SpamAssassin 3.1 & Pyzor/Razor/DCC

[Scott Silva]

    [ The following text is in the "ISO-8859-1" character set. ]
    [ Your display is set for the "US-ASCII" character set.  ]
    [ Some characters may be displayed incorrectly. ]

Peter Bates spake the following on 11/2/2005 6:33 AM:
> Hello all/Steve!
> 
> 
>>smf at F2S.COM 02/11/05 13:21:35 >>>
>>Curious - next best thing is going to be to pick a quiet time and run
> 
> a
> 
>>batch through MailScanner with 'Debug = yes' and 'Debug SpamAssassin
> 
> =
> 
>>yes' and see if anything shows up there.
> 
> 
> Well, I've done that, and I have a few odd results, which probably
> explain why bits aren't working...
> 
> Right at the top we have:
> 
> [16029] dbg: config: using "/usr/etc/mail/spamassassin" for site rules
> pre files
> [16029] dbg: config: using "/usr/share/spamassassin" for sys rules pre
> files
> [16029] dbg: config: using "/usr/share/spamassassin" for default rules
> dir
> 
> ... the second two are OK, but the first is an empty directory
> (init.pre is in /etc/mail/spamassassin)
> 
> [16029] dbg: config: using "/usr/etc/mail/spamassassin" for site rules
> dir
> [16029] dbg: config: using "/etc/MailScanner/spam.assassin.prefs.conf"
> for user
> prefs file
> 
> Same problem here... it has found the .conf file, but site rules is set
> to an empty dir.
> 
> [16029] info: config: failed to parse line, skipping:
> auto_whitelist_path
>  /var/spool/MailScanner/spamassassin/auto-whitelist
> [16029] info: config: failed to parse line, skipping:
> auto_whitelist_file_mode
>  0600
> [16029] info: config: failed to parse line, skipping: pyzor_path
> /usr/bin/pyzor
> [16029] info: config: failed to parse line, skipping: dcc_path
> /usr/local/bin/dc
> cproc
> [16029] info: config: failed to parse line, skipping: razor_timeout 10
> [16029] info: config: failed to parse line, skipping: pyzor_timeout 10
> 
> Whoops... I think that begins to explain my problem.
> 
> I can also see:
> 
> [16029] info: config: failed to parse line, skipping: urirhssub
> URIBL_JP_SURBL
> multi.surbl.org.        A   64
> 
> [16029] dbg: uri: running uri tests; score so far=0.96
> [16029] dbg: rules: no method found for eval test check_uridnsbl
> rules: failed to run URIBL_JP_SURBL test, skipping:
>         (Can't locate object method "check_uridnsbl" via package
> "Mail::SpamAssa
> ssin::PerMsgStatus" at
> /usr/lib/perl5/site_perl/5.8.5/Mail/SpamAssassin/PerMsgSt
> atus.pm line 2581.
> )
> 
> ... which also explain why I seem to be seeing no hits on URIs since
> upgrading.
> 
> Any thoughts?
> 
> There's clearly something that was okay in my MailScanner.conf (or
> spam.assassin.prefs.conf) for 3.0.4 but is now very *unhappy*.
> 
> In MailScanner.conf:
> 
> SpamAssassin Install Prefix =
> SpamAssassin Site Rules Dir = /etc/mail/spamassassin
> SpamAssassin Local Rules Dir =
> SpamAssassin Default Rules Dir =
> 
> Oh, just as it seemed an obvious thing to do,
> I symlinked /usr/etc/mail/spamassassin to /etc/mail/spamassassin,
> and guess what? It now works...
> 
> Is there some default setting here I'm missing in what is a fairly
> 'out of the box' install?
> 

I had this problem also, temporarily solved it with softlinks from
/usr/etc/mail/spamassassin to /etc/mail/spamassassin

I think Julian's install package needs to be tweaked to fix the default
SYSCONFDIR. Something must have changed in spamassassin with 3.1.0.
I would have a look at it, but Julian would probably have it fixed and
posted before I could get it e-mailed to him.


-- 

/-----------------------\           |~~\_____/~~\__  |
| MailScanner; The best |___________ \N1____====== )-+
| protection on the net!|                   ~~~|/~~  |
\-----------------------/                      ()

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the Wiki (http://wiki.mailscanner.info/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!