It all started with one of those rare spam mails that got through.
This one was hawking address lists, much like the ones I occasionally receive to addresses that I can not turn into spamtraps. The message was addressed to, of all things, email@example.com. (The message with full headers has been preserved here for reference).
Yes, that's right, they sent their spam to root@. And a quick peek at the headers revealed that like most of those attempts at hawking address lists for spamming that actually make it to a mailbox here, this one had been sent by an outlook.com customer.
The problem with spam delivered via outlook.com is that you can't usefully blacklist the sending server, since the largish chunk of the world that uses some sort of Microsoft hosted email solution (Office365 and its ilk) have their usually legitimate mail delivered via the very same infrastructure.
And since outlook.com is one of the mail providers that doesn't play well with greylisting (it spreads its retries across no less than 81 subnets (the output of 'echo outlook.com | doas smtpctl spf walk' is preserved here), it's fairly common practice to just whitelist all those networks and avoid the hassle of lost or delayed mail to and from Microsoft customers.
I was going to just ignore this message too, but we've seen an increasing number of spammy outfits taking advantage of outlook.com's seeming right of way to innocent third parties' mail boxes.
So I decided to try both to do my best at demoralizing this particular sender and alert outlook.com to their problem. I wrote a messsage (preserved here) with a Cc: to firstname.lastname@example.org where the meat is,
The address email@example.com has never been subscribed to any mailing list, for obvious reasons. Whoever sold you an address list with that address on it are criminals and you should at least demand your money back.
Whoever handles firstname.lastname@example.org will appreciate the attachment, which is a copy of the message as it arrived here with all headers intact.
Peter N. M. Hansteen
What happened next is quite amazing.
If my analysis is correct, it may not be possible for senders who are not themselves outlook.com customers to actually reach the outlook.com abuse team.
Almost immediately after I sent the message to Ms Farell with a Cc: to email@example.com, two apparently identical messages from firstname.lastname@example.org, addressed to email@example.com appeared (preserved here and here), with the main content of both stating
This is an email abuse report for an email message received from IP 220.127.116.11 on Sat, 17 Feb 2018 01:59:21 -0800.
The message below did not meet the sending domain's authentication policy.
For more information about this format please see http://www.ietf.org/rfc/rfc5965.txt.
In order to understand what happened here, it is necessary to look at the mail server log for a time interval of a few seconds (preserved here).
The first few lines describe the processing of my outgoing message:
2018-02-17 10:59:14 1emzGs-0009wb-94 <= firstname.lastname@example.org H=(greyhame.bsdly.net) [192.168.103.164] P=esmtps X=TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128 CV=no S=34977 id=31b4ffcf-bf87-de33-b53a-0 email@example.com
My server receives the message from my laptop, and we can see that the connection was properly TLS encrypted
2018-02-17 10:59:15 1emzGs-0009wb-94 => peter <firstname.lastname@example.org> R=localuser T=local_delivery
I had for some reason kept the original recipient among the To: addresses. Actually useless but also harmless.
2018-02-17 10:59:16 1emzGs-0009wb-94 [18.104.22.168] SSL verify error: certificate name mismatch: DN="/C=US/ST=WA/L=Redmond/O=Microsoft Corporation/OU=Microsoft Corporation/CN=mail.protection.outlook.com" H="outlook-com.olc.protection.outlook.com"
2018-02-17 10:59:18 1emzGs-0009wb-94 SMTP error from remote mail server after end of data: 451 4.4.0 Message failed to be made redundant due to A shadow copy was required but failed to be made with an AckStatus of Fail [CO1NAM03HT002.eop-NAM03.prod.protection.outlook.com] [CO1NAM03FT002.eop-NAM03.prod.protection.outlook.com]
2018-02-17 10:59:19 1emzGs-0009wb-94 [22.214.171.124] SSL verify error: certificate name mismatch: DN="/C=US/ST=WA/L=Redmond/O=Microsoft Corporation/OU=Microsoft Corporation/CN=mail.protection.outlook.com" H="outlook-com.olc.protection.outlook.com"
What we see here is that even a huge corporation like Microsoft does not always handle certificates properly. The certificate they present for setting up the encrypted connection is not actually valid for the host name that the outlook.com server presents.
There is also what I interpret as a file system related message which I assume is meaningful to someone well versed in Microsoft products, but we see that
2018-02-17 10:59:20 1emzGs-0009wb-94 => email@example.com R=dnslookup T=remote_smtp H=prospectingsales-net.mail.protection.outlook.com [126.96.36.199] X=TLSv1.2:ECDHE-RSA-AES256-SHA384:256 CV=yes K C="250 2.6.0 <firstname.lastname@example.org> [InternalId=40926743365667, Hostname=BMXPR01MB0934.INDPRD01.PROD.OUTLOOK.COM] 44350 bytes in 0.868, 49.851 KB/sec Queued mail for delivery"
even though the certificate fails the verification part, the connection sets up with TLSv1.2 anyway, and the message is accepted with a "Queued mail for delivery" message.
The message is also delivered to the Cc: recipient:
2018-02-17 10:59:21 1emzGs-0009wb-94 => email@example.com R=dnslookup T=remote_smtp H=outlook-com.olc.protection.outlook.com [188.8.131.52] X=TLSv1.2:ECDHE-RSA-AES256-SHA384:256 CV=no K C="250 2.6.0 <firstname.lastname@example.org> [InternalId=3491808500196, Hostname=BY2NAM03HT071.eop-NAM03.prod.protection.outlook.com] 42526 bytes in 0.125, 332.215 KB/sec Queued mail for delivery"
2018-02-17 10:59:21 1emzGs-0009wb-94 Completed
And the transactions involving my message would normally have been completed.
