You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Nick Gilbert <ni...@nickgilbert.com> on 2005/05/10 00:26:23 UTC
Difference in scores between spamc and spamassassin
Hi,
Why do I get (VERY) different results on the same machine if I call
spamassassin in two different ways? Eg:
spamassassin -D < spam.txt
Gives:
X-Spam-Status: Yes, score=10.8 required=5.0
tests=AWL,HELO_DYNAMIC_COMCAST,INFO_TLD,RAZOR2_CF_RANGE_51_100,RAZOR2_CHECK,RCVD_IN_BL_SPAMCOP_NET,
RCVD_IN_SORBS_DUL,URIBL_OB_SURBL,URIBL_SBL autolearn=unavailable
version=3.0.3
but:
spamc < spam.txt
gives:
X-Spam-Status: No, score=4.8 required=5.0
tests=HELO_DYNAMIC_COMCAST,INFO_TLD
autolearn=no version=3.0.3
Surely they should both give exactly the same result?
Thanks,
Nick...
Re: Difference in scores between spamc and spamassassin [RESOLVED]
Posted by Matt Kettler <mk...@evi-inc.com>.
Nick Gilbert wrote:
>>>Surely they should both give exactly the same result?
>>>
>>>
>>Maybe.. Depends a LOT on the options you pass to spamd when you start it.
>>
>>
>
>Yes you're right - there's an error in the scripts on Bill Shupps
>shupp.org site which I used to make my box - it was calling spamd with
>-L for some reason. I expect someone turned it on for debugging and
>forgot to remove it again.
>
Well, some people do prefer to run it that way... You get some
substantial speed benefits by doing so, but your accuracy suffers.
Re: Difference in scores between spamc and spamassassin [RESOLVED]
Posted by Nick Gilbert <ni...@nickgilbert.com>.
>> Surely they should both give exactly the same result?
>
> Maybe.. Depends a LOT on the options you pass to spamd when you start it.
Yes you're right - there's an error in the scripts on Bill Shupps
shupp.org site which I used to make my box - it was calling spamd with
-L for some reason. I expect someone turned it on for debugging and
forgot to remove it again.
Thanks!
Nick...
Re: Difference in scores between spamc and spamassassin
Posted by Matt Kettler <mk...@comcast.net>.
At 06:26 PM 5/9/2005, Nick Gilbert wrote:
>X-Spam-Status: Yes, score=10.8 required=5.0
>tests=AWL,HELO_DYNAMIC_COMCAST,INFO_TLD,RAZOR2_CF_RANGE_51_100,RAZOR2_CHECK,RCVD_IN_BL_SPAMCOP_NET,
> RCVD_IN_SORBS_DUL,URIBL_OB_SURBL,URIBL_SBL autolearn=unavailable
> version=3.0.3
>
>but:
>
>spamc < spam.txt
>
>gives:
>
>X-Spam-Status: No, score=4.8 required=5.0 tests=HELO_DYNAMIC_COMCAST,INFO_TLD
> autolearn=no version=3.0.3
>
>Surely they should both give exactly the same result?
Maybe.. Depends a LOT on the options you pass to spamd when you start it.
In particular, it looks like spamassassin used network checks, but
spamc/spamd did not.
Any chance you started spamd with the -L parameter? (which forcibly
disables all network checks)
Also, what user did you do this as? Root? If so, spamd would have setuid'ed
itself to nobody before scaning the mail, and would have used a different
AWL database.
Re: Difference in scores between spamc and spamassassin
Posted by Craig McLean <cr...@craig.dnsalias.com>.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Nick Gilbert wrote:
| Hi,
|
| Why do I get (VERY) different results on the same machine if I call
| spamassassin in two different ways? Eg:
|
| spamassassin -D < spam.txt
|
| Gives:
|
| X-Spam-Status: Yes, score=10.8 required=5.0
|
tests=AWL,HELO_DYNAMIC_COMCAST,INFO_TLD,RAZOR2_CF_RANGE_51_100,RAZOR2_CHECK,RCVD_IN_BL_SPAMCOP_NET,
|
| RCVD_IN_SORBS_DUL,URIBL_OB_SURBL,URIBL_SBL autolearn=unavailable
| version=3.0.3
|
| but:
|
| spamc < spam.txt
|
| gives:
|
| X-Spam-Status: No, score=4.8 required=5.0
| tests=HELO_DYNAMIC_COMCAST,INFO_TLD
| autolearn=no version=3.0.3
|
| Surely they should both give exactly the same result?
In an ideal world ;-)
It looks like the checks not hitting when you use spamc are all network
checks (razor and DNS checks).
Perhaps the user running spamd has options disabling razor and other
network checks in its user_prefs, or perhaps spamd has not been
restarted since a change to include razor and DNS checks was made in
local.cf?
If none of the above, the output from 'spamassassin -D --lint' run as
the user controlling spamd might show something...
Kind Regards,
Craig.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
iD8DBQFCf+XWMDDagS2VwJ4RAh2gAJ9F2iRvEiswS04y0BUTcvNdjYfyDQCdHob/
XZr9SX3++KydfN6UJChjB9w=
=e9uB
-----END PGP SIGNATURE-----