|
List Info
Thread: Re: Confirming SPAM/HAM
|
|
| Re: Confirming SPAM/HAM |

|
2007-01-30 10:11:12 |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Jan 30, 2007, at 5:14 AM, Torsten Stauder wrote:
> Ok I did. Here's the output so far:
> [30-Jan-2007 12:11:52] maia: set_item_confirmations DB
Error: already
> exists UPDATE maia_mail_recipients SET type = ?,
token=CONCAT
> ('expired',
> recipient_id, mail_id) WHERE recipient_id = ? AND
mail_id IN
>
(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?
,?,?,?,?,?
> ,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)
> 2
>
69159,66923,72767,66994,67282,71707,69343,68857,68869,68883,
72616,7263
>
5,72652,72474,72482,72488,72502,73215,72411,72418,72427,7234
5,72350,72
>
358,72368,69162,71266,71268,71271,71272,71274,71256,71262,71
264,65554,
>
65551,71241,71246,71249,69974,71225,71229,71232,72993,70759,
69026,7125
> 3,67422,67068,68328
>
> But there was another strange thing: today I was able
to delete the
> first batch (50 Spams). Then the old behaviour was
shown :-(
I don't see any duplicate ID's, unless it's with another set
of values.
Besides the expire-quarantine-cache.pl, have you been
running the
process-quarantine.pl script?
One corner case that I need to address, would be if say, you
had
another user with an id of say 26, and they had a mail with
an id of
9159. that routine changes the token to make it expired,
and it's
supposed to be unique, but it may not be.
expired + id + mail_id
"expired" + "2" + "69159" and
"expired" + "26"+ "9159"
would both
be "expired269159" and fail the uniqueness test.
But the expired entries should not remain after a
process-quarantine
run, so the conflict should clear itself. Plus, for this
conflict to
happen, mail id's have to be an order of magnitude in
difference as
do the id's in question.... It's probably not likely to be
triggered often. (unless process-quarantine is not run)
David Morton
Maia Mailguard http://www.maiamailguard
.com
mortonda dgrmm.net
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)
iD8DBQFFv24jUy30ODPkzl0RAo6AAKCHjdBZbijb+2VLqSMU2FsNG6nyAACg
1o0F
CbKIREFAypJSGZrn7XfRcZk=
=SGId
-----END PGP SIGNATURE-----
_______________________________________________
Maia-users mailing list
Maia-users renaissoft.com
http://www.renaissoft.com/mailman/listinfo/maia-users
|
|
| Re: Confirming SPAM/HAM |

|
2007-01-30 12:39:03 |
Well... process-quarantine.pl finally solved my problem. Via
cron I ran
it every hour but there was a problem with the learning
procedure. So
finally this was my mistake - sorry!
I'll keep this in mind and have a look on it!
Thanks for your help!!
David Morton wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> On Jan 30, 2007, at 5:14 AM, Torsten Stauder wrote:
>
>> Ok I did. Here's the output so far:
>> [30-Jan-2007 12:11:52] maia: set_item_confirmations
DB Error: already
>> exists UPDATE maia_mail_recipients SET type = ?,
token=CONCAT('expired',
>> recipient_id, mail_id) WHERE recipient_id = ? AND
mail_id IN
>>
(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?
,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)
>>
>> 2
>>
69159,66923,72767,66994,67282,71707,69343,68857,68869,68883,
72616,72635,72652,72474,72482,72488,72502,73215,72411,72418,
72427,72345,72350,72358,72368,69162,71266,71268,71271,71272,
71274,71256,71262,71264,65554,65551,71241,71246,71249,69974,
71225,71229,71232,72993,70759,69026,71253,67422,67068,68328
>>
>>
>> But there was another strange thing: today I was
able to delete the
>> first batch (50 Spams). Then the old behaviour was
shown :-(
>
>
> I don't see any duplicate ID's, unless it's with
another set of values.
>
> Besides the expire-quarantine-cache.pl, have you been
running the
> process-quarantine.pl script?
>
> One corner case that I need to address, would be if
say, you had another
> user with an id of say 26, and they had a mail with an
id of 9159.
> that routine changes the token to make it expired, and
it's supposed to
> be unique, but it may not be.
>
> expired + id + mail_id
>
> "expired" + "2" +
"69159" and "expired" + "26"+
"9159" would both be
> "expired269159" and fail the uniqueness
test.
>
> But the expired entries should not remain after a
process-quarantine
> run, so the conflict should clear itself. Plus, for
this conflict to
> happen, mail id's have to be an order of magnitude in
difference as do
> the id's in question.... It's probably not likely to
be triggered
> often. (unless process-quarantine is not run)
>
>
> David Morton
> Maia Mailguard http://www.maiamailguard
.com
> mortonda dgrmm.net
>
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.5 (Darwin)
>
>
iD8DBQFFv24jUy30ODPkzl0RAo6AAKCHjdBZbijb+2VLqSMU2FsNG6nyAACg
1o0F
> CbKIREFAypJSGZrn7XfRcZk=
> =SGId
> -----END PGP SIGNATURE-----
_______________________________________________
Maia-users mailing list
Maia-users renaissoft.com
http://www.renaissoft.com/mailman/listinfo/maia-users
|
|
[1-2]
|
|