Jump to filtering
instructions What is Spam? According to WikiPedia,
"Spamming is the abuse of electronic messaging systems to send
unsolicited bulk messages, which are universally undesired."
See the full entries for spam
and email
spam for more detailed information. Where Does it Come From? Often, spam messages you receive
come from a forged address. That means the email address or even
the name of the sender of a spam message, as it appears in your mail
client, is not necessarily the person who really sent it. Commonly,
people who send spam will try to make it look like someone else in
your organization is the sender of the spam, in order to trick you
into thinking the message is a legitimate email. Spam messages are
commonly trying to sell you something. Why am I getting spam? People and organizations that send
spam use numerous methods to find addresses to which they send their
spam. The first, and simplest, method for gathering addresses is to
scrape them off web sites and Usenet Newsgroups. Next, various
Windows viruses and malware started appearing that would try to
harvest anything and everything on infected computers that looked
like an address book or even a single email address, and send them
to a spammer for later use. Recently, spammers have resorted to
dictionary attacks on mail servers, hammering the servers with
requests to send to every user name conceivable in hopes that even a
few messages actually find real recipients. By far the most successful methods for address harvesting have been
from web site scraping and address-harvesting viruses on PCs, though
as collections of addresses grow, the spammers are having more and
more success with dictionary-style attacks when sending their
spam. What can I do about it? One possibility is to be proactive
and avoid giving out your email address to anyone on the web who you
suspect might turn around and sell your name to others or use it for
unsolicited mailings. In practice, this is not an easy thing to
determine, and there are numerous times when you do need to give out
your email address. Some people keep a second mail account somewhere, either through
their ISP or via one of the free email services. They use this
address for anything questionable. Another option is a temporary,
disposable address. www.spamgourmet.com is
one site that hosts these. When you do receive spam, we suggest that you do not reply,
not even to complain. If the return address is even valid, all
complaining will do is tell the spammer that they have sent their
junk email to a valid email address and that someone on the other
end took the time to read their email. If anything, this makes you
an even more valuable target. Department Spam Processing The CSE Department has
recently aquired and installed a Barracuda
Spam Firewall appliance which scans for viruses and spam
in all incoming email to the CSE Department for @cs.ucsd.edu and
@cse.ucsd.edu email addresses. The Barracuda is updated hourly with
new spam and virus definitions by Barracuda
Networks in order to try to provide more effective spam
and virus filtering for our email systems. What is the Barracuda? The Barracuda is a
specially-designed email server that acts as the incoming mail
server for the CSE Department and actively examines all incoming
email to see if it is Spam or contains a virus, worm or trojan
program. The Barracuda system is backed by a support service that
provides hourly updates to both the virus definitions and the spam
filtering technology, in order to keep up as best as possible with
the ever changing nature of Spam and virus email messages.
Additionally, users can elect to create an account for themselves on
the Barracuda, which will give access to additional features such as
Quarantine of spam messages and the ability to custom-tune the spam
filter based on declaring held messages as spam or not spam. How does the Barracuda work? By default, the CSE
Department's Barracuda system is configured to evaluate every
incoming mail message and score each message based on how
"spammy" it is. Messages end up falling into one of the
following categories:
- Not spam, and not a virus payload.
- Looks a little like spam, but might not be spam.
- Matches known spam profiles, or evaluates as
"spammy" enough that there is very little chance
that it is a legitimate email message.
- Message contains URLs that have been determined to be
illigitimate phishing sites or the message happens to match,
exactly, known spam or phishing messages, or the message
contains a virus or an attachment with a virus.
In cases 1 through 3, the messages are delivered to your mailbox. In
the final case (4), the message is never accepted by CSE Department
mail servers and is instead rejected, so you will never see it in
your mailbox. How do I make the Barracuda work for me? The typical way
to use the system is to set up a mail filter using a method of your
choice (see below for instructions using various mail clients.) We recommend you set up this filter to save aside any messages
marked as "likely spam" to a separate folder. You then periodically
go through this spam folder, quickly making sure no non-spam ("ham")
was put there by mistake. Then delete the spam. There are several ways to filter the spam so that it doesn't clutter
up your mailbox. You can set up rules to file messages into a spam
folder as they arrive to your Inbox using:
If your mail account is on the Zimbra server, we strongly recommend
that you create your mail filters within Zimbra. To create additional rules in mail clients to more aggressively
filter spam, see these
instructions. Additionally, Outlook users can install a
plugin from Barracuda to train the system on individual
messages by marking them as spam or not-spam via two new toolbar
buttons. The Barracuda offers additional capabilities and features, and
documentation on what those are and how to use them will be
forthcoming. The system is by no means perfect. It has the potential for both
marking ham as spam, and vice-versa. As a result, we strongly
recommend using the spam scores only to filter the potential spam to
a separate folder for later skimming, and not to just
delete the potential spam without any human review. Since the advent of filtering options in mail clients and software
packages that perform spam tagging, spammers have been working to
constantly re-craft the messages they send out. Because of this,
some spam may not be scored as highly as you think it should. Isn't there someone I can complain to? Generally, Internet
Service Providers are supposed to have and monitor an email address
called "abuse" for people to register complaints about
problem users. Sending a complaint to this address with the
complete raw source of the spam message in question might get a
response, if the information is clear enough and you've managed to
trace the spam to the correct ISP. Often, though, there is simply
too much to deal with, and you may receive no reply. If you receive email that is objectionable, abusive, or threatening
beyond the bounds of "normal" spam, please save the email
and forward it and all headers to csehelp@cs.ucsd.edu.
 |