Fake Microsoft 365 outage alerts used in large phishing wave threatening to delete held emails
On November 13, 2026, Microsoft Threat Intelligence observed a coordinated, high-volume phishing campaign aimed at organisations across multiple sectors worldwide. The emails were disguised as Microsoft 365 service-health notifications and claimed that messages had been held during a service disruption and would be deleted within 24 hours unless the recipient acted to restore them.
The lures followed a recognisable pattern. Subjects read "Outage" followed by a random token, and display names combined a three-character, organisation-style prefix with "Outage Notification" or "Service Health". One sample referred to an Exchange Online incident on November 12 and said four messages were being held, with a prominent "RESTORE MESSAGES" button. Sender domains included moments-notice[.]com, planetmaison[.]com and newbooker[.]com. Email-delivery and click-tracking services generated recipient-specific links, and several samples led to reallifeaccounting[.]com.
The campaign was link-based. Microsoft saw no QR codes, no non-EML attachments and no evidence of malware delivery. Because the destination was off-brand and the emails impersonated Microsoft 365, Microsoft assesses the activity as likely credential phishing. However, it did not have the final page source, credential fields, submission telemetry or post-compromise activity, so credential capture is unconfirmed and no named actor or malware family is attributed.
Microsoft says messages in environments protected by Defender for Office 365 are being blocked or quarantined, and it has not observed indicators of compromise from this attack. For anyone who clicked and entered credentials, the risks include account compromise, mailbox access, business email compromise, internal phishing, inbox-rule manipulation and access to cloud collaboration data.
Why it matters
The lure exploits a trusted, routine kind of message: a service-health notice, with a deadline that pressures people to skip checks. Because the links are unique per recipient and the sender domains rotate, blocking a single domain or subject line is not enough, so any business relying on simple pattern matching is exposed. Since credential capture is unconfirmed, defenders must decide who clicked and treat those accounts as potentially compromised rather than waiting for proof.
What to do
- Search email logs for the subject pattern "Outage" plus a random token and the sender domains moments-notice[.]com, planetmaison[.]com and newbooker[.]com.
- Check click and web logs for visits to reallifeaccounting[.]com and identify which users clicked.
- Review users who clicked for suspicious sign-ins, new inbox rules and unusual mailbox access.
- Validate indicators in your environment before blocking, as shared email-delivery infrastructure can cause false positives.