Re: [vps-mail] Fwd: [SAtalk] [SA-Announce] SpamAssassin 2.60 rc3 released

[Bennett Lanford <ben@xxxxxxxxxxxxx>]

On Thu, 28 Aug 2003, Abigail Marshall wrote:

> Scott, I thought you might be interested in this - the new
> version of SA has a feature (UNIX-domain sockets) that might
> reduce memory load, if it is something that can be
> implemented on a VPS.

As long as we're on the subject of Unix sockets ... there is also a milter
version of Spamassassin (called spamass-milter). It is available in the
FreeBSD ports collection.

I have it running on one of my "real" FreeBSD servers. (I compiled it on
my VPS2, but haven't configured it yet ...) Milters can "talk"  to
sendmail either through Unix domain sockets (on the same machine) or
through regular IP sockets (running on the same or on a different machine
...)

Both the VPS1 and VPS2 versions of the sendmail executable have MILTER
support compiled in.

-Ben

>
> -Abigai
>
>
> From: Justin Mason <jm@xxxxxxxxxx>
> To: SpamAssassin-talk@xxxxxxxxxxxxxxxxxxxxx
> Date: Thursday, August 28, 2003, 5:18:07 PM
> Subject: [SAtalk] [SA-Announce] SpamAssassin 2.60 rc3 released
>
> ===8<==============Original message text===============
> Pick it up from:
>
>   http://SpamAssassin.org/released/Mail-SpamAssassin-2.60-rc3.tar.gz
>   http://SpamAssassin.org/released/Mail-SpamAssassin-2.60-rc3.zip
>
> md5sums:
>
>   141a73b07d46324865ea7ead49375043  Mail-SpamAssassin-2.60-rc3.tar.gz
>   94f196bfe9301233c848934908cc96ac  Mail-SpamAssassin-2.60-rc3.zip
>
> Changes since 2.5x:
>
> - spamd now supports UNIX-domain sockets for low-overhead scanning, thanks
>   to Steve Friedl for this.  Strongly recommended if you're running spamc
>   on the same host as the spamd server
>
> - DNSBLs overhauled; lots of new lists, including SORBS and SpamCop BL;
>   dropped Osirusoft due to their retirement, brought on by a DDOS
>
> - better support for detecting 'invisible text' in HTML messages
>
> - new Bayes tweaks -- tokenization of partial address and URI elements
>
> - RBL timeouts are now handled much more efficiently; if one RBL is taking
>   much longer than all the others, it will be cut short and ignored
>
> - now reduces amount of IP addresses queried, by inferring trust relationships
>   between Received headers; trust can be manually specified using
>   "trusted_networks"
>
> - generalised Received-header parser to more easily detect HELO string
>   forgery and faked relay lines
>
> - default report updated to list the hostname of the machine where
>   SpamAssassin is running; also asks installer for administrative address
>   or URL, to reduce user confusion
>
> - "add_header": more flexible header rewriting, with a simple template system
>
> - workaround for red hat kernel with NPTL reporting "application bug" in
>   spamd due to use of waitpid() when SIGCHLD is set to SIG_IGN
>
> - dccifd support added
>
> - Razor 1 support dropped
>
> - support for domain-based blacklists
>
> - new bayes backend, including new database format, ability to learn messages
>   to the journal, etc.  Improved expiration mechanism
>
> - lots of bugs fixed and new rules added. ;)
>
>
> Bayes FYI: 2.60 has a new Bayes backend and database format.  Your old
> database(s) will automatically be upgraded the first time 2.60 tries
> to write to the DB, and any journal, if it exists, will be wiped out
> without being synced.
>
> In addition, we have had to drop support for Bayes databases in formats
> other than DB_File, due to a large number of serious issues (including
> crash and concurrency bugs) with those formats.
>
> So, what you want to do is something like this:
>
> - stop running spamassassin/spamd (ie: you don't want it to be running
>   during the upgrade)
> - run "sa-learn --rebuild", this will sync your journal.
> - upgrade SA to 2.60
> - install DB_File module if necessary
> - if you were using another database module, run "sa-learn --import"
>   to migrate the data into new DB_File files
> - run "sa-learn --rebuild", this will cause the db format to be upgraded
> - start running spamassassin/spamd again
>
> Obviously the steps will be different depending on your environment,
> but you get the idea. :)
>
> Bayes FYI 2: Because of the new database format, "check_bayes_db" will
> no longer function properly.  The functionality from that script was
> added to sa-learn via the "--dump" parameter.  Please see the sa-learn
> man/pod documentation for more info.
>
>
> Changes since 2.60 rc2:
>
> - bug 2334: trusted network inference ignores 'local tests only'
>
> - bug 2370: removed support for non-DB_File database modules in Bayes
>
> - bug 2365: Osirusoft rules dropped
>
> - bug 2356: report_safe_copy_headers was broken
>
> - bug 2338: Razor static srand() worked around
>
> - bug 2360: Bonded Sender queries changed zone
>
> - bug 2347: add_header did not support multiple newlines
>
> - bug 2354: header options now more backward-compatible
>
> - bug 2359: "make install" logic for non-standard PREFIXes broken
>
> --j.
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Spamassassin-announce mailing list
> Spamassassin-announce@xxxxxxxxxxxxxxxxxxxxx
> https://lists.sourceforge.net/lists/listinfo/spamassassin-announce
>
> ======================================================================
> Technical questions regarding this list may be sent to
> <vps-mail-owner@xxxxxxxxxxxx>. You may request an automated help
> response by sending an email with the word 'help' (w/o quotes) in the
> BODY of the message (subject is ignored) to <vps-mail-request@xxxxxxxxxxxx>.
> ======================================================================
>
======================================================================
Technical questions regarding this list may be sent to
<vps-mail-owner@xxxxxxxxxxxx>. You may request an automated help
response by sending an email with the word 'help' (w/o quotes) in the
BODY of the message (subject is ignored) to <vps-mail-request@xxxxxxxxxxxx>.
======================================================================