BF works under Fedora Core 2 on an AMD

[Matthias Andree]

David Relson wrote:
> On Tue, 13 Dec 2005 17:28:26 -0500
> Al Donaldson wrote:
> 
>> I've downloaded bogofilter-1.0.0-1.i586.rpm and it runs fine on a Fedora 
>>   Core 3 (Celeron).  The RPM installs on a Red Hat 9 system (AMD Duron), 
>> but bogofilter dumps core.

What signal does it catch?

>> Don't have a debugger on that machine, so 
>> not sure why.  Both machines show up as i686 (uname -m), which leaves me 
>> wondering if it is CPU issue since the RPM is i586.rpm.
>>
>> Or is it a compatibility problem with Red Hat 9?  I have customers 
>> running everything from RH 7.1 to Fedora Core 3, and would prefer to 
>> point them to a bogofilter-static RPM instead of having to build it for 
>> them on n different platforms.  Any suggestions?
>>
>> This started out as a follow-up to ??David?? in a posting of the same 
>> subject found by Google.  Apologies if the matter is explained 
>> elsewhere, but I haven't found it.
>>
>> Al
> 
> Hello Al,
> 
> Sorry for the delay in getting your message to the list.  As moderator
> I only check for non-subscriber posts once per day.  I suggest you
> subscribe to the list for quicker response :->
> 
> AMD Duron should be not be a problem.  The 1.0.0 rpms were built on my
> Athlon running Mandriva 2006.0.
> 
> Under what sorts of conditions is bogofilter crashing?  What options
> are you using in your command line?  Have you had bogofilter on the
> system before now?  In the past we've seen problems with obsolete
> config file options causing trouble.  We think all those problems are
> fixed, but it's hard to be certain.  If there's a config file, try
> using the "-C" option (no config file) with bogofilter and see if that
> helps.  One can also run "strace -feopen bogofilter ..." to see how far
> bogofilter is getting.
> 
> If you can gzip a core file (or two) and email it to me
> (relson at bogofilter.org will do), I can take a look.  I'll be off-line
> for a couple of days starting tomorrow, so act quickly (if you can).
> 
> Alternatively, someone else on the list may be aware of the problem and
> have a solution.

I'd also offer to take a look at the core files if necessary. I'd expect CPU 
issues to show up as SIGILL, perhaps (rare) SIGBUS. SIGSEGV might either be an 
issue with library versions (if you could show the output of
ldd `which bogofilter`) or some genuine bug, which might then depend on the 
command line.