
> > > Hi, everyone,
> > >
> > > I mentioned in #virt a few weeks ago that I was working on making
> > > gtk-vnc be nominally as fast as TightVNC.  This came out from an
> > > Important Customer(tm) who wanted to use virt-viewer, but who found it
> > > to be too slow.
> > >
> > > It turns out that they were on a crazy setup with a high-latency remote
> > > X connection.  My patches implement some tricks from TightVNC to handle
> > > such connections:
> > >   
> > >       
> > Can I ask, why in the world is your Important Customer(tm) forwarding 
> > gtk-vnc over SSH via X instead of doing VNC over X?  That would solve 
> > all of these problems.
> >     
>
> The former is securely authenticated and encryption, the latter is often
> not - unless using the GTK-VNC + QEMU  VeNCrypt extension, but I can well
> imagine there's plenty of people who don't do that, particularly if they
> are a SSH GSSAPI+Kerberos enabled environment.
>   

You can forward the VNC traffic over SSH.

> Both are valid deployment use cases and we shouldn't presume one vs the
> other, although obviously we can make recommendations.
>   

I'm not sure it's worth jumping through hoops to support high-latency 
connections between the X client and server.  That seems counter 
productive to me.

Regards,

Anthony Liguori

> Daniel
>