[OpenAFS] Windows: aklog: ktc 7 (11862791) while obtaining tokens

[Mickey Lane]

I have Windows 7 Ultimate on the desktop machine here.

With no changes:

AFS works fine at startup. If I create a VPN (Microsoft software) to an adj=
acent Server 2008 machine, AFS fails. If I disconnect the VPN, AFS works ag=
ain (no reboot). This is repeatable.

If I check the two Link-layer Topology boxes on the loopback adaptor proper=
ties box and reboot, I get:

AFS works fine at startup. Create a VPN to Server machine, AFS fails. Wait =
3-4 minutes, AFS starts working again.

I have not had anyone verify this yet. I've observed the same on another Wi=
ndows 7 virtual machine but I don't remember what flavor of Windows 7 it wa=
s.

Mickey.


> -----Original Message-----
> From: openafs-info-admin@openafs.org [mailto:openafs-info-
> admin@openafs.org] On Behalf Of Richard Brittain
> Sent: Friday, July 30, 2010 11:56 AM
> To: Jeff Blaine
> Cc: jaltman@secure-endpoints.com; openafs-info@openafs.org
> Subject: Re: [OpenAFS] Windows: aklog: ktc 7 (11862791) while obtaining
> tokens
>=20
> On Wed, 23 Jun 2010, Jeff Blaine wrote:
>=20
> > * There is a bug in Windows that will prevent
> >  access to \\AFS after an IP address has
> >  been removed or assigned after boot.  When
> >  the bug is triggered, all attempts to connect
> >  to \\AFS will result in a "Bad Network Name"
> >  error.  Please reproduce this issue locally
> >  and submit bug reports to Microsoft. .............. below
> > """
>=20
> For what it's worth, we now have two examples of "Windows 7 Enterprise"
> (32-bit) systems which seem to have no problem at all accessing AFS
> after
> an address change (typically 2 minutes needed to re-establish access
> after
> a change).  We also have one example of "Windows 7 Professional" which
> clearly follows the pattern described above.
>=20
> From the published descriptions of differences between different
> versions
> on Windows 7, I could believe that the virtualization features of
> Enterprise might be responsible for the difference.  From the
> description,
> it seems that "Ultimate" stands a chance of working too, but I've got
> no
> Ultimate system to test on.
> My guess is that all the home editions would also fail as above.
>=20
> I also discovered that on Ultimate laptops using only secure wireless
> connections, this is a complete show stopper, since they boot with no
> network, and transition to a real address only after login and wireless
> authentication, and the loopback adaptor comes 'pre-hosed' to the login
> session.
>=20
> Richard
> --
> Richard Brittain,  Research Computing Group,
>                     Kiewit Computing Services, 6224 Baker/Berry Library
>                     Dartmouth College, Hanover NH 03755
> Richard.Brittain@dartmouth.edu 6-2085
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info