


Perceptive Security
SOC/SIEM Consultancy

Azure CLI Password Spray Hits at Least 78 Microsoft Accounts in 81M+ Attempts
Published:
1 juli 2026 om 05:46:03
Alert date:
1 juli 2026 om 06:04:06
Source:
thehackernews.com
Cloud & Virtualization, Identity & Access
Cybersecurity researchers at Huntress have identified a massive, ongoing, automated password spray attack targeting Microsoft's Azure CLI. The campaign compromised at least 78 Microsoft accounts across more than 81 million attempts. The attack activity was observed between June 12 and June 26, originating from an IPv6 address range (2a0a:d683::/32) controlled by internet infrastructure provider LSHIY LLC (AS32167). The scale and automation of the attack suggest a sophisticated threat actor leveraging bulk credential stuffing techniques against Azure environments. Organizations using Azure CLI are urged to review their authentication logs and enforce strong account protections.
Technical details
A massive, ongoing, automated password spray attack targeted Microsoft's Azure CLI between June 12 and June 26, 2026. The threat actor made over 81 million login attempts and successfully compromised at least 78 Microsoft accounts across 64 organizations. The attack originated from an IPv6 address range (2a0a:d683::/32) controlled by LSHIY LLC (AS32167), with some IPs resolving to the U.S. and a few to China. The campaign leveraged the deprecated OAuth 2.0 Resource Owner Password Credentials (ROPC) flow, which bypasses Conditional Access Policies (CAP) because ROPC does not route through the standard authorization endpoint where CAP policies are enforced. This allowed attackers to bypass MFA in scenarios where: MFA was enforced only for specific apps (not 'All Cloud Apps'), MFA was enforced only for specific user groups (e.g., Admins only), or MFA was enforced only for requests from non-trusted locations. The attack used old, previously breached username/password combinations that had never been rotated. Compromise rates averaged 2-4 accounts per day from June 12-21, with a spike of 12 accounts on June 19, and 30 identities across 23 businesses on June 22. Eight impacted businesses had no MFA policy at all. Huntress observed credential spray attack volume surge by over 155 times across its customer base, with a current mean of ~1,964 failed attacks per month per protected tenant.
Mitigation steps:
1. Require MFA for All Users, All Cloud Apps, and All Client App types when configuring Conditional Access Policies (CAP) — do not scope MFA to specific apps, groups, or locations only. 2. Block or restrict the Azure CLI application for non-admin users. 3. Disable or block the legacy ROPC OAuth 2.0 flow where possible, as it is deprecated in OAuth 2.1 and incompatible with MFA. 4. Rotate all credentials that appear in known breached password combo lists, especially old/stale passwords that have never been changed. 5. Block or monitor traffic from the IPv6 range 2a0a:d683::/32 (LSHIY LLC, AS32167). 6. Prioritize incident response based on credential validity — focus on accounts where login was successful. 7. Audit existing CAP configurations to ensure legacy authentication protocols like ROPC cannot bypass enforcement. 8. Monitor for unusual Azure CLI authentication events, especially those using the ROPC flow from unexpected IP ranges.
Affected products:
Microsoft Azure CLI
Microsoft Azure / Entra ID (Conditional Access Policies)
Microsoft 365 / Microsoft accounts using OAuth 2.0 ROPC flow
Related links:
https://www.huntress.com/blog/lshiy-password-spray-attack
https://bgp.he.net/net/2a0a:d683::/32
https://bgp.he.net/AS32167#_prefixes6
https://auth0.com/docs/get-started/authentication-and-authorization-flow/resource-owner-password-flow
https://learn.microsoft.com/en-us/entra/identity-platform/v2-oauth-ropc
Related CVE's:
Related threat actors:
IOC's:
IPv6 range: 2a0a:d683::/32 (LSHIY LLC, AS32167), ASN: AS32167 (LSHIY LLC), OAuth flow: Resource Owner Password Credentials (ROPC) grant type, Azure CLI application used as authentication client by threat actor
This article was created with the assistance of AI technology by Perceptive.
