Skip to content

Entra Connect Health flaw exploited to read tenant metadata

Microsoft says an Entra Connect Health API flaw gave unauthorized read-only access to tenant metadata and admin emails. Active since July 23, fixed server-side November 19. No CVE assigned.

Published 3 min read

Microsoft has told affected customers that an attacker read tenant metadata through a flaw in the Entra Connect Health API. In the notification, Microsoft concluded that "unauthorized read-only access to tenant-level metadata associated with your organization had occurred" — not a theoretical exposure, but access that the company's investigation says took place. There is no public advisory and no CVE as of this writing; Microsoft is notifying affected tenants directly, and this post is based on those notifications and reporting at BleepingComputer and Neowin.

What was exposed

Per the notification, the accessed metadata includes:

  • Email addresses of tenant technical contacts and administrator accounts
  • Tenant IDs, display names, initial domain names, country codes, and regional information
  • Entra license tiers and directory synchronization status, including whether sync is enabled and the last sync timestamp
  • Connect Health onboarding status
  • An expired certificate once used to register Connect Health agents

Microsoft says the flaw did not expose passwords or password hashes, authentication tokens, valid cryptographic keys, on-premises server or Active Directory topology, or Azure subscription data, and that it found no unauthorized configuration changes. The data is low-sensitivity on its own, but admin and technical-contact email addresses plus tenant naming are exactly the inputs for targeted phishing and tenant-identification work ahead of a follow-on attempt.

Timeline and exploitation status

Microsoft's dates, as relayed in the notifications:

  • July 23, 2026 — the vulnerability was introduced.
  • November 9, 2026 — Microsoft detected the issue.
  • November 19, 2026 — a server-side fix was deployed.
  • November 20, 2026 — the investigation concluded and customer notifications began.

Microsoft has not said when exploitation actually started, how many tenants were affected, or who was behind it. July 23 is the earliest the flaw could have been abused, not a confirmed start date. Because the fix is server-side, there is no patch to install for the flaw itself.

Action checklist

  1. Check whether your organization received a notification in the Microsoft 365 message center or by email; do not assume you are clear because no public advisory exists.
  2. Treat the exposed admin and technical-contact addresses as burned for phishing purposes — brief those users and your help desk to expect targeted lures referencing your tenant by name.
  3. Review sign-in and audit logs for the affected admin accounts across the July 23–November 19 window for anomalous access, even though Microsoft reports no configuration changes.
  4. Rotate the expired Connect Health agent registration certificate if it is still present in your environment, and confirm current agents use a valid one.
  5. Nothing to patch locally for this flaw — the fix was server-side — but use the notification as a prompt to re-check conditional-access and MFA coverage on privileged accounts.

Context

This lands in a run of identity-plane disclosures around Entra: metadata and token-handling bugs in the service that fronts authentication for millions of tenants, where a provider-side flaw becomes every customer's exposure at once and there is nothing for the customer to patch. The reassuring part — no credentials, tokens, or keys — is also the limit of what Microsoft has confirmed during an open investigation. Read "no evidence of configuration changes" as a provisional statement, not an all-clear, and keep the affected admin identities under closer watch than usual.

Related stories