Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Kerberos armouring wraps the authentication exchange inside a tunnel keyed by the computer’s own credential, closing the offline password attacks that target pre-authentication.

What Is Kerberos Armouring (FAST)?
The pre-authentication step in Kerberos has an awkward property. To prove it knows the password, a client encrypts a timestamp with a key derived from that password and sends it. Anyone who captures that message, or who can persuade a domain controller to produce a comparable structure, holds material that can be attacked offline at whatever speed their hardware allows.
Kerberos armouring closes that gap. Defined by RFC 6113 as Flexible Authentication Secure Tunnelling—FAST—it wraps the AS and TGS exchanges inside an outer tunnel whose key comes from a credential the user does not supply: the computer account’s own ticket-granting ticket. The user’s password-derived material is then protected by a 120-character machine password that no human chose and no human knows.
Three things follow from armouring the exchange.
FAST also enables the features that depend on a protected channel: compound authentication, where a service receives both the user’s identity and the device’s, and Dynamic Access Control, which uses claims about user and device attributes to make file access decisions. If you have ever wanted an access rule that says “this document opens only from a managed device” inside an on-premises file estate, this is the plumbing it needs.
Armouring is not a single switch. Both ends must support and be configured for it.
The KDC setting offers three levels beyond disabled: supported, where armouring is used when the client can; Always provide claims; and Fail unarmored authentication requests, which rejects anything that cannot armour. That last one is the security objective and the operational hazard.
Enable client support first, broadly, and leave it to settle. Then set the KDC to supported and let armoured and unarmoured authentication coexist while you measure. Only when you can account for every remaining unarmoured request should you consider failing them.
The things that will still be unarmoured are predictable: non-Windows systems using Kerberos, such as Linux servers joined with realmd or SSSD and Java applications using their own Kerberos stack; workgroup machines; appliances that authenticate against the domain; and anything in a trusting domain that has not been upgraded. Across a forest trust, armouring does not extend automatically, which catches organisations whose user and resource domains were raised at different times.
Use domain controller event data to find them before you enforce. The cost of getting this wrong is an authentication outage for the systems you did not know about, which in a hospital or a manufacturing plant is not a theoretical inconvenience.
It is worth being precise, because FAST is sometimes described as solving Kerberos security generally. It does not stop Kerberoasting: that attack uses a legitimate service ticket request from an already-authenticated user, and the ticket is encrypted with the service account’s key regardless of whether the request was armoured. Removing RC4 and using Group Managed Service Accounts is the answer there.
It does not prevent golden tickets, because an attacker holding the KRBTGT key is forging tickets rather than participating in an exchange. It does not address pass-the-ticket, delegation abuse, or anything that follows from endpoint compromise, since a compromised machine can armour perfectly well using its own credential.
Treat it as one control in a layered Active Directory security programme: cheap to enable, genuinely effective against a specific class of credential attack, and worth combining with sensible ticket lifetimes, the Protected Users group, and AES-only encryption.
At Privia Security, we review domain authentication hardening and test it against current tooling in our infrastructure and architecture review services.
No. Kerberoasting works by an authenticated user requesting a service ticket for an account with a service principal name, and that ticket is encrypted with the service account’s own key whether or not the request was armoured. The effective controls are Group Managed Service Accounts, very long random service account passwords, and disabling RC4 ticket encryption.
It configures domain controllers to reject any Kerberos authentication that does not use FAST. That is the strongest setting and also the one most likely to cause an outage, because non-Windows Kerberos clients, appliances, and machines in unupgraded trusting domains cannot armour. Measure unarmoured traffic with the setting at supported before you enforce it.
Support is uneven. MIT Kerberos implements FAST and SSSD can be configured to use it on domain-joined Linux systems, but many applications bundle their own Kerberos stack that does not. Java applications are a frequent exception. Inventory your non-Windows Kerberos clients and test each one before moving the KDC to enforcement.
It is a FAST-dependent feature where the service receives an identity for both the user and the device they are using, so authorisation can depend on both. Combined with Dynamic Access Control claims, it supports rules such as allowing a file share only from a managed workstation. It requires armouring because the device assertion travels inside the protected tunnel.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.