Not sure how this happened, but one of our client’s domains was running about 23 minutes ahead of time.  Short of it having a flux-capacitor this needed to be fixed;

NB: Thanks to the very informative article at Microsoft (http://support.microsoft.com/kb/816042) and Windows Networking (http://www.windowsnetworking.com/articles_tutorials/Configuring-Windows-Time-Service.html)

Open up regedit on the PDC and navigate to;

• HKLM\SYSTEM\CurrentControlSet\Services\W32Time\Parameters\Type

This is the how the time service should be updated, change this to NTP (for sync’ing externally) NT5DS will check the server’s own CMOS clock.

• HKLM\SYSTEM\CurrentControlSet\Services\W32Time\Config\AnnounceFlags

Change this from 10 to 5.

• HKLM\SYSTEM\CurrentControlSet\Services\W32Time\Parameters\NtpServer

Now this is your source of time.  I use pool.ntp.org,0×1 (note the 0×1 is required when you use the DNS name);  if using IP address (in this case pool.ntp.org), you can use 202.174.111.169.

Then restart your W32Time service;  updates can take up to an hour so you can force update by using 

• w32tm /resync /rediscover

Then just double-check your eventlogs and ensure that are no errors/warnings.