- WEB HOSTING
- RESELLER HOSTING
- Cloud VPS HOSTING
- Dedicated Servers
- About Us
This is a discussion on Spam in the Dedicated and Cloud Servers forum
Originally Posted by masood What does this sa-learn.raw do? Er... This is like a joke, right? ...
Question to wife: Are you getting less spam now? I've been playing around with the spam program...
Answer: Oh, yeah!
Redundant Question: So, it's better now?
Answer: Yes! I love it!
My wife doesn't love anything...
BTW, one of the rules in my user_prefs failed a lint, so I redid it.
Plus, I'm gonna try doing manual syncs for a while. If this works out, I'll do it through a daily cron, e.g.
Anyway, here's the latest...Code:/usr/bin/sa-learn --sync
Code:############################################################# # SpamAssassin user preferences file - VinDSL (Lenon.com) # ############################################################# # How many hits before a message is considered spam required_hits 2.5 # Text to prepend to subject if spam is suspected rewrite_header Subject *** SPAM (_HITS_ points, _REQD_ required) *** # Show message was processed by SpamAssassin at Lenon.com add_header all Checker-Version SpamAssassin _VERSION_ (_SUBVERSION_) on VinDSL's Lenon.com # Add X-Spam-Report to header add_header all Report _REPORT_ # Encapsulate spam in an attachment # 0=no, 1=yes, 2=safe report_safe 2 # Enable the Bayes system use_bayes 1 # Enable Bayes auto-learning bayes_auto_learn 1 # Disable auto_expire bayes_auto_expire 0 # Enable learning to a journal bayes_learn_to_journal 1 # Enable unlimited journal size bayes_journal_max_size 0 # Enable RBL tests & increase timeout skip_rbl_checks 0 rbl_timeout 30 # Reweight some existing rules score DATE_IN_FUTURE_03_06 2.5 score DATE_IN_FUTURE_06_12 2.5 score DATE_IN_FUTURE_12_24 2.5 score DATE_IN_FUTURE_24_48 2.5 score DATE_IN_FUTURE_48_96 2.5 score DATE_IN_FUTURE_96_XX 2.5 score MIME_BASE64_TEXT 2.5 score MISSING_DATE 2.5 score RCVD_IN_BL_SPAMCOP_NET 2.5 score RCVD_IN_NJABL_DUL 2.5 score RCVD_IN_SORBS_DUL 2.5 score SUBJECT_SEXUAL 2.5 score TO_EMPTY 2.5 # Whitelist and blacklist options whitelist_from *.lenon.com whitelist_from *@asu.edu whitelist_from *@mobile.mycingular.com whitelist_from *@jaguarpc.com blacklist_from email@example.com # Networks or hosts that are 'trusted' trusted_networks 18.104.22.168/18 # For English only use the following ok_languages en ok_locales en
Here's another cutie for you...
I've been playing around with backing up my SA dbs, but the results have been spotty. Many times the backup fails because the db files won't lock. However, I've noticed if I sync and prune just before doing the backup, it works every time.
So, if you wanna give it a try:
I'll probably set this up as a cron, if it continues to work manually...Code:/usr/bin/sa-learn -D --sync /usr/bin/sa-learn -D --force-expire /usr/bin/sa-learn -D --backup > backup_`date +%m-%d-%y`.txt
Last edited by Vin DSL; 12-06-2006 at 12:39 PM.
This will save you some space...
Code:/usr/bin/sa-learn -D --sync /usr/bin/sa-learn -D --force-expire /usr/bin/sa-learn -D --backup | gzip > /home/example/.spamassassin/backup_`date +%m-%d-%y`.gz
Here's an example from today.
The score of this message was 2.4 before AWL.Code:Content analysis details: (5.7 points, 5.0 required) pts rule name description ---- ---------------------- -------------------------------------------------- 1.1 EXTRA_MPART_TYPE Header has extraneous Content-type:...type= entry -2.6 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0002] 0.0 HTML_MESSAGE BODY: HTML included in message 2.0 RCVD_IN_SORBS_DUL RBL: SORBS: sent directly from dynamic IP address [XX.XXX.XX.14 listed in dnsbl.sorbs.net] 1.9 RCVD_IN_NJABL_DUL RBL: NJABL: dialup sender did non-local SMTP [XX.XXX.XX.14 listed in combined.njabl.org] 3.3 AWL AWL: From: address is in the auto white-list The original message was not completely plain text, and may be unsafe to open with some email clients; in particular, it may contain a virus, or confirm that your address can receive spam. If you wish to view it, it may be safer to save it to a file and open it with an editor.
