On Fri, Sep 10, 2004 at 02:44:51PM -0400, Theo Van Dinter wrote:
> On Fri, Sep 10, 2004 at 11:38:33AM -0700, Michael Hall wrote:
> > I just thought I'd try enabling DCC again with 'rc4' and also (still)
> > get the same error message:
> >
> > Sep 10 11:25:57 ukiah dccproc[43025]: missing message body; fatal error
> >
> > Has anybody got DCC working with 3.x? Reset my old 2.63 up again and DCC
> > works fine.
>
> Still works fine for me:
>
> debug: executable for dccproc was found at /usr/local/bin/dccproc
> debug: DCC is available: /usr/local/bin/dccproc
> debug: entering helper-app run mode
> debug: setuid: helper proc 16677: ruid=501 euid=501
> debug: DCC: got response: X-DCC-sgs_public_dcc_server-Metrics:
> eclectic.kluge.net 1199; Body=2 Fuz1=5 Fuz2=many
> debug: leaving helper-app run mode
> debug: DCC: Listed! BODY: 2 of 999999 FUZ1: 5 of 999999 FUZ2: 999999 of 999999
>
> What does your -D output say? Are you running spamd, spamassassin,
> or something else? There aren't any useful details in your message.
Sorry about the delayed reply, I had a family emergency and just got back
into town this weekend and finally getting caught up with my mail :-(
Anyway, I was using 'rc4', perl 5.8.5, and amavisd-new-2.1.2, all under
FreeBSD. Feeding a message directly to spamassassin shows DCC working,
yet running from amavisd-new results in 'missing message body; fatal error'
for every message. After enabling and setting up dccifd it works fine
versus using dccproc.
Enabling debugging in amavisd-new and grepping the output I had:
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
DCC -> check failed: Died at /usr/local/lib/perl5/site_perl/5.8.5/Mail/SpamAssas
sin/Dns.pm line 802.
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
debug: DCC -> check timed out after 10 secs.
debug: DCC -> check timed out after 10 secs.
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
debug: DCC -> check timed out after 10 secs.
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
debug: DCC -> check timed out after 10 secs.
debug: DCC -> check timed out after 10 secs.
debug: DCCifd is not available: no r/w dccifd socket found.
debug: DCC is available: /usr/local/bin/dccproc
which seems to indicate its actually a timeout issue. Interesting enough
after reading the 3.0 announcement thread regarding 'pyzor' not working
(bugzilla ID 3649) this sounded similar. After applying Mark Martinec's
fix logic for the pyzor issue to dcc it now works (SA 3.0, dccproc,
and amavisd-new). What I did was just uncomment the old line and comment
out the new line(s) in Dns.pm:
sub dcc_lookup
...
my $pid = open(DCC, join(' ', $path, "-H", $opts, "< '$tmpf'", "2>&1", '|'))
|| die "$!\n";
#my $pid = Mail::SpamAssassin::Util::helper_app_pipe_open(*DCC,
# $tmpf, 1, $path, "-H", split(' ', $opts));
...
--
Mike Hall,
System Admin - Rock Island Communications <[EMAIL PROTECTED]>
System Admin - riverside.org, ssdd.org <[EMAIL PROTECTED]>
