A real outage ID is better bait than a fake one. After Microsoft 365 suffered a widespread disruption on 12 November 2026 under incident EX1550543, attackers flooded inboxes with service-health style recovery mail that named that same code and told staff held messages would vanish within 24 hours unless they clicked restore. According to Guardz, ten email waves sent 5,481 malicious messages that reached 1,184 organizations across a three-day span, landing users on a credential-harvesting page on reallifeaccounting.com. Microsoft Threat Intelligence characterized the activity as credential phishing, Microsoft products blocked the campaign, and public reporting has not detailed confirmed account takeovers.
How EX1550543 turned into the lure
Microsoft’s own disruption on 12 November 2026 was a network capacity upgrade glitch, not a cyberattack. Outlook, Teams, SharePoint, and Copilot felt it. Recovery talk filled every tenant’s status page. That noise was the gift.
Operators did not invent a crisis. They forged notices that sounded like the recovery window staff already expected. Guardz describes lure body text such as “Mail delivery to your mailbox failed during the Microsoft 365 disruption on 12 November 2026” and “Action required: (4) messages held after the outage.” Subjects used an Outage-plus-incident-code pattern. The call to action was a “RESTORE MESSAGES” style link, not a random invoice.
Earlier waves on 9 and 10 November already used domains including newbooker.com and lesterscompany.com. The heavy fire came on 13 November, while EX1550543 recovery was still live. Microsoft Threat Intelligence detected the campaign that day. Peak-day mail rode Amazon SES from a short list of aged sender domains that included moments-notice.com, planetmaison.com, and newbooker.com, each with valid SPF and DKIM, so the path looked mail-authenticated even though the brand was fake.
Each wave dumped its batch in 3 to 22 minutes. That pace is industrial, not a lone actor typing one message at a time. Links moved through SendGrid-style tracking, then redirected to reallifeaccounting.com/themelist0.html behind Cloudflare. The hostname looks like a bookkeeping blog. It was a front for a harvest kit.
What the harvest kit actually did
Landing pages wore generic mindfulness-blog shells. Per-victim tracking beacons and unique session IDs fired before the form rendered, so operators could measure who opened what and which lure worked. The form collected credentials and posted them to a local index.php handler on the attacker side.
That is classic password capture on an attacker-controlled page. Nothing in the public Guardz analysis shows a reverse proxy sitting between the user and the real Microsoft login, live OTP relay, or session-cookie theft. Microsoft Threat Intelligence has not verified stolen passwords or MFA details. Guardz telemetry names volume and reach; it does not publish named victim organizations, phishing-kit file hashes, or a threat-actor brand.
If anyone did submit a password, the sensible response list is familiar: revoke active sessions, force password resets, and audit sign-in telemetry. Those steps clean up after a submission. They do not rewrite the fact that a reusable secret was the thing the form wanted.
| Control in the path | What this campaign used | Why it failed the user |
|---|---|---|
| Outage context | Real EX1550543 ID and timing | Staff already expected recovery mail |
| Email authentication | Amazon SES with valid SPF/DKIM | Auth of the pipe, not of Microsoft |
| Landing site | reallifeaccounting.com harvest form | Password typed to the wrong origin |
| Endpoint filters | Microsoft products blocked the wave | Stops delivery, not the lure design |
Valid SPF on a SES stream only proves the sending infrastructure was allowed to send for those aged domains. It never proves the message came from Microsoft. Helpdesks still hear “I got the outage restore mail” for days after a real incident ID hits the news. That human pressure is the point of riding a genuine service-health event.
Volume, blocks, and open gaps
Guardz counts 5,481 messages across 1,184 customer organizations. Ten distinct waves spanned roughly three days. Microsoft products blocked the campaign in the telemetry Guardz reviewed. No public write-up has confirmed successful account takeovers, named the organizations that clicked through, or attributed a known group.
Adjacent context matters only as cover story. The EX1550543 disruption itself supplied the incident number and the recovery window. It was not the identity compromise path. The compromise path, where it existed, was a human typing a password into a form that looked urgent because the outage was real.
Password-only harvest dies when the factor the form can steal is no longer enough to finish a login. Device-bound, origin-bound authentication leaves nothing reusable for a blog-shell form to post to index.php. That is the mechanical gap this wave aimed at, whether or not every recipient fell for it.
Operators will keep mining the next status-page incident the same way. The next EX-number will show up in subject lines within hours of a real glitch. Filters catch many of those waves. The ones that land still win if the only secret required is something a user can type on a wrong-origin page.
The defensive side of this case is covered in the related article on mfa2point0.com: preventing Microsoft 365 outage-themed credential phishing.
FAQ
How did the EX1550543 phishing campaign reach so many tenants so fast?
The EX1550543 phishing campaign reached 1,184 organizations because operators sent ten rapid email waves totaling 5,481 messages across a three-day span through Amazon SES with valid SPF and DKIM, according to Guardz. Each wave completed in 3 to 22 minutes, and the lures cited the real 12 November 2026 Microsoft 365 outage so the restore story matched what staff already saw on service health pages.
Did this campaign use an adversary-in-the-middle proxy?
Public reporting on the EX1550543-themed wave describes a credential-harvesting page on reallifeaccounting.com that posted submitted passwords to a local handler, not a confirmed adversary-in-the-middle reverse proxy or session-cookie theft. Microsoft Threat Intelligence characterized the activity as credential phishing and has not verified MFA capture or stolen session tokens for this campaign.
Were any account takeovers confirmed after the EX1550543 lures?
No confirmed account takeovers have been publicly detailed for the EX1550543 outage phishing wave. Guardz reports that Microsoft products blocked the campaign in observed telemetry, and Microsoft Threat Intelligence has not verified stolen passwords. Organizations that still fear a submission should revoke sessions, reset passwords, and review sign-in logs.
Why did valid SPF and DKIM fail to protect recipients?
Valid SPF and DKIM on the EX1550543 phishing mail only authenticated the Amazon SES sending path for the attacker-controlled domains, not Microsoft as the brand. Recipients saw authenticated mail about a real outage ID, which is exactly how outage-themed credential phishing defeats the mental shortcut that “authenticated equals safe.”
What should IT do if a user clicked a RESTORE MESSAGES link from this wave?
IT should treat a click-through on the EX1550543-themed restore lure as a possible password exposure: revoke active Microsoft 365 sessions for that account, force a password reset, and audit sign-in telemetry for unfamiliar clients or locations. Public sources have not confirmed mass takeovers, but a submitted password on reallifeaccounting.com is still a reusable secret until it is rotated and sessions are killed.