At 21:49 23.10.2003, you wrote:
>Which version of Perl do you have? I had similar problems (endless loop with
>some HTML messages from one source) with Perl 5.0. Upgrade to 5.6 or 5.8 had
>solved the problem.
thanks for the info, however i am using perl 5.8.0 atm.
as i said i am not sure if this problem is sanitizer related.
the other possibility would be that the perl call to the sanitizer binary
under some conditions.
for those interested, this is what the call looks like atm:
my $pid = open3(*SAN_IN, *SAN_OUT, *SAN_ERR,
print SAN_IN $mail;
#read output mail
my @newmail = <SAN_OUT>;
my @errors = <SAN_ERR>;
howerver, the important point is that the sanitizer process is not
terminating, i am not sure if a blocked
stdout or stderr would prevent the sanitizer process from exiting, but i
guess that might well be possible.
any thoughts on this?
>On 23 Oct 2003 at 21:32, V. Weck wrote:
>Subject: [anomy-list]: problems with sanitizer blocking on
>Date sent: Thu, 23 Oct 2003 21:32:28 +0200
>From: "V. Weck" <firstname.lastname@example.org>
> > hi,
> > i recently had a couple of problems with blocking sanitizer processes,
> > meaning that they accepted a message on stdin (from another program) but
> > somehow didn't finish properly.
> > this led to exim and sanitizer processes being blocked and effectively
> > stoped the corresponding emails from being delivered.
> > unfortunatly those emails were lost so i am unable to check if they
> > had any similarities.
> > however most of the time everything works flawlessly (i'd say it only
> fails for
> > about 0,1% of the emails).
> > my setup might be a little different from most others as i am using a self
> > written anti-spam parsing script which is called by procmail and calls
> > in turn. sanitizer uses the antivir virus checker. so it looks like
> this: exim
> > -> procmail -> own perl script -> sanitizer -> antivir
> > the problematic emails do pass all stages and end in sanitizer, meaning the
> > virus scanner is never called (or at least does not create it's usual
> log entry
> > and isn't visible in the ps output).
> > this might well be caused by some other problems and i will continue to
> > look into
> > the matter, but i wanted to ask if someone might have encountered such
> > with sanitizer before.
> > thanks.
> > mfg
> > v. weck