-------- Original-Nachricht --------
> Datum: Mon, 24 Dec 2007 11:29:02 +0100
> Von: Tony Earnshaw <[EMAIL PROTECTED]>
> An: [email protected]
> Betreff: Re: [dspam-users] RPM?

> Steve skrev, on 24-12-2007 10:08:
> 
> [...]
> 
> > Tonni. You self are using patches from Gentoo in your RPM. I see that
> you use the BNR debug fix. How do you think this patch has emerged? It was
> one Gentoo user reporting it to the Gentoo bugzilla
> (http://bugs.gentoo.org/show_bug.cgi?id=199700) and it was me fixing it. Now 
> if you look at the
> whole conversation in the bug report, then you will see that we take at Gentoo
> our users very serious. Alin and me took the time to install PostgreSQL and
> configure a separate instance of DSPAM just to reproduce the users error
> and attempt to fix it. If you look at the report then you will see that the
> user helped us to spot the error. And this is exactly what I am asking from
> you. If you are serious about the error and about wanting that
> error/segmentation fault to be fixed, then I need some help form you in order 
> to
> reproduce the error. Just saying "it segfaults on me" is not going to help you
> or me, because I could easily say "it works for me".
> > If I can't reproduce the error, then I can not debug it and then I can
> not find it. Don't get me wrong. I am since months looking into DSPAM source
> code and digging around. The code is in in some parts very prone to
> errors. And I know that. I have started months ago to rewrite some parts of 
> the
> code and I am fixing many things. But I do that mostly for me. I do that
> because I want my DSPAM to be rock solid and I want to understand the internal
> of DSPAM better. I want this beast to be fast and solid. I don't want it
> to crash. For me crashing is a no-go.
> > So if you want me or others on the list to help you with that
> segmentation fault, then you need to provide us some more info. It is a give 
> and
> take. You give us more info and we try to fix it and give back the patch to
> eliminate that error. Easy. Isn't it?
> 
> The bnr patch is one of two patches I use; one  (a CGI patch from Jonz 
> aeons ago) y'all don't include. For the rest as I said, my stuff doesn't 
> crash and yours does. Etc, etc.
> 
Well.. I took now the time and compiled CVS with the SPEC file (had to add the 
autogen stuff because CVS needs it. The result can be seen in the attached 
dspam.log file.


I classified successful 50 spam mails on my CentOS 5.1 without the DSPAM binary 
crashing. I have not started the daemon on CentOS but used the one I already 
have running on Gentoo. If you want, then I could configure the daemon on 
CentOS and run a pure local test. I reused the existing daemon because you 
wrote that the daemon does not crash and only the DSPAM binary is crashing.

Well.. I can not (jet) reproduce your crash on my end. Would it be possible for 
you to post some more info about what storage driver you are using, what 
version of the storage software (MySQL Server version, PostgreSQL version, etc) 
and maybe post a anonymized version of dspam.conf?
 


> In the mean time I'm trying to cope with your latest spec, which is 
> croaking in various different places (it's unnecessarily  complicated 
> and does things it shouldn't); you'll have to give me time.
> 
It is complicated because I tried to make a very generic spec file. What things 
does the spec do, which it should not?


> As to Red Hat using /usr/share for all http stuff, see the RH 
> squirrelmail rpm (use the CentOS repo, since RH's strictly license): 
> http://mirror.centos.org/centos/5.0/updates/i386/RPMS/squirrelmail-1.4.8-4.0.1..el5.centos.1.noarch.rpm.
> 
Okay. Will look into that.


> Best,
> 
> --Tonni
> 
Kind Regards

Steve

> -- 
> Tony Earnshaw
> Email: tonni at hetnet dot nl

-- 
GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS.
Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail

-- 
Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten 
Browser-Versionen downloaden: http://www.gmx.net/de/go/browser

Attachment: dspam.log.tgz
Description: Binary data

Reply via email to