AOL Now Publishing SPF Records 340
SPF Fan writes "It looks like SPF is starting to catch on with the bigger ISPs. AOL is now publishing SPF records which you can verify with 'dig aol.com txt'. Will Hotmail and Yahoo be far behind? Who else is publishing SPF records for their domains? Slashdot has covered SPF in the past a couple times."
I publish SPF records (Score:4, Informative)
I've only done the publishing side, I have not yet enabled my mail servers to use them.
Even though SPF may not be a complete or perfect solution, I see no harm in announcing to the world that if it purports to come from my domain than it also comes from my designated mail servers.
omg... (Score:2, Informative)
Make/break it (Score:1, Informative)
Anyone can develop standards, but still it's the ISPs that can make it or break it. Big ISPs can push some standard, and force the whole internet to use SPF or be cut off.
Re:boo (Score:5, Informative)
$ dig aol.com txt
; <<>> DiG 9.2.2 <<>> aol.com txt
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 49576
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 4, ADDITIONAL: 4
;; QUESTION SECTION:
;aol.com. IN TXT
;; ANSWER SECTION:
aol.com. 300 IN TXT "v=spf1 ip4:152.163.225.0/24 ip4:205.188.139.0/24 ip4:205.188.144.0/24 ip4:205.188.156.0/24 ip4:205.188.157.0/24 ip4:205.188.159.0/24 ip4:64.12.136.0/24 ip4:64.12.137.0/24 ip4:64.12.138.0/24 ptr:mx.aol.com -all"
;; AUTHORITY SECTION:
aol.com. 3071 IN NS dns-02.ns.aol.com.
aol.com. 3071 IN NS dns-06.ns.aol.com.
aol.com. 3071 IN NS dns-07.ns.aol.com.
aol.com. 3071 IN NS dns-01.ns.aol.com.
;; ADDITIONAL SECTION:
dns-02.ns.aol.com. 3273 IN A 205.188.157.232
dns-06.ns.aol.com. 1887 IN A 149.174.211.8
dns-07.ns.aol.com. 431 IN A 64.12.51.132
dns-01.ns.aol.com. 192 IN A 152.163.159.232
;; Query time: 110 msec
;; WHEN: Fri Jan 9 09:06:32 2004
;; MSG SIZE rcvd: 405
Re:Some of us have reasons for spoofing our addres (Score:5, Informative)
SPF was invented especially to cater for your situation. The quick way out would have been to use MX records as the only validation, but this was not done.
Re:How does this reduce spam in any shape or form? (Score:2, Informative)
1) since it effectively kills sender forgeries, it's a LOT easier to maintain white/blacklists
2) a domain needs to be purchased, and the registration takes time; this increases the cost of spam and hopefully might also make spammers more traceable (credit card transactions for registration)
I am totally convinced this will make the spam problem manageable. I'll probably add my own SPF this weekend.
Re:Some of us have reasons for spoofing our addres (Score:2, Informative)
You wouldn't. But that is part of the problem as legitimate uses can't be differentiated from SPAM when taking this approach.
Its one of those great "lose liberty in the name of enforcement" style things.
Or of course you could just set up SMTP on that remote server of yours.
NewsFlash (Score:1, Informative)
Why this is a big deal (Score:5, Informative)
SPF [pobox.com] is a proposed standard for a domain owner to tell mailers where mail From: that domain may originate. The domain owner publishes a DNS TXT record for their domain with (at the simplest) list of IP addresses. Participating mail transfer agents can then look this record up and make a policy decision on whether the mail is likely to be legitimate. The presence of an SPF record on a domain at present means that while you still can't be sure when you're handling spam, you can be sure when you have a piece of non-spam because the SPF record tells you so.
SPF is not a wholly original idea (e.g. up "designated mailer protocol"), and certainly not the simplest implementation but the important factor is that its proponent, Meng Wong, is an excellent lobbyer and spokesperson, as well as someone who as the nous to put forward a useful protocol (he founded pobox.com). It's currently at the point where lots of implementation are being written, with the canonical version being Meng's Perl modules. Currently I'm helping to finish the C implementation which will shortly be integrated into qmail [cr.yp.to] and exim [exim.org].
The tipping point (I hope) will be when a domain not publishing an SPF record or publishing a globaly permissive one will be considered "obviously" untrustworthy. Combining SPF authorisation with a more traditional "From: domain blacklist" will give spammers a very very hard time indeed forging mail. But AOL publishing a record (we hope) shows the way the wind is blowing: the rest of the world does seem to have to change their mail server configuration to keep mail flowing to AOL.
