[OpenAFS] Version (was Re: Error when moving volumes)

[Horst Birthelmer]

On Wednesday, June 30, 2004, at 01:09  PM, Christian Pfaffel wrote:

> Dr A V Le Blanc <LeBlanc@mcc.ac.uk> writes:
>
>> On Mon 28 Jun 2004 at 17:57:30 -0400, Jeffrey Hutzelman 
>> <jhutz@cmu.edu> wrote:
>>> To: Frode Nilsen <mailing-lists@cyberpunks.no>
>>> Cc: openafs-info@openafs.org
>>> Subject: Re: [OpenAFS] Error when moving volumes
>>>
>>> I suggest you send the output of
>>>
>>> rxdebug marvin 7005 -version
>>>
>>> which will indicate the actual version of the running volserver.
>>
>> I'm a little doubtful of the accuracy of this; I'm running server
>> binaries compiled from Debian's openafs-1.2.11-1 source and diffs;
>> when I run rxdebug on my servers, I get this result:
>>
>>> rxdebug xxxxx 7005 -version
>>      Trying 130.88.xxx.xx (port 7005):
>>      AFS version:  OpenAFS 1.2.10 built  2004-01-12
>>
>> (the source was released on January 11 at
>> www-openafs.central.org/release/latest.html.)
>>
>> Of course, the number reported is the one in the compiled source;
>> someone has released a version which contains the file 
>> configure-libafs
>> apparently generated from an old template.  If I look at the debian
>> binary for vlserver from the same site, I find this:
>>
>>      OpenAFS 1.2.10 built  2004-01-11
>>
>> Admittedly Debian's build procedure is a little different, but
>> can it not be that other binaries have the wrong version number
>> built into them?
>>
>
> I filed a bug report concerning this yesterday:
>
>         http://rt.central.org/rt/Ticket/Display.html?id=5553
>

The OpenAFS version 1.2.11 was just a "hotfix" for a ubik problem 
occured on Jan. 11.
It makes you loose your quorum ...

So, if somebody just applied the patch and never changed the version 
number that's no big deal and IMHO no sin at all :-))
Maybe the Debian maintainer should have done that to avoid some 
irritation.

Horst