Spam is being sent using my domain, what can I do?
Since it was released I've been using Google Apps FYD for stackednotion.com. All of the email I send goes through Google's servers and I use Gmail to view my email. I haven't had any issues before, however recently I've been seeing weird bouncebacks ending up in the catch all account. It looks like somebody is using my domain to send spam. I don't really want my domain getting marked with a bad reputation, so how can I stop this?
I have setup SPF, DMARC and DKIM on the domain by following the guides on Google Apps, here is my zone file:
; stackednotion.com [9548]
$TTL 86400
@ IN SOA ns1.linode.com. luca.stackednotion.com. 2012072633 7200 7200 1209600 86400
@ NS ns1.linode.com.
@ NS ns2.linode.com.
@ NS ns3.linode.com.
@ NS ns4.linode.com.
@ NS ns5.linode.com.
@ MX 1 ASPMX.L.GOOGLE.COM.
@ MX 5 ALT1.ASPMX.L.GOOGLE.COM.
@ MX 5 ALT2.ASPMX.L.GOOGLE.COM.
@ MX 10 ASPMX2.GOOGLEMAIL.COM.
@ MX 10 ASPMX3.GOOGLEMAIL.COM.
@ MX 30 ASPMX4.GOOGLEMAIL.COM.
@ MX 30 ASPMX5.GOOGLEMAIL.COM.
@ TXT "v=spf1 include:_spf.google.com ~all"
google._domainkey TXT "v=DKIM1; k=rsa; p=MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQKBgQDi19ipSdqDEpnJEWrVF7MarSLnlzXi0wPOHws2BY6oMQInbY5OHzdw9LcFr1biVvipErm4odyJfjZAIp5s8r6z50ZxQdW5Uwdy9krA1A9HMPaqVN+fm2xpntU//uXn0wD8sGc9CljYQIl+MusxQ690PfVGnAz/QeLqaZFxpHHmmQIDAQAB"
_dmarc TXT "v=DMARC1; p=quarantine; rua=mailto:[email protected]"
@ A 178.79.164.64
* A 178.79.164.64
_xmpp-server._tcp SRV 5 0 5269 xmpp-server.l.google.com.
_xmpp-server._tcp SRV 20 0 5269 alt1.xmpp-server.l.google.com.
Also here are the headers of a spam message (somebody tried to susbscribe me to a Zend mailing list, what kind of sick people are they?!?):
Return-Path: <[email protected]>
Received: (qmail 20117 invoked from network); 27 Jul 2012 06:51:01 -0000
Received: from exprod7mx200.postini.com (HELO psmtp.com) (64.18.2.92)
by rsmx2.zend.com with SMTP; 27 Jul 2012 06:51:01 -0000
Received: from source ([188.51.41.223]) by exprod7mx200.postini.com ([64.18.6.13]) with SMTP;
Fri, 27 Jul 2012 02:51:00 EDT
To: <[email protected]>
Subject: Invoice #48469883494
From: "Order" <[email protected]>
Date: Sat, 28 Jul 2012 09:40:03 +0300
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: IPS PHP Mailer
MIME-Version: 1.0
Content-type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
Message-ID: <20120728094003.9312B884F9D66F02CE7C@DELL-PC>
X-pstn-neptune: 500/484/0.97/100
X-pstn-levels: (S: 0.00346/89.11253 CV:99.9000 FC:95.5390 LC:95.5390 R:95.9108 P:95.9108 M:97.0282 C:98.6951 )
X-pstn-dkim: 0 skipp
Answer
At present, the way to reduce miscreants ability to send spam purportedly from your domain is to inform other mail servers what servers are allowed to send mail on your domains behalf. The mechanism is SPF and you already have a SPF record:
TXT "v=spf1 include:_spf.google.com ~all"
If blocking forgery attempts is your desire, this can be improved upon. Read the SPF Record Syntax page that describes what your SPF policy should be. If you have other mail servers sending mail on behalf of your domain, add them to the SPF record and change your policy to fail:
TXT "v=spf1 include:_spf.google.com -all"
Because SPF is so widely deployed, this will make a difference. But SPF has edge cases (forwards, email lists, etc.) where SPF policy fails, so most sites choose to be more liberal with SPF policy than you request. For example, if your policy is set to reject, and the message appears to be from an email list, most servers notate it somehow (the Authentication-Results header is defined for this purpose) and allow it to pass.
This is where DMARC comes in. You have already added a DMARC record:
_dmarc TXT "v=DMARC1; p=quarantine; rua=mailto:[email protected]"
Your policy is only to quarantine failing DMARC messages. If the DMARC reports do not indicate any valid messages being blocked, and/or you are willing to live with some edge cases where valid messages are rejected, then you can improve upon this with p=reject.
Not surprisingly, getting bounces from mail servers for spam purporting to be from one of my domains is exactly what compelled me to start DKIM signing my messages, so that I could deploy DMARC. DMARC is a policy mechanism that combines SPF and DKIM, so that domain owners can assert to other mail servers that, "If it's not from this list of IPs (SPF) and it's not DKIM signed, then [reject|quarantine|allow] it."
DMARC works brilliantly. Instead of getting bounce messages, now I get DMARC reports. I use Mail::DMARC to parse the reports and put the summaries into a database.
DMARC is still an IETF draft and it's not widely deployed. However, most large email providers have implemented it and coverage is surprisingly good. After deploying DMARC for my domain, I wrote a DMARC plugin for Qpsmtpd, so I could validate incoming messages against DMARC policy. I published some of my findings as a DMARC operator in a DMARC FAQ.
I mentioned edge cases earlier, so I feel compelled to share one.
Google handles misaligned messages (those that fail both SPF and DKIM alignment) by dropping them into the users Spam folder. I have become familiar with this because emails sent from my domain are generally treated well by gmail. The exception is for messages relayed through some email lists such as [email protected]. Messages I sent to that list get modified by the list processing software, invalidating my DKIM signature. When the message is forwarded to gmail recipients, those messages are marked as spam because a) I have published a reject DMARC policy, and b) that email list isn't a valid SPF sender of email from tnpi.net, and c) the DKIM signature bearing my domain fails validation.
There are workarounds, besides fixing the list software, such as adding the offending mailing list server to my SPF record. Some DMARC implementations will detect messages from mailing lists and reduce the policy severity (ie, Google quarantines my list messages rather than rejecting).
At present, there is no better way to inhibit phishing and spoofing attempts using your domain than a well implemented DMARC policy.