Foxtrot048 wrote OFF-TOPIC:
> Win XP Sp 3, Thunderbird 17.0.4, AVG antivirus
>
> Occasionally I will get a spam email message with my own email address in
> the "From" column. How is this possible?
This is an anti-VIRUS newsgroup. That's nothing to do with spam which
is your issue. In the future, post spam queries in a anti-spam
newsgroup.
Anyone, even you, can put anything they want in the From header. Same
for the To/Cc headers. They are NOT used to specify delivery. When
your e-mail client connects to the SMTP server, it aggregates a list
of all recipients from the To, Cc, and Bcc headers and issues a
RCPT-TO command to the SMTP server for each recipient. Some clients,
like mass mailers, use a list of recipients from a file rather than
what the user specified in To, Cc, and Bcc *fields* shown in the GUI.
Similarly, the From header has nothing to do with where the e-mail
gets delivered. The To, Cc, From, and Reply-To fields are part of the
body of your post transferred by a following DATA command. Nothing
sent with the DATA command is used to specify delivery. It's just
data in the message.
RCPT TO: That command is how the client tells the server to where the
e-mail(s) get delivered. Might match on what you put in the To/Cc/Bcc
headers, might not.
DATA: That command is how the data of the message gets sent. That
includes whatever To, CC, From, Reply-To, Date, or other "headers" in
the body of the message that your client chose to include. This is
data, not destination information.
> I normally just delete these emails without reading them.
Why do you even see them? Tis easy to define a rule or filter that
looks for e-mails sent from yourself. Use a rule to delete, trash, or
junk any e-mails where your e-mail address is in the From header.
Anyone can put anything they want in the From header. It is extremely
rare that an SMTP server checks the *data* of a message to check the
From and Reply-To headers specify the account through which the e-mail
was sent. Even you can change your From header to whatever you want
to claim is you in your e-mail client. A mass mailer can easily put
the same value in the To and From header which is in the *data* of the
message (sent in the DATA command) and then specify whatever they want
in the RCPT-TO commands (one for each recipient) issued by the client
to the server.
> Question 2: The other day I got 6 identical emails with different
> senders. Obviously this was another scam and I just deleted those.
> How can I block such emails? Can AVG detect problems with emails
> such as these?
Define rules. You can define them up at the server (i.e., in your
account). You can define them in your local e-mail client. Filtering
out e-mails sent from yourself to yourself is easy. Filtering out a
spammer by their e-mail address is futile since they change it to
whatever they want each time they spew their crap. You'll never get
rid of spam by blacklisting their comment or e-mail address values in
the From header. Same sender can use any name and address they want.
You could build a blacklist of 3000 entries and that same spammer will
still get into your Inbox.
Ding dong. Butler answer's the door. Tell's you there is a visitor.
You go to the door.
Them: Hi, I'm Jack and am selling ...
You: No thanks. You close the door. Add Jack to your blacklist.
Ding dong. You go to the door.
Them: Hi, I'm Mark and ...
You: Hey, your Jack now claiming to be Mark. Close the door. Add
Mark to your blacklist.
Ding dong.
Them: I'm Susie and ...
You: Your Mark aka Jack. Shut door. Add Susie to blacklist.
And the looping continues while you build a worthless blacklist.
They can claim to be whoever they want to be. Their name changes
every time. You build a list that will never match on their next nym.
Blacklisting by name (address) is stupid and employed only by ignorant
users. It will never get rid of spam because they have more nyms to
use than you have space to add to a blacklist. There's only one
unwanted sender for which blacklisting will work: senders that always
use the same nym. Spammers don't do that; else, spam filtering would
be immensely easy by compiling public blacklists on spammer nyms and
spam would disappear overnight.
