The Mysterious 10-Minute Authentication Delay
During a Windows 11 enterprise deployment, we encountered a perplexing issue: users logging in with Windows Hello for Business (WHfB) faced 10-minute delays when accessing DFS shares, followed by sudden resolution. Here’s how we diagnosed and fixed it.
The Root Cause: Kerberos + Zscaler Timing Conflict
Key Findings:
- Zscaler Private Access (ZPA) took time to establish tunnels to domain controllers.
- WHfB authentication completed before ZPA was ready.
- Initial Kerberos requests failed silently, with Windows retrying after 10 minutes (default FarKdcTimeout).
Proof of Concept:
Running this command immediately fixed access:
klist purge_bind # Forces Kerberos ticket renewal
The Role of FarKdcTimeout in Kerberos Retries
What is FarKdcTimeout?
A Windows Registry value that controls how long Windows waits before retrying failed Kerberos requests. Default: 10 minutes.

Registry Location:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa\Kerberos\ParametersShould You Modify It?
- Decrease value (e.g., 5 mins): Faster retries but more network traffic
- Keep default (10 mins): Stable but longer user delays
- Our Solution: Scheduled klist purge_bind (no registry changes)
Note: Modifying FarKdcTimeout requires careful testing.
The Fix: Automated Kerberos Ticket Renewal
Option 1: Scheduled Task (Recommended)
- Create a PowerShell script (Reset-Kerberos.ps1):
klist purge_bind
- Deploy via Intune/GPO to run 10 seconds after logon. (Create a scheduled task for this)
Option 2: Adjust FarKdcTimeout (If Needed)
Set-ItemProperty -Path "HKLM:\SYSTEM\CurrentControlSet\Control\Lsa\Kerberos\Parameters" -Name "FarKdcTimeout" -Value 0x00000258 -Type DWordLessons Learned
- WHfB + Zscaler can have hidden timing dependencies.
- Kerberos retries are governed by FarKdcTimeout.
- Scheduled tasks beat registry hacks for reliability.
