Xen users - Feb 2009 - Xen 3.3.1 + Windows 2008 x64 + GPLPV 0.9.12-pre13 = corruption?

Hi James,

I''m testing GPLPV 0.9.12-pre13 on Windows 2008 x64, and found some
interesting results :
- The drivers are NOT signed, even though I''ve install certs. Had to
use F8 on boot.
- both QEMU and Xen block device are active, which usually means
corruption is coming. Is this a known issue?
http://i44.tinypic.com/2vmypvk.png
http://i42.tinypic.com/2mewnia.png

Regards,

Fajar

_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users

> 
> Hi James,
> 
> I''m testing GPLPV 0.9.12-pre13 on Windows 2008 x64, and found some
> interesting results :
> - The drivers are NOT signed, even though I''ve install certs. Had
to
> use F8 on boot.
> - both QEMU and Xen block device are active, which usually means
> corruption is coming. Is this a known issue?
> http://i44.tinypic.com/2vmypvk.png
> http://i42.tinypic.com/2mewnia.png
> 
Yes, that probably isn''t good news. With a recent version of Xen and
phy: or tap:aio, windows seems to pick up that one drive is an
''alternate path'' to the other and ignores one, otherwise
things will go
bad pretty quickly.

I''d need to see the windows debugger output on boot to know what was
going on though.

James

_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users

James Harper wrote:
>> - both QEMU and Xen block device are active, which usually means
>> corruption is coming. Is this a known issue?
>> http://i44.tinypic.com/2vmypvk.png
>> http://i42.tinypic.com/2mewnia.png
>>
>>     
>
> Yes, that probably isn''t good news. With a recent version of Xen
and
> phy: or tap:aio, windows seems to pick up that one drive is an
> ''alternate path'' to the other and ignores one, otherwise
things will go
> bad pretty quickly.
>
> I''d need to see the windows debugger output on boot to know what
was
> going on though.
>
>   
And how can I get that? It should involve something like a serial
console so that the output can be captured, right?



_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users

> >
> > Yes, that probably isn''t good news. With a recent version of
Xen and
> > phy: or tap:aio, windows seems to pick up that one drive is an
> > ''alternate path'' to the other and ignores one,
otherwise things will go
> > bad pretty quickly.
> >
> > I''d need to see the windows debugger output on boot to know
what was
> > going on though.
> >
> >
> 
> And how can I get that? It should involve something like a serial
> console so that the output can be captured, right?
I think I documented it on the wiki. Basically yes, you use a virtual serial
port on the DomU and connect the debugger to it.

I''m nearing the end of a pretty major set of changes to gplpv, but
haven''t
tested x64 yet... hopefully I''ll have some good news shortly.

James


_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users