# Microsoft says attackers used Entra Connect Health flaw to read tenant data

Microsoft is notifying customers that a flaw in the Entra Connect Health API was exploited to read administrator email addresses and tenant configuration metadata. The number of affected tenants has not been disclosed.

Source: https://junglewise.ai/news/microsoft-says-attackers-used-entra-connect-health-flaw-to-read-tenant-data  
Byline: Junglewise News Desk, https://junglewise.ai/news  
Topic: Breaches and incidents  
Published: 22 November 2026  
Reading time: 5 min

This is a developing story.

![Close-up of fiber optic cables plugged into a network patch panel in a server rack](https://images.unsplash.com/photo-1594915440248-1e419eba6611?ixid=M3w0NDU4NjF8MHwxfHNlYXJjaHw0fHxlbnRlcnByaXNlJTIwbmV0d29yayUyMGFjY2VzcyUyMGNvbnRyb2x8ZW58MXwwfHx8MTc5NTMwNzA2OXww&ixlib=rb-4.1.0&w=1600&q=80&fm=jpg&fit=crop)

Picture: Photo by [Kirill Sh](https://unsplash.com/@kirill2020?utm_source=junglewise&utm_medium=referral) on [Unsplash](https://unsplash.com/?utm_source=junglewise&utm_medium=referral)

Microsoft has begun notifying organisations that attackers exploited a vulnerability in Entra Connect Health to read administrator email addresses and tenant configuration data, [BleepingComputer reported](https://bleepingcomputer.com/news/security/microsoft-entra-connect-health-flaw-exploited-to-steal-tenant-data). The flaw sat in an API belonging to the service, and the data it handed over was read-only metadata about the tenant itself: who administers it, what it is called, where it is, what licences it holds and whether directory synchronisation is running.

Nothing that would let an intruder authenticate was exposed, according to the same reporting: no passwords or password hashes, no authentication tokens, no usable cryptographic keys, no details of on-premises server or Active Directory layout, and no Azure subscription data. What was taken is nonetheless a usable target list. An attacker who knows the technical contact and administrator addresses for a specific tenant, along with its display name and initial domain, has useful material for a targeted phishing message about an identity service those people administer. [Microsoft](/threats/vendors/microsoft)'s own guidance to affected customers warns about phishing that references Entra or Connect Health.

Entra Connect Health is the cloud service administrators use to monitor the health, performance and alerts of on-premises identity infrastructure from the Microsoft Entra admin centre. The incident has not been independently confirmed beyond BleepingComputer's reporting and Microsoft's statements to that outlet, and several key questions remain open: Microsoft has not said who exploited the vulnerability, how many tenants were affected, or when the attacks began.

## What to do now

Microsoft's notification states that no software updates or other customer remediation steps are required, because the fix was applied server side. That does not mean there is nothing to do.

1. Check whether you received a notification. Microsoft said it contacted impacted customers through Azure Service Health, so review Service Health notifications for the tenant rather than waiting for an email to surface in an inbox. Managed service providers reportedly received separate notices per customer tenant, in some cases hours apart, so one notification does not mean only one tenant was touched.
2. Tell your security team and your service desk to expect targeted phishing and social engineering aimed at administrator accounts, specifically lures that reference Entra, Connect Health or tenant onboarding. The attacker already knows the names involved, which removes the usual tell of a generic message.
3. Confirm that every privileged account uses phishing-resistant multifactor authentication. This is the control that holds when a convincing message gets through.
4. Review your Connect Health configuration and check Entra audit and sign-in logs for suspicious activity around and after the exposure window, which runs from 23 July 2026, when the flaw was introduced, to the 19 November 2026 fix.
5. Run Microsoft's PowerShell script, published at `https://aka.ms/ConnectHealth/scripts`, to see which categories of tenant metadata could have been exposed. Read the output with care: it reports current configuration values, not a record of what was actually accessed, so it tells you what an attacker would have seen if they queried your tenant today rather than what they saw at the time.

There is no patch to deploy and no agent to update. The practical work is assuming the metadata is in someone else's hands and making the accounts it points at harder to compromise.

## Technical details

The vulnerability was in the Entra Connect Health API. Microsoft discovered it on 9 November 2026 and determined that it had been introduced on 23 July, meaning the exposure window ran for roughly four months before detection. A server-side fix was deployed on 19 November 2026, ten days after discovery and 119 days after the flaw first shipped.

Microsoft's investigation found that the flaw had been exploited, giving unauthorised parties access to tenant and service-monitoring metadata, with the potential to modify Connect Health monitoring configurations. On 20 November 2026 the company concluded its investigation and said the access that occurred was read-only, with no unauthorised configuration changes identified. Telemetry was reviewed to identify which customers had been affected by unauthorised API calls, and notifications began after that review closed.

The data reportedly exposed:

- email addresses belonging to administrator accounts and to the technical contacts listed for the tenant
- tenant identifiers, display names and initial domain names, plus country and regional fields
- the tenant's Microsoft Entra licence tier, and directory synchronisation status including whether sync was turned on and when it last ran
- whether the tenant had been onboarded to Connect Health
- a certificate, now expired, that Connect Health agents once used to register with an internal Microsoft service retired in late 2024

The certificate is the detail worth pausing on. Microsoft said it was no longer in use and presented no risk. It had nevertheless remained in Connect Health storage since the service it belonged to was retired, and the `GetTenant` API kept returning it to callers. A certificate tied to a service retired in late 2024 was still being returned by the API in 2026.

A Microsoft spokesperson told BleepingComputer: "We identified and addressed a security issue affecting Microsoft Entra Connect Health that could have resulted in unauthorized access to tenant-level contact and configuration metadata." The company confirmed it had fixed the vulnerability and notified impacted customers, and said it provided guidance and support through Azure Service Health. It declined to give further details.

Scale is unclear. Microsoft has not published a number of affected organisations, and the only public signal of breadth so far is administrator reports: more than a dozen said on Reddit that they had received the notification, and some managed service providers said they received notices for multiple customer tenants. Those accounts are unverified. Microsoft has not published indicators of compromise that would let an organisation check its own logs for the attacker's activity.

## On Junglewise Threat Intelligence

- [Microsoft](https://junglewise.ai/threats/vendors/microsoft)

## Related articles

- [Microsoft signing credential suspected in poisoned agent-framework PyPI releases](https://junglewise.ai/news/microsoft-agent-framework-python-package-briefly-poisoned-with-a-credential)
- [Phishing emails posing as Microsoft 365 outage notices ask staff to restore held mail](https://junglewise.ai/news/phishing-emails-posing-as-microsoft-365-outage-notices-ask-staff-to-restore-held)
- [FBI investigating apparent breach of nationwide police database, CNN reports](https://junglewise.ai/news/fbi-investigating-apparent-breach-of-nationwide-police-database-cnn-reports)
- [Russian group Star Blizzard used one hacked US firm to break into another](https://junglewise.ai/news/russian-group-star-blizzard-used-one-hacked-us-firm-to-break-into-another)
