Enrico M. V. Fasanelli wrote:

> Dear Stephan,
>
> the solution comes from google (as it happens frequently... see
> https://lists.openafs.org/pipermail/openafs-devel/2004-May/
> 010428.html).
>
> I've  applied the last patch supplied by Chris Wing (I've renamed it
> into openafs-krb5-64bit.patch) and produced the
> openafs-1.2.11-15.10.SL.src.rpm and corresponding binaries with
> *WORKING* aklog (yust checked !).
>
> You can  download them  starting from
> ftp://ftp.le.infn.it/linux/scientific/[SRPMS || RPMS] (or copy directly
> from /afs/le.infn.it/ftp/pub/linux/scientific/)
>
> Troy, Connie,  I think that is better to include this new version in
> the distribution (if Scientific Linux release policies allows).
>
> Ciao
>                        Enrico
>
>
> On 15 Jan 2005, at 10:18, Stephan Wiesand wrote:
>
>> Dear Enrico,
>>
>> this is a known "feature" of the krb5 packages - see the thread on
>> scientific-linux-users titled "krb5/afs on SL 3.0.2 (x86_64) ?".
>>
>> Workarounds I've found so far:
>>
>>    - the 32bit aklog binary works
>>
>>    - rebuilding krb5 with patch37 enabled makes afslog work,
>>      but neither aklog nor pam_krb5afs.so
>>
>>    - SuSE 9.0/amd64 ships with a pam_krb5afs.so that works and can
>>      be used on SL3/amd64 if you pull in the needed shared libs
>>      (heimdal, ...) as well, but there are issues with at least the
>>      configuration (heimdal slightly incomatible with MIT) and
>>      ticket lieftimes
>>
>> If anyone comes up with something better, I'd be glad to hear about it.
>>
>> Cheers,
>>       Stephan
>>
>> --
>>
>>   ----------------------------------------------------
>> | Stephan Wiesand  |                                |
>> |                  |                                |
>> | DESY     - DV -  | phone  +49 33762 7 7370        |
>> | Platanenallee 6  | fax    +49 33762 7 7216        |
>> | 15738 Zeuthen    |                                |
>> | Germany          |                                |
>>   ----------------------------------------------------
>>
>>
> ---
> Enrico M. V. Fasanelli - I.N.F.N. Sezione di Lecce

Whoo hoo.
I haven't tested it yet, but I believe you and will run the tests
tomorrow.  If all works well we'll  get it integrated in in time for
S.L. 3.0.4
Troy