Skip to content

Est. 2011ยทMicrosoft Partner ID 7033487ยทDelivery under 3 minยทSupport 7 days a week

Your vault is empty.

Free Fix Event ID 4771

Event ID 4771 Kerberos pre-authentication failed: tracing account lockouts

9 min read Updated October 5, 2026 Windows Server: AD, DNS & Group Policy

Fix it now

The domain controller refused a credential before issuing any ticket and wrote it down. 4771 is a record rather than a fault, and its value is in three fields: the account, the failure code, and the client address. The code says what was wrong; the address says which machine keeps offering it.

Run this on the domain controller, in an elevated PowerShell session

Get-WinEvent -FilterHashtable @{LogName='Security'; Id=4771} -MaxEvents 50 |
  Format-List TimeCreated, Message
  1. Read the failure code first. 0x18 means the password offered was wrong. 0x12 means the account was already disabled, expired or locked out when the request arrived, so the credential itself may be fine.
  2. Read the Client Address. That is the machine to go to. An address of ::1 means the request came from the domain controller itself, so look at what runs there rather than hunting a client.
  3. On that machine, look for the stale credential: a saved entry in Credential Manager, a service or scheduled task running as the account, a mapped drive, or a mobile device with an old password.
  4. Update or remove it, unlock the account with Unlock-ADAccount -Identity <name>, and watch whether new 4771 entries stop arriving from that address.

4771 is generated on domain controllers only. Collect it from all of them, because a client that fails against one DC will simply try another.

If the entries stop and the account stays unlocked, you have found it. The next section covers the cases where 4771 has nothing to do with a password.

Why it happens

Kerberos pre-authentication is the step where the client proves it knows the password before the KDC will issue a ticket granting ticket. Event ID 4771 is written every time that fails, on the domain controller that handled the request. Because it is generated at the moment of failure and carries the source address, it is the most direct tool there is for finding what keeps locking an account out.

Microsoft names three causes for the event, and only one of them is a bad password: the wrong password was provided, the user’s password has expired, or the domain controller does not have a certificate installed for smart card authentication. That third one catches people out badly, because the account is fine, the password is fine, and every attempt still fails.

Two fields decide which of those you have. Failure Code 0x18 is KDC_ERR_PREAUTH_FAILED, pre-authentication information was invalid – the wrong password. Failure Code 0x12 is KDC_ERR_CLIENT_REVOKED, the client’s credentials have been revoked, which covers a disabled, expired or locked-out account. Pre-Authentication Type tells you how the credential was offered: 2 is normal password pre-authentication, while 15, 16 and 17 are the smart card exchange.

A stale password cached on one machine

You have this one if 4771 with code 0x18 arriving in bursts from the same client address, often at a regular interval.

  1. Go to the machine at that address and check Credential Manager, services, scheduled tasks and mapped drives for the account.
  2. Update or remove the credential, then unlock the account.
  3. Watch for new events from that address before assuming you are finished – a machine can hold more than one stale copy.

After a password change, NTLM network access keeps working with the old password for a short period – five minutes by default, set by OldPasswordAllowedPeriod. That is why a stale credential can look intermittent rather than constant.

The account is already locked or disabled

You have this one if Failure code 0x12 rather than 0x18, sometimes from several addresses at once.

  1. Treat 0x12 as a symptom of an earlier lockout, not as the cause. Look further back for the 0x18 entries that filled the counter.
  2. Check whether the account is disabled or expired rather than locked, which produces the same code and no amount of unlocking will help.
  3. Unlock or re-enable, then go back and fix whatever produced the original failures.

Smart card logon with no usable certificate

You have this one if 4771 with a Pre-Authentication Type in the smart card range, and no wrong password anywhere.

  1. Check the domain controller has a valid certificate for smart card authentication – Microsoft names a missing one as a cause of this event.
  2. Verify it with certutil -dcinfo verify on the domain controller.
  3. If the certificate is the problem, the account and the card are not, and no password reset will change anything.

Something automated is guessing

You have this one if A run of 0x18 entries in a short window, often against several accounts, from an address you do not recognise.

  1. Microsoft’s own monitoring guidance for 4771 is that a number of 0x18 events in a short period can indicate a brute-force attempt against the password.
  2. Identify the source address before unlocking anything, and treat the unlock as the last step rather than the first.
  3. Check whether the address is inside your estate. A machine with malware behaves exactly like a stale credential except for the variety of accounts it touches.