But ten seconds later this happens:
2018-02-17 10:59:31 1emzHG-0004w8-0l <= email@example.com H=bay004-omc1s10.hotmail.com [184.108.40.206] P=esmtps X=TLSv1.2:ECDHE-RSA-AES256-SHA384:256 CV=no K S=43968 id=BAY0-XMR-100m4KrfmH000a51d4@bay0-xmr-100.phx.gbl
2018-02-17 10:59:31 1emzHG-0004w8-0l => peter <firstname.lastname@example.org> R=localuser T=local_delivery
2018-02-17 10:59:31 1emzHG-0004w8-0l => peter <email@example.com> R=localuser T=local_delivery
That's the first message to my domain's postmaster@ address, followed two seconds later by
2018-02-17 10:59:33 1emzHI-0004w8-Fy <= firstname.lastname@example.org H=bay004-omc1s10.hotmail.com [220.127.116.11] P=esmtps X=TLSv1.2:ECDHE-RSA-AES256-SHA384:256 CV=no K S=43963 id=BAY0-XMR-100Q2wN0I8000a51d3@bay0-xmr-100.phx.gbl
2018-02-17 10:59:33 1emzHI-0004w8-Fy => peter <email@example.com> R=localuser T=local_delivery
2018-02-17 10:59:33 1emzHI-0004w8-Fy Completed
a second, apparently identical message.
Both of those messages state that the message I sent to firstname.lastname@example.org had failed SPF verification, because the check happened on connections from NAM03-BY2-obe.outbound.protection.outlook.com (18.104.22.168) by whatever handles incoming mail to the email@example.com address, which apparently is where the system forwards firstname.lastname@example.org's mail.
Reading Microsoft Exchange's variant SMTP headers has never been my forte, and I won't try decoding the exact chain of events here since that would probably also require you to have fairly intimate knowledge of Microsoft's internal mail delivery infrastructure.
But even a quick glance at the messages reveals that the message passed SPF and other checks on incoming to the outlook.com infrastructure, but may have ended up not getting delivered after all since a second SPF test happened on a connection from a host that is not in the sender domain's SPF record.
In fact, that second test would only succeed for domains that have
in their SPF record, and those would presumably be Outlook.com customers.
Any student or practitioner of SMTP mail delivery should know that SPF records should only happen on ingress, that is at the point where the mail traffic enters your infrastructure and the sender IP address is the original one. Leave the check for later when the message may have been forwarded, and you do not have sufficient data to perform the check.
Whenever I encounter incredibly stupid and functionally destructive configuration errors like this I tend to believe they're down to simple incompetence and not malice.
But this one has me wondering. If you essentially require incoming mail to include the contents of spf.outlook.com (currently no less than 81 subnets) as valid senders for the domain, you are essentially saying that only outlook.com customers are allowed to communicate.
If that restriction is a result of a deliberate choice rather than a simple configuration error, the problem moves out of the technical sphere and could conceivably become a legal matter, depending on what outlook.com have specified in their contracts that they are selling to their customers.
But let us assume that this is indeed a matter of simple bad luck or incompetence and that the solution is indeed technical.
I would have liked to report this to whoever does technical things at that domain via email, but unfortunately there are indications that being their customer is a precondition for using that channel of communication to them.
I hope they fix that, and soon. And then move on to terminating their spamming customers' contracts.
The main lesson to be learned from this is that when you shop around for email service, please do yourself a favor and make an effort to ensure that your prospective providers actually understand how the modern-ish SMTP addons SPF, DKIM and DMARC actually work.
Otherwise you may end up receiving more of the mail you don't want than what you do want, and your own mail may end up not being delivered as intended.
Update 2018-02-19: Just as I was going to get ready for bed (it's late here in CET) another message from Ms Farell arrived, this time to an alias I set up in order to make it easier to filter PF tutorial related messages into a separate mailbox.
I wrote another response, and as the mail server log will show, despite the fact that a friend with an Office365 contract contacted them quoting this article, outlook.com have still not fixed the problem. Two more messages (preserved here and here) shot back here immediately.
Update 2018-02-20: A response from Microsoft, with pointers to potentially useful information.
A message from somebody identifying as working for Microsoft Online Safety arrived, apparently responding to my message dated 2018-02-19, where the main material was,
Hi,Based on the information you provided, it appears to have originated from an Office 365 or Exchange Online tenant account.To report junk mail from Office 365 tenants, send an email to email@example.com and include the junk mail as an attachment.This link provides further junk mail education https://technet.microsoft.com/en-us/library/jj200769(v=exchg.150).aspx.
I have asked for clarification of some points, but no response has arrived by this getting close to bedtime in CET.Kindly,
However I did take the advice to forward the offending messages as attachment to the junk@ message, and put the outlook.com abuse address in the Cc: on that message. My logs indicate that the certificate error had not gone away, but no SPF-generated bounces appeared either.
If Microsoft responds with further clarifications, I will publish a useful condensate here.
Update 2019-07-16: If you were wondering how I make the output of smtpctl spf walk useful as mentioned in this article, please see the article Goodness, Enumerated by Robots. Or, Handling Those Who Do Not Play Well With Greylisting for some specifics.
In other news, there will be PF tutorial at the 2018 AsiaBSDCon in Tokyo. Follow the links for the most up to date information.