As mentioned, certificate based authentication is the way to go. But a patch on the problem you have before you can implement certificate based authentication on all devices is to do the following to only treat the device with the faulty password with the badPwdCound rule.
- Create a custom attribute in Endpoints database, i.e. BadPassword
- Create two new enforcement profiles with type Endpoint Update. One to set the attribute BadPassword to True and one to set it to False
- Create two duplicates of the AD source, one with (&(&(sAMAccountName=%{Authentication:Username})(objectClass=user))(!(badPwdCount>=3))) the other a normal with out (!(badPwdCount>=3)).
Also change the 3 to just 1 instead.
- Create a copy of your 802.1x enforcement policy, add the Endpoint update enforcement profile to set the BadPassword attribute to False
- Duplicate your 802.1x service. In the copy add a service filter to check the endpoint database for the BadPassword=True
Name the service something BadPassword
Make sure the AD source is the one with the query (&(&(sAMAccountName=%{Authentication:Username})(objectClass=user))(!(badPwdCount>=1)))
Place it above the normal 802.1x service
Use same role mapping and the updated enforcement policy setting BadPassword to False
- In the normal 802.1x service add a new rule as rule number one, or at least very high
(Radius:Microsoft:MS-CHAP-Error EXISTS) and the Enforcement policy to set the BadPassword attribute to True
With this configuration a device with a bad password configured will not be able to send more authentications to the AD, but other devices will still be able to do so. When the user has updated the faulty password on the device, a successful authentication must take place on any of the other devices to reset the badPwdCount in AD. If I remember correct the counter is also reset after some time.
This isn't by any means a "best practice" configuration, but it will solve you problem. And yes, I have seen a customer run this "temporary fix" configuration for 10+ years... ;-)
------------------------------
Best Regards
Jonas Hammarbäck
MVP Guru, ACEX, ACDX #1600, ACCX #1335, ACX-Network Security
Aranya AB
If you find my answer useful, consider giving kudos and/or mark as solution
------------------------------
Original Message:
Sent: Oct 27, 2025 08:07 AM
From: Lord
Subject: 802.1x causing AD lockouts
With badPwdCount, you can only treat the symptoms, not the cause. You yourself mentioned the cause: "users change their Active Directory password and forget to update it on their wireless clients."
Our customers use certificate-based authentication, i.e., EAP-TLS. This means that no passwords are used and users are not blocked in AD. With EAP-TLS, however, you must ensure that the certificates on the end devices are renewed before they expire.
------------------------------
Regards,
Waldemar
ACCX # 1377, ACEP, ACX - Network Security
If you find my answer useful, consider giving kudos and/or mark as solution
------------------------------
Original Message:
Sent: Oct 27, 2025 04:59 AM
From: Sisant
Subject: 802.1x causing AD lockouts
The EAP method is not good when the password is expired. Windows cached the expired password.
The way to avoid it is EAP-TLS or EAP-TEAP, which is more convenient to do.
------------------------------
Give me a Kudo if it is useful.
Ratchapas
https://www.facebook.com/Aruba-News-Update-1401095559960142