[OpenAFS] com_err hell (WAS: asetkey: failed to set key, code 70354694)

[Marcus Watts]

Derrick J Brashear <shadow@dementia.org> writes:
> From: Derrick J Brashear <shadow@dementia.org>
> To: Marcus Watts <mdw@umich.edu>
> Cc: openafs-info <openafs-info@openafs.org>
> Subject: Re: [OpenAFS] com_err hell (WAS: asetkey: failed to set key, code
>  70354694) 
> Date: Tue, 10 Apr 2007 15:25:08 -0400 (EDT)
> 
> On Tue, 10 Apr 2007, Marcus Watts wrote:
> 
> > Carson Gaspar <carson@taltos.org>
> >> From: Carson Gaspar <carson@taltos.org>
> >> To: openafs-info <openafs-info@openafs.org>
> >> References: <200704091946.PAA15130@quince.ifs.umich.edu>
> >> In-Reply-To: <200704091946.PAA15130@quince.ifs.umich.edu>
> >> Subject: [OpenAFS] com_err hell (WAS: asetkey: failed to set key, code 70354694)
> >> Sender: openafs-info-admin@openafs.org
> >> Date: Tue, 10 Apr 2007 09:13:58 -0700
> >>
> >> Re: the whole com_err mess...
> >>
> >> If AFS has an incompatible implementation, why not just rename the
> >> exported functions / vars and get it over with? This should be no more
> >> than a day's work, and would solve the problem very nicely. Or am I
> >> missing something?
> >
> > From an application standpoint that would be a very
> > disappointing answer - it means you can't really have
> > one error number space anymore.
> 
> Well, for the moment you can't anyway, really. We need to find a way to 
> unify with Heimdal, ext2 and MIT Krb5 but until we do we can at least make 
> it possible to not lose.

The implementation of com_err that's in the rxk5 patch actually can
do this.  The last asetkey session I posted had errors from both mit
kerberos & afs going through one com_err.

Granted, it's not as pretty as it should be, and it would be good
for all those groups you named to come to a better consensus as to
how this should all work.

			-Marcus Watts