You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by "Sven Juergensen (KielNET)" <s....@kielnet.de> on 2008/01/03 08:01:06 UTC

Spamassassin fallback to process owner.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi list,

i was posting this subject a while
ago and recently had some time to
look into it some more.

Apparently, whenever $SOMETHING isn't
part of the envelope and/or body,
spamassassin falls back to the user
which is owning the process. If that
user isn't listed in the SQL-Backend it
queries against, spamassassin just uses
the global setting which, in my case,
does nothing at all.

Does anyone of you reckon what causes
spamassassin to score on the process
owner? As stated, I *believe* it has
something to do with something missing,
but I have yet to recreate this.

Thanks for any clue and best regards,

sven03

Mit freundlichen Gruessen

i. A. Sven Juergensen

- --
Fachbereich
Informationstechnologie

KielNET GmbH
Gesellschaft fuer Kommunikation
Preusserstr. 1-9, 24105 Kiel

Telefon : 0431 / 2219-053
Telefax : 0431 / 2219-005
E-Mail  : s.juergensen@kielnet.de
Internet: http://www.kielnet.de

AS# 25295
Key fingerprint:
65B6 90FC 010A 39CE DCA5  336D 9C45 3B7A B02D E132

"221 2.7.0 Error: I can break rules, too. Goodbye."

Geschaeftsfuehrer Eberhard Schmidt
HRB 4499 (Amtsgericht Kiel)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)

iD8DBQFHfIgynEU7erAt4TIRAvoVAKDhWjFzwUk0VFDgul98XSXaI/CxbACfXvp0
QC9kOG1Pn4j9g6FpwzOf7/c=
=dkwp
-----END PGP SIGNATURE-----

Re: Spamassassin fallback to process owner.

Posted by Matt Kettler <mk...@verizon.net>.
Sven Juergensen (KielNET) wrote:
> Hello Matt,
>
> turns out that the script supplied by the
> designers of the whole mess is, well,
> suboptimal. I kicked it entirely and used
> a single command line to run a per-recipient
> checking.
>
> Just for info: whenever spamc can't find
> the $recipient or is lacking a '-u', it
> falls back to the process owner. If there
> are no userprefs defined for that recipient,
> the $GLOBAL settings are applied.
>   
That's mostly right, except spamassassin *NEVER* tries to find 
$recipient. That's never been a feature of SA, and likely never will.

If both spamd and spamc ar lacking a -u, it uses the process owner 
calling spamc. Period.


Re: Spamassassin fallback to process owner.

Posted by "Sven Juergensen (KielNET)" <s....@kielnet.de>.
Hello Matt,

turns out that the script supplied by the
designers of the whole mess is, well,
suboptimal. I kicked it entirely and used
a single command line to run a per-recipient
checking.

Just for info: whenever spamc can't find
the $recipient or is lacking a '-u', it
falls back to the process owner. If there
are no userprefs defined for that recipient,
the $GLOBAL settings are applied.

Thanks for the input!

Best regards,

sven03

Mit freundlichen Gruessen

i. A. Sven Juergensen

-- 
Fachbereich
Informationstechnologie

KielNET GmbH
Gesellschaft fuer Kommunikation
Preusserstr. 1-9, 24105 Kiel

Telefon : 0431 / 2219-053
Telefax : 0431 / 2219-005
E-Mail  : s.juergensen@kielnet.de
Internet: http://www.kielnet.de

AS# 25295
Key fingerprint:
65B6 90FC 010A 39CE DCA5  336D 9C45 3B7A B02D E132

"221 2.7.0 Error: I can break rules, too. Goodbye."

Geschaeftsfuehrer Eberhard Schmidt
HRB 4499 (Amtsgericht Kiel)


Matt Kettler wrote:
> Sven Juergensen (KielNET) wrote:
>> Hi list,
>>
>> i was posting this subject a while
>> ago and recently had some time to
>> look into it some more.
>>
>> Apparently, whenever $SOMETHING isn't
>> part of the envelope and/or body,
>> spamassassin falls back to the user
>> which is owning the process.
> Spamassassin *NEVER* examines the envelope or body to determine what
> user to run as. *NEVER*.
> 
> It *ALWAYS* runs as the user that invoked it, or whatever user it was
> explicitly told to run as (ie: via spamc's -u parameter).
> 
> It's possible that whatever tool you're using invokes SA using some part
> of said envelope or body to pass to -u, but but you didn't specify
> what's calling SA, so...
>> If that
>> user isn't listed in the SQL-Backend it
>> queries against, spamassassin just uses
>> the global setting which, in my case,
>> does nothing at all.
> nothing at all? That seems rather odd. Even with no user preferences at
> all, SA should run with fairly "normal" defaults. (ie: score threshold
> of 5.0, etc)
>> Does anyone of you reckon what causes
>> spamassassin to score on the process
>> owner? 
> As above, nothing. SA does what it's told, and if it's not told, it uses
> the calling userid.
>> As stated, I *believe* it has
>> something to do with something missing,
>> but I have yet to recreate this.
> Can you tell us more about your setup? Are you calling spamc from
> procmail? using a milter? ec.
> 

Re: Spamassassin fallback to process owner.

Posted by Matt Kettler <mk...@verizon.net>.
Sven Juergensen (KielNET) wrote:
> Hi list,
>
> i was posting this subject a while
> ago and recently had some time to
> look into it some more.
>
> Apparently, whenever $SOMETHING isn't
> part of the envelope and/or body,
> spamassassin falls back to the user
> which is owning the process.
Spamassassin *NEVER* examines the envelope or body to determine what
user to run as. *NEVER*.

It *ALWAYS* runs as the user that invoked it, or whatever user it was
explicitly told to run as (ie: via spamc's -u parameter).

It's possible that whatever tool you're using invokes SA using some part
of said envelope or body to pass to -u, but but you didn't specify
what's calling SA, so...
> If that
> user isn't listed in the SQL-Backend it
> queries against, spamassassin just uses
> the global setting which, in my case,
> does nothing at all.
nothing at all? That seems rather odd. Even with no user preferences at
all, SA should run with fairly "normal" defaults. (ie: score threshold
of 5.0, etc)
>
> Does anyone of you reckon what causes
> spamassassin to score on the process
> owner? 
As above, nothing. SA does what it's told, and if it's not told, it uses
the calling userid.
> As stated, I *believe* it has
> something to do with something missing,
> but I have yet to recreate this.
Can you tell us more about your setup? Are you calling spamc from
procmail? using a milter? ec.