If AWL always "pushes towards 0" as SA claims, the score for AWL should have been negative.
Instead it was 3.3 and pushed the message over the limit.
How do I raise this as an issue? With whom, SA or Jag? Should I bother? Are they just gonna tell me to reload the AWL, it must be corrupt?
Ron has a corrupt/incompatible AWL... and he's perfect... so this has to be a mistake, right? Because Ron's AWL represents him... even though he hasn't got a clue how it works -- doesn't even know who to contact, 'cause nobody else has a clue either, except his arch-nemesis, Vin DSL, who gets it sort of right, sometimes. It doesn't matter 'cause Ron's sh!t is perfect, just like him, or it wouldn't be Rons' AWL, right!
Am I getting the drift, dogman?
Oh, I'm sorry! Is this permissible?
Ron, isn't a homo, AFAIK?
He likes dogs and women....
Okay let's make it 6600 posts, minus the ones I recently deleted...
Or, did I just edit them?
Life is so confusing, eh, Ronny?
Apologies to everyone offended! Shall we continue, or am I boring you?
If there's one thing I've learned about Spamassassin... it requires vigilance, just like any other proggie...
If you think this is all magic, I would suggest that you are no better than somebody that just bought their first Dell, my webMASTER brethren.
Think about what I just said...
I've been watching and trying the changes to the prefs file. I haven't tried your latest things from the command line as yet.
But so far I've noticed no decrease in the amount of spam over the last day or so. If anything, I'd say I'm getting a bit more.
However, email for me comes via two accounts, and I haven't made the changes on the second one as yet.
I appreciate you keeping us all in the loop on this.
Best Regards, Lloyd Borrett.
With my wife's account, I've setup a rule where spam is automagically downloaded to the 'Deleted Items' folder in her mail client, instead of her 'Inbox'. So, for all intensive purposes, she SEES less spam.Question to wife: Are you getting less spam now? I've been playing around with the spam program...
Answer: Oh, yeah!
Spamassassin is a system for DETECTING spam, .e.g. a mail filtering system. So, on face value, I would say it's a good thing that you're getting a bit more spam now. Allow me to explain...
Recently, SA has NOT been doing a good job at detecting spam. In my case, I think it's because the dbs were jacked and the AWL impotent.
Why did I think this? Hold on to yourself...
The last time my Bayes dbs were updated was 12-1-2003
And, my AWL had grown to 187 MB (it's 679 KB now), which is suspicious enough in itself, but the corker was when I started poring over undetected spam, and realized mail from '777 Online Casino', or whatever, was being white-listed in AWL...
That's when I decided to start from scratch. And, yes, SA is detecting spam like mad now!
As I told Ronny, above, I'm becoming aware that SA is just like any other proggie. There are a LOT of things YOU can do with it, but this is going to require investigation, customization, and so forth, and so on... just like PHP-Nuke, WordPress, et cetera.
To SA's credit, it seems to work fairly well, even if you don't do anything. However, having a corrupt/incompatible/outdated AWL and dbs is NOT helping the situation.
So, personally, I'm gonna keep going forward with this, until I get bored... and, I'll keep posting the fruits of my tinkering.
The pieces ARE starting to fall into place. For instance, I never did understand the importance of having a spam folder, e.g. saving your spam. However, after getting into it a bit more, I'm starting to realize SA uses these spam/ham folders for 'learning'. LoL! And, all these years, I've just been deleting spam, as soon as I get it...
Last edited by Vin DSL; 12-08-2006 at 01:41 PM.
It seems to me that using a program like Mail Washer is a lot simpler. You guys jump through hoops and in the end you still have to review the spam folder to see if you want to delete the email.
Worked 19 hours yesterday, and need to go back in 45 minutes, so I'll keep this short.
Here's the latest...
I was thinking about 'SA spam learning', in the wee hours of the morn, and decided to give it a whirl before going to bed. Worked great!
I added a folder called 'Junk' to one of my mail accounts, and have been dumping (detected and undetected) spams over there for a couple of days. Then, I ran the following command this morning:
As you've probably noticed, I've been running all these manual SA commands in debug mode, so I can watch/read the output. I haven't had a chance to pore over this output, but the '--spam' option is obviously doing 'its thing' (no warnings).Code:/usr/bin/sa-learn -D --spam /home/example/mail/lenon.com/john/.Junk/cur
When I get a chance, I'm gonna try 'SA ham learning' next. Ham is non-spam emails, in Spamassassin parlance, e.g. good mail.
Last edited by Vin DSL; 12-10-2006 at 12:45 PM.