Filtering by sender's name or e-mail address won't work to get rid of
spam. Filter by the source of the e-mail. There are anti-spam
programs that incorporate DNSBLs (DNS blacklists) of known sources of
spam, like Spamhaus, Spamcop, and other publicly accessible
blacklists. The anti-spam program looks at the headers in received
e-mails to see where it came from. If it is from a spam source then
that e-mail gets tagged, flagged, held, trashed, or whatever action is
performed by the anti-spam program. You do have spam filtering
enabled up in your account, right? Another method of detecting spam
is using Bayes which statistically guesses at whether an e-mail is
spam or ham. Alas, it takes awhile for a Bayesian filter to learn
your e-mails during which you have to train it by identifying the
spams. Microsoft pushes an update their Bayesian spam filter in
Outlook or Windows Live Mail (if you leave junk filtering enabled) in
their monthly update. I don't know what Mozilla does to update their
Bayesian filter in Thunderbird. I don't use Bayes because it
generates too many false alerts because I don't get enough e-mail
traffic to have a large enough sampling in the database for it to make
good guesses as to what is spam or ham. Bayes is a guessing scheme
based on content and past flagging. Microsoft or Mozilla pushing
updates to their Bayes filters means you get a database not based on
your personal e-mail traffic but on theirs.
You never mentioned who is your e-mail provider. With some of them,
you can turn your account into an exclusive one where only known good
senders can get their e-mails delivered to you. That is, it is an
opt-in account. In Hotmail, you can set your account to exclusive
mode and only senders in your contacts or Safe Senders list will get
their e-mails into your Inbox. For other providers, you might be able
to define a "Safe Sender" rule that has a list of good senders with
their e-mails moved or left in the Inbox followed by another rule that
junks or trashes everything else. That does mean you have to
occasionally review your junk or trash folders to see if you missed
adding someone you want to you safe senders list or rule, especially
if you are expecting e-mails from them. For example, when I signup at
forums, I add them to my safe senders list but have to wait until they
send an e-mail so I know what to add. E-mails from anyone other than
your good senders gets filtered out of the Inbox.
Of course, you divulging your e-mail address everywhere just because
someone said they wanted it pretty much means you chose to have that
account targeted by spam. If you don't protect your e-mail address
then expect it to get abused and you won't be disappointed. Doesn't
matter if it is some unknown web site that wants your e-mail address
or a well-known big company asking for it. For example, if you
register at
gamespot.com, a long-time and well-known gaming review
site, you WILL get spammed in about 3 days. Don't give out your true
e-mail address just because someone asks for it. Give them an alias.
There are forwarding services that pretend they give you an alias:
senders send to that e-mail address and the service forwards the
e-mails to your true account. Alas, if you reply to those aliased
e-mails, it's likely that you divulge your true address in your reply.
Use a real aliasing service that masks the return address (what's in
the From and Reply-To headers) so your replies go back through the
aliasing service to wipe out your SMTP's headers, any headers you put
in the *data* of your message (From, Reply-To, etc), along with
scanning the body of your e-mail to replace any e-mail strings there
with your true e-mail address, like you append a signature with your
contact info. Some good alias providers are Sneakmail (paid) and
Spamgourmet (free).
You could also create disposable accounts, like at Hotmail, Gmail, or
Yahoo. If you poll them in your e-mail client, your replies will go
back through those accounts. If you don't want to poll them all the
time in your e-mail client, and don't expect to keep them for more
than 9 months (or whatever is their idle expiration period), configure
them to forward any incoming e-mails there to your true e-mail
address. Just remember that any replies will probably be through your
true account and divulge it in those replies, or remember to issue
your replies through those disposable accounts by using their webmail
interface. You can even use Gmail as a collector account since it
will poll up to 5 POP3 accounts, and just poll the Gmail account. If
you have more than 5 disposable addresses, use the 5th entry in a
Gmail account to poll another Gmail account where you get another 5
accounts to poll from there. The downside is Gmail will slack off to
1-hour polling intervals if it finds no new e-mails in the accounts
that it polls.
There are lots of ways to hide your true e-mail address. Only divulge
it to known and good senders that you trust, not to every bozo that
asks for it. You really trust someone you meet at a party that you've
never seen before with your true e-mail address? Nah, give them a
Spamgourmet alias since you can create those on-the-fly (no login
required, no software, nothing). Then decide when you start getting
their e-mails if you want to reply to them, give your true address, or
just kill or disable the alias so you never hear from them again. If
you do use aliases, make sure you give a UNIQUE alias to ever
potential sender. That way, if the alias gets abused, you will know
exactly to whom you gave that alias and now the culprit to avoid. If
you dole out the same alias to multiple senders (which is what happens
when you dole out the same disposable account to multiple senders),
you won't know who betrayed you when you start getting spam.
For more help on spam avoidance, visit the spam newsgroups. This
isn't one of them.
alt.spam
alt.stop.spamming
(look for "spam" in the newsgroup name)