So go on, it's dead easy, publish a record for your domain now. Tell people where your mail comes from. Look, there's even a wizard to help you [pobox.com].
Re:boo (Score:3, Informative)
This does reduce spam (Score:5, Informative)
If Red Hat adds SPF verification to their default spamassassin configuration files, a lot of companies will start to add SPF records to their DNS.
If I send an e-mail to a RoadRunner mailbox, it is rejected. Why? Because my mailserver is a Linux box on my ADSL internet connection, and RoadRunner blocks all e-mails from residential IP ranges. With SPF, such filtering can be made much more careful, making it possible for me to send e-mails to RoadRunner customers again.
Re:How does this reduce spam in any shape or form? (Score:5, Informative)
I bet your parents are proud!
Re:How about dynamic IPs? (Score:3, Informative)
In the UK we have plenty of choice for broadband ISPs who offer fixed IPs at no extra cost (which is why I'm moving away from BT Openworld who charge an extra 10 a month for the privilege)
Comment removed (Score:4, Informative)
Spamassassin will support it in 2.70 (Score:4, Informative)
Anyway, it seems SpamAssassin will be adding support for SPF in 2.70, at least according to bug 2143 [spamassassin.org]. That's cool!
Re:boo (Score:4, Informative)
As to the second point that is already easily dealt with by most intelligent MTA's, heck my ISP's email servers already flag any message which has a different sending IP and host identifier, and they have informed us that they plan to dump the connection on this condition "real soon now". SPF just makes this easier since it can be used to eliminate false positives from semi-clued admins.
Re:Suggestion for submitter (Score:3, Informative)
Dynamic IP addresses (Score:3, Informative)
Re:How does this reduce spam in any shape or form? (Score:1, Informative)
Re:Why this is a big deal (Score:3, Informative)
It does seem to work (Score:4, Informative)
This is what I got:
Jan 8 19:34:01 scrat sendmail[16839]: i08IY0ON016839: Milter: from=<larhondabeirne@aol.com>, reject=550 5.7.1 Command rejected
Jan 9 05:34:47 scrat sendmail[16305]: i094YlON016305: Milter: from=<krbsnag2gs@aol.com>, reject=550 5.7.1 Command rejected
Jan 9 08:59:45 scrat sendmail[25027]: i097xiON025027: Milter: from=<clairacree@aol.com>, reject=550 5.7.1 Command rejected
Re:Would someone explain this to a simpleton? (Score:5, Informative)
One day, you start getting a lot of bounced spam. Some spammer, for some reason, has decided that he would forge his latest batch of spam from @sharpfang.com email addresses. What a dick!
So, you set up SPF records for your domain. The SPF records are basically a way of telling other mail servers, "I only send mail from my cable modem connection, which will always have an IP of 24.95.x.x. If you get mail claiming to be from sharpfang.com, but it didn't come from an IP address inside 24.95.0.0/24, it's bogus!"
Now, enlightened mail server admins can reject any email with an @sharpfang.com return address but an origin IP of somewhere outside of 24.95.0.0/24. Of course, if your IP address or range changes (e.g. you're traveling, you switch ISPs) you simply update your SPF records in DNS.
SPF has dual benefits: it can reduce the load you get from joe-jobs (assuming some of the recipients' mail servers honor SPF), and it helps everyone else identify spam.
Re:Dynamic IP addresses (Score:4, Informative)
Re:boo (Score:2, Informative)
(Well okay, it's not quite true. You could just about manage to spoof IPs for machines on the same ethernet segment as you. However, if you're on the same segment as an outbound mail server, you're probably allowed to send via that server anyway.)
Re:How about dynamic IPs? (Score:3, Informative)
Re:Why this is a big deal (Score:5, Informative)
So, as a spammer, you only have to publish an SPF for your own domain, and your mail is garanteed to be nonspam?
No, you have it wrong: Mail coming from hosts not allowed by the SPF, is guaranteed to violate the policy of the sender domain. SPF is basically saying: ``Hey, to whom is interested, mail coming from one of oud adresses, will always be send by these mailservers. So if you receive them from other means... We didn't do it!''
