I also had a similar problem after upgrading to 4.2-release, an

upgrade to the snapshot from Nov. 17 didn't help:
http://marc.info/?l=openbsd-misc&m=119602370303500&w=2
Even though I didn't fix the hardware yet, now with the snapshot from

Dec. 4 installed, ntpd synced the clock fine, at least so far (for a

few days), and I didn't dare to reboot the system yet as I fear that

the clock might not sync again after that. ;-)
If the clock of my system makes problems again, I'll try Otto's diff.

Thank you for that, Otto! :-)
> As you can see below, it needed to have a correction allow > then the

quoted text
> limit of 500 ppm. So, the 5000 now allow for quick catch up and then

> keep it in sync as well.

My log from the last few days:
Dec  5 14:31:08 darkone ntpd[12025]: ntp engine ready

Dec  5 14:31:29 darkone ntpd[12025]: peer 192.168.0.22 now valid

Dec  5 14:31:31 darkone ntpd[12025]: peer 192.168.0.21 now valid

Dec  5 14:32:20 darkone ntpd[8351]: adjusting local clock by 1.062893s

Dec  5 14:33:26 darkone ntpd[8351]: adjusting local clock by 0.969532s

Dec  5 14:36:47 darkone ntpd[8351]: adjusting local clock by 0.859755s

Dec  5 14:40:35 darkone ntpd[8351]: adjusting local clock by 0.736032s

Dec  5 14:41:05 darkone ntpd[8351]: adjusting local clock by 0.714376s

Dec  5 14:42:07 darkone ntpd[8351]: adjusting local clock by 0.680028s

Dec  5 14:43:41 darkone ntpd[8351]: adjusting local clock by 0.627119s

Dec  5 14:46:53 darkone ntpd[8351]: adjusting local clock by 0.517440s

Dec  5 14:49:32 darkone ntpd[8351]: adjusting local clock by 0.430305s

Dec  5 14:51:07 darkone ntpd[8351]: adjusting local clock by 0.374174s

Dec  5 14:54:51 darkone ntpd[8351]: adjusting local clock by 0.252689s

Dec  5 14:58:05 darkone ntpd[8351]: adjusting local clock by 0.148064s

Dec  5 15:00:13 darkone ntpd[8351]: adjusting local clock by 0.070505s

Dec  5 15:03:30 darkone ntpd[12025]: clock is now synced

Dec  5 15:24:14 darkone ntpd[8351]: adjusting clock frequency by

-57.519746 to -57.519746ppm

Dec  5 15:46:44 darkone ntpd[8351]: adjusting clock frequency by

-2.664841 to -60.184588ppm

Dec  5 16:09:54 darkone ntpd[8351]: adjusting clock frequency by

-0.514653 to -60.699241ppm

Dec  6 01:16:41 darkone ntpd[8351]: adjusting clock frequency by

0.747444 to -59.951797ppm

Dec  6 13:33:19 darkone ntpd[8351]: adjusting clock frequency by

-0.257248 to -60.209045ppm

Dec  6 22:45:05 darkone ntpd[8351]: adjusting clock frequency by

0.282931 to -59.926114ppm

Dec  7 09:34:22 darkone ntpd[8351]: adjusting clock frequency by

-0.233533 to -60.159647ppm

Dec  7 22:13:15 darkone ntpd[8351]: adjusting clock frequency by

0.152604 to -60.007043ppm

Dec  8 16:54:34 darkone ntpd[8351]: adjusting clock frequency by

-0.058112 to -60.065155ppm

Dec  9 08:55:18 darkone ntpd[8351]: adjusting clock frequency by

-0.041860 to -60.107015ppm

Dec  9 21:24:34 darkone ntpd[8351]: adjusting clock frequency by

0.070473 to -60.036542ppm

Dec 10 10:03:34 darkone ntpd[8351]: adjusting clock frequency by

-0.111900 to -60.148442ppm

Dec 11 00:54:45 darkone ntpd[8351]: adjusting clock frequency by

-0.002014 to -60.150456ppm
$ cat /var/db/ntpd.drift

-6.015046e-05
If the limit of what ntpd can correct is +/- 500 ppm, then my clock

isn't as bad as I thought, but it made problems anyway, see the link

above.
I think the clocks in some computers are really cheap, but are still

in use, so if Otto's diff has no negative side effects, a bit less

strict limits could be useful for us all.
Tas.