Microsoft consumer mail rejects legitimate messages lacking DMARC
Original title: Microsoft’s email security is unexpectedly blocking legitimate business mail
Bounces citing 550 5.7.1 or 550 5.7.23 began November 5 and also hit small senders.
Mail sent to Outlook, Hotmail, Live, and MSN has been rejected when domains lack DMARC, SPF, or DKIM, CSO Online reports. Microsoft’s rule since May 2025 required DMARC only from domains sending more than 5,000 messages a day. A Microsoft moderator called it a service degradation under investigation. MailRoute says a monitoring DMARC record can stop these bounces when SPF or DKIM aligns.
Unconfirmed: Microsoft has not confirmed a policy change; a moderator calls it a service degradation, and a researcher says a bug may also be involved.
- Refusals began November 5 and were ongoing at publication.
- Some domains with working SPF, DKIM, and DMARC still bounced.
- Non-delivery reports cite 550 5.7.1 or 550 5.7.23 errors.
- Microsoft has not published a policy or service-health notice.
- Search logs since November 5 for missing SPF, DKIM, or DMARC.
Impact: Legitimate mail to Microsoft consumer inboxes is rejected
| Product | Affected |
|---|---|
| Microsoft Outlook, Hotmail, Live, and MSN | — |
- 1.Publish a DMARC TXT record of v=DMARC1; p=none if one is missing.
- 2.Search mail logs since November 5 for SPF, DKIM, or DMARC bounce text.
- 3.Enable DKIM on Microsoft 365 domains that do not already use it.
- Vendors
- Microsoft
- Organizations
- MailRouteDickson Research
From the original article
5 min read at csoonline.com
Microsoft’s email filters meant to protect users from malicious messages appear to be rejecting legitimate ones. Over the last few days, some mail sent to Microsoft consumer addresses have bounced if they did not publish a domain-based message authentication, reporting, and conformance (DMARC) record. This includes messages to Outlook, Hotmail, Live, and MSN.
csoonline.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.