But indeed, if the domain and its users are trustworthy, you may decide that spam isn't likely to come from them. While ISP's might be trustworthy themselves, their users as a whole are not.
the rest of the world does seem to have to change their mail server configuration to keep mail flowing to AOL.
Wrong again, it's about mail flowing FROM @aol.com adresses. Mail going TOWARDS aol has nothing to do with it. Even if AOL will be implementing SPL while recieving mail themselves, if you don't use SPL, you're not blocked, and also, you need to change your DNS, not your mail server, if you want to implement SPL for outgoing mail of your domain.
Re:How about dynamic IPs? (Score:1, Informative)
You should read the SPF RFC. I just did, and just added SPF records to my DNS server. SPF includes support for specifying other IP ranges, domain names, and even specifying exceptions at a per-email-address level to address cases where someone has no idea what country they'll send email from next.
I'm a skeptical bastard with an extreme dislike for change, but a few hours after hearing about SPF for the first time, I'm serving '-all' SPF records from my domains. I was able to see it was a good idea when I finally understood that
SPF records provide a way to recommend restrictions on those who attempt to spoof your domain in their outbound email.
Re:Why this is a big deal (Score:5, Informative)
Re:boo (Score:3, Informative)
G:\>nslookup
> set q=txt
> aol.com
Server: XXXXXXXXXXXXXXX
Address: XXXXXXXXXXXXXXX
Non-authoritative answer:
aol.com text =
"v=spf1 ip4:152.163.225.0/24 ip4:20....
Damni! "RMX" was such a cooler acronym! (Score:3, Informative)
and for each bounced message, who knows how many are getting through. A friend of mine (an AOL user) actually had a spammer us his personal email address, and got not only a bunch of bounces, but angry emails and IMs.
The sooner this goes into effect, the better. It'll probably be a long time before we can block all email that doesn't come from a domain with SPF, but hopefully soon we can get rid of emails that are explicitly not authorized. (like those claming to be from my servers...)
Re:Why this is a big deal (Score:4, Informative)
However, in order to get things off the ground without having to wait for DNS servers and tools to support a new record type, it is also possible to publish the same information in a TXT record:
If your DNS server supports the SPF *type*, then you should ideally use that and provide the TXT record as a backup. Query tools that properly support SPF will probably look for the SPF type first and then requery for TXT on a failure, but it's up to the developer of course.
AOL will likely remove these SPF records today (Score:5, Informative)
See: this message on the SPF mailing list [gmane.org]
SPF is NOT a problem for you, (Score:5, Informative)
Two solutions.
1) The "hard" but proper way, setup SPF records from all the machines you will be sending mail from or
2) Simply send all your mail out through the box you get it in from. What's so hard about that?
Anyway, I'll be happy to let anon mail through just for your convenience, so you don't have to setup SPF once every 6 months, or wait for your email to get forwarded through your own mail server, if you'd be willing to go through and delete the hundred or so SPAMs I get each day. Sound like a fair deal?
Breaks Forwarding (Score:4, Informative)
Re:The really important question is... (Score:3, Informative)
As a matter of fact, there is nothing stopping spammers from registering a bogus domain, and making the entire internet part of their SPF
But it kills domain forging; they have to use their own bogus domains which can be quickly and easily blacklisted by other methods if they spam a lot. SPF says "This machine can be held accountable for mail sent for this domain," there's no magic if you're not willing to actually hold people accountable. But the contrapositive to that is, if someone says they're host is accountable and mail from that host is otherwise sound, then you should give them the benefit of the doubt.
What is needed is SPF and some sort of a trust between domains.
Mechanisms based on trust are either expensive or doomed to failure. So it has always been and so it will always be.
Re:Wrong. (Re:Nitpick (Re:Tag it)) (Score:2, Informative)
Unfortunately the W3C's sites seem to be ambiguous about this. However, somewhere it does state that ACRONYM is for pronouncable acronyms and ABBR is for unpronouncable acronyms and abbveviations (although I can't find the link to back this up). They probably could've made this less confusing, but they didn't.
At http://www.w3.org/TR/html4/struct/text.html#edef-A CRONYM [w3.org] where they actually define the standard, they give WWW as an example for ABBR.
Again, I'm just saying it's ambiguous, I'm not trying to start a flamewar.
Re:Tag it (Score:4, Informative)
More pertentely in this context: Slashcode doesn't support it. Even if the original submitter included it in their submission it would have been stripped out before it got to the editors.
Re:boo (Score:1, Informative)