Full reference

The fields that matter in the event

Field What to do with it
Account Name The account the ticket was requested for. A trailing $ means it is a computer account, which changes where you look
Client Address The machine that sent the request. ::1 means the domain controller itself
Failure Code The diagnosis: 0x18 wrong password, 0x12 account state
Pre-Authentication Type 2 for standard password pre-authentication, 15/16/17 for the smart card exchange, 0 for none

Failure codes you will meet alongside these two

Code Name Meaning
0x18 KDC_ERR_PREAUTH_FAILED Pre-authentication information was invalid. The wrong password was provided
0x12 KDC_ERR_CLIENT_REVOKED The client’s credentials have been revoked: disabled, expired or locked out
0x25 KRB_AP_ERR_SKEW The clock skew is too great, measured against the Kerberos policy tolerance
0xE KDC_ERR_ETYPE_NOTSUPP No supported encryption type in common. A different problem, and a different article

Which event to use for which question

  • 4771 answers “which machine offered a bad credential, and when”. It is generated on domain controllers only.
  • 4768 records a ticket granting ticket request, successful or failed, and is the AS-stage view of the same exchange.
  • 4769 records a service ticket request, which is where encryption type problems show up rather than password ones.
  • 4740 is the account lockout itself, and names the caller computer – useful as the marker you work backwards from.
  • 4625 covers logon failures that never reached Kerberos at all, such as NTLM authentication against a member server.

Working an account lockout end to end

  1. Collect 4771 from every domain controller for the account, not just the one you happened to open.
  2. Group by Client Address. One address dominating is a stale credential; many addresses is a person, a script, or something worse.
  3. Note the interval between attempts. A tidy repeating gap points at a service or scheduled task rather than a human.
  4. Visit the machine and clear the credential at source before unlocking the account.
  5. Unlock, then watch for an hour. If the events resume, you cleared one copy and there is another.

Computer accounts in this log

An account name ending in $ is a machine, and the same three fields apply. The usual causes are different, though: a machine whose account password no longer matches the directory, a cloned virtual machine sharing an identity, or a device restored from an image. Those show up as secure channel failures too, so check the machine’s own System log before assuming a person typed something wrong.

Every code this article covers

Code What it points at Source
Event ID 4771 Kerberos pre-authentication failed. Generated on domain controllers when the KDC does not issue a ticket granting ticket; carries the account, client address and failure code Microsoft Learn
0x18 KDC_ERR_PREAUTH_FAILED: pre-authentication information was invalid, which Microsoft glosses as the wrong password being provided Microsoft Learn
0x12 KDC_ERR_CLIENT_REVOKED: the client’s credentials have been revoked – the account is disabled, expired or locked out Microsoft Learn

Confirm the fix worked

  1. No new 4771 entries for the account on any domain controller for at least an hour.
  2. The account stays unlocked through a full working day.
  3. The machine that was the source no longer holds the old credential in Credential Manager, a service, or a scheduled task.
  4. A deliberate sign-in with the current password succeeds and produces a 4768 rather than a 4771.
  5. If the cause was smart card related, certutil -dcinfo verify passes on the domain controllers.

Questions people ask about this

Is 4771 always someone typing the wrong password?

No. Microsoft lists three causes: a wrong password, an expired password, and a domain controller with no certificate installed for smart card authentication. The failure code and the pre-authentication type separate them.

The client address is ::1. Which machine is that?

The domain controller itself. Look at what runs there under that account – a service, a scheduled task, or an administrative session – rather than searching the network.

Why do I see 0x12 with no 0x18 before it?

Because the account was already locked or disabled when those requests arrived. The failures that filled the lockout counter are earlier in the log, possibly on a different domain controller.

Do I need to collect this from every domain controller?

Yes. A client that fails against one will try another, so a single DC gives you a partial picture and a misleading count.

Does auditing 4771 cost anything?

No. It is an audit policy setting on domain controllers. Nothing about tracing a lockout requires a purchase.

Related error codes

Was this article helpful?

Your feedback helps us improve our documentation.

Related articles

Free Fix Kerberos Event ID 4 KRB_AP_ERR_MODIFIED: a duplicate SPN in the directory License Error DHCP Event ID 14: the scope address pool is exhausted and leases now fail License Error DFSR Event ID 5002: SYSVOL partners cannot hold an RPC session open Free Fix Event ID 1056: DHCP has no credentials for dynamic DNS registration on a DC
โ† Back to Knowledge Base