Subscribe
Cybersecurity

Microsoft confirms Entra Connect Health flaw was exploited to steal tenant data

Sunday, November 22, 2026 · 5 min read

Microsoft confirmed on Nov 21 that attackers exploited a flaw in its Entra Connect Health service to steal tenant data; the flaw went undetected for 109 days, from its July 23 introduction to its Nov 9 detection, and stayed live for 119 days until the server-side fix on Nov 19.

Bearish$MSFT

📊 Scorecard: our Bearish call on $MSFT is scored from $484.52, the last close before it ran. Its first result comes after the next close.

Microsoft confirms Entra Connect Health flaw was exploited to steal tenant data

The News

Microsoft confirmed on Nov 21 that attackers exploited a vulnerability in its Entra Connect Health service to gain unauthorised access to administrator email addresses and tenant configuration information. The flaw was in the Connect Health API: introduced on July 23, 2026, detected on Nov 9 — 109 days later — fixed server-side on Nov 19, and the investigation concluded on Nov 20 with a confirmed finding that unauthorised read-only access to tenant-level metadata had occurred. Notifications to affected organisations began on Nov 20.

The exposed data included email addresses of tenant technical contacts and administrator accounts; tenant IDs, display names, initial domain names, country codes and regional information; Microsoft Entra licence tiers and directory synchronisation status, including whether sync was enabled and the last sync timestamp; Connect Health onboarding status; and an expired certificate once used to register Connect Health agents with a Microsoft internal service decommissioned in late 2024, which was still returned by the API. Microsoft says the flaw did not expose user passwords, password hashes, authentication tokens, valid cryptographic keys, on-premises server and Active Directory topology details, or Azure subscription information.

The flaw could potentially have allowed changes to Connect Health monitoring configurations, but Microsoft found no evidence of configuration changes. Microsoft has not disclosed how many organisations were impacted, who exploited the flaw, or when the unauthorised access began; BleepingComputer counted more than a dozen administrators reporting receipt of the notification in a Reddit thread. No customer action is required — the fix is server-side.

A flaw in Microsoft's own identity plumbing went undetected for 109 days — and the company still cannot say who used it or when.

Why It Matters

This matters because Entra is the front door to Microsoft's cloud. When a company signs its staff into Microsoft 365, Azure or Teams, the identity check runs through Entra, and Connect Health is the monitoring layer that watches that connection. A flaw in that layer does not just affect one product — it touches every organisation that trusts Microsoft to know who its administrators are and how its directories sync. Tenant metadata sounds dull, but administrator email addresses plus licence tiers plus sync status is a targeting list.

The 109-day window is the second problem. The flaw was introduced on July 23 and not detected until Nov 9, which means it was available to exploit for more than three months before anyone noticed. Microsoft has not said when the unauthorised access began, so the exposure window could be nearly that long — or much shorter. Either way, a flaw that lives undetected for a quarter in a security product is a process failure, not just a bug.

The third problem is what was not stolen. Microsoft is clear that passwords, tokens and valid keys were not exposed, and that matters — it limits the direct damage. But an expired certificate from a service decommissioned in late 2024 was still being returned by the API, which suggests old plumbing was left connected long after it should have been switched off. The lesson for customers is that even the identity provider's housekeeping matters, and for Microsoft it is a reminder that its most sensitive product is also its most scrutinised.

Analyst's Take

Our call is Bearish. A vulnerability in Microsoft's identity platform that stayed live for 109 days, was exploited to steal tenant data, and Microsoft has not disclosed who exploited the flaw or when the unauthorised access began is a bad look for the company that runs the sign-in system for much of the corporate world. The exposed data was metadata, not passwords — but metadata is exactly what a targeted attacker needs to pick its next victims.

What to watch next is whether Microsoft says more about the scale: how many organisations were hit, and when the unauthorised access began. The more than a dozen administrators reporting notifications on Reddit is BleepingComputer's tally of a thread, not Microsoft's count, and the real number could be far larger. Also watch whether any affected organisation reports follow-on phishing or intrusion attempts built on the stolen administrator email addresses.

For investors, the near-term risk is reputational rather than financial — no customer action is required and the fix is already in place — but identity is the product Microsoft sells as its moat. Microsoft closed Friday at $484.52, down 2.8%, in a broad market rout, so this story is not the reason the stock fell. The reason it still matters is trust: every enterprise that routes its sign-ins through Entra has just been reminded that the plumbing itself can be the weak point. Reader poll: Does a 109-day flaw shake your trust in Microsoft — yes or no? Vote: Yes, it does, No, still trust it, Need more detail.

Are you bullish or bearish on this story?

By The Numbers

109 daysFlaw live from introduction to detection
Jul 23–Nov 19Flaw introduced to server-side fix
Nov 20Investigation concluded; notifications began
>12Administrators reporting the notification, per BleepingComputer's Reddit tally
0Passwords, tokens or valid keys exposed, per Microsoft
$484.52Microsoft close on Friday, down 2.8%

FAQ

What is Entra Connect Health?

Entra Connect Health is a Microsoft service that monitors the health of the systems companies use to connect their own user directories to Microsoft's cloud — it watches whether sign-ins and synchronisation are working. It is part of Microsoft Entra, the identity platform behind Microsoft 365 and Azure sign-ins.

What is tenant metadata?

A tenant is one organisation's account inside Microsoft's cloud. Tenant metadata is information about that organisation — its name, its domain, which licence tier it pays for, and whether its user directory is syncing — rather than the contents of its emails or files. It is useful to an attacker planning a targeted attack.

What is a server-side fix?

A server-side fix is a repair Microsoft makes on its own computers, with no action needed from customers. Because the flaw lived in Microsoft's Connect Health API rather than in software on customer machines, Microsoft could close it centrally — which is also why no customer action is required now.

What does read-only access mean?

Read-only access means the attacker could look at data but not change it. Microsoft found no evidence that monitoring configurations were changed, though it says the flaw could potentially have allowed such changes.

👁

Source: BleepingComputer

Get the daily briefing. AI news with a markets lens — tickers, bullish/bearish reads, zero hype. Subscribe free on Substack

More on $MSFT

All $MSFT coverage →

More from AI Agent Update