x86-64 rawhide update obnoxiousness

[Panu Matilainen]

On Sun, 16 Oct 2005, Michal Jaegermann wrote:
> On Sun, Oct 16, 2005 at 02:02:52PM +0100, Paul Jakma wrote:
>> On Fri, 14 Oct 2005, Panu Matilainen wrote:
>>
>>> when you install/upgrade both the i386 and x86_64 packages
>>> simultanously, rpm "swallows" file conflicts between multilib
>>> packages.
>>
>> Aha. Thanks for explaining that.
>
>>
>>> Whether that's a sane thing to do is another question.
>>
>> It's dangerous / not sane IMHO.
>
> I am curious how you would propose to resolve that taking into
> account all existing packages and this detail that people making
> original packages may be often unaware of the issue and even if they
> are they may not have suitable installations for testing.  Multiply
> that by 'extras', external repositories, and so on.  If you will get
> strict here then resulting pains will surely far exceed your worst
> current hiccups.
>
>> The worst case I had was where some key arch-dependent glibc files
>> got clobbered (FCtest days) - rescue CD time.
>
> AFAICT dependencies are really "swallowed" during an installation if
> files in question are the same - which is easy to test.  At least if
> the whole concept is not buggy It appears that you were bitten by
> something else or you run into some traps in testing but that what
> testing is for.

If it only "swallowed" identical files I wouldn't mind either but consider 
this:
[root at linox01 xxx]# rpm -q tcsh
package tcsh is not installed
[root at linox01 xxx]# rpm -Uvh tcsh-6.13-9.i386.rpm tcsh-6.13-9.x86_64.rpm
Preparing...                ########################################### 
[100%]
    1:tcsh                   ########################################### [ 
50%]
    2:tcsh                   ########################################### 
[100%]
[root at linox01 xxx]# file /bin/tcsh
/bin/tcsh: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for 
GNU/Linux 2.4.0, dynamically linked (uses shared libs), stripped

Notice the lack of any warnings or complaints from rpm. In fact it doesn't 
even need to be in the same transaction apparently:

[root at linox01 xxx]# rpm -e --allmatches --nodeps tcsh
[root at linox01 xxx]# rpm -ivh tcsh-6.13-9.x86_64.rpm
Preparing...                ########################################### 
[100%]
    1:tcsh                   ########################################### 
[100%]
[root at linox01 xxx]# rpm -ivh tcsh-6.13-9.i386.rpm
Preparing...                ########################################### 
[100%]
    1:tcsh                   ########################################### 
[100%]
[root at linox01 xxx]# file /bin/tcsh
/bin/tcsh: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for 
GNU/Linux 2.4.0, dynamically linked (uses shared libs), stripped

..and again zero complaints from rpm, the 32bit version of tcsh only 
exists in rpmdb in reality.

 	- Panu -