Date: Fri, 14 Mar 1997 10:17:56 -0800 (PST) From: mark thompson <thompson@tgsoft.com> To: freebsd-questions@freebsd.com Subject: ntpdate Message-ID: <199703141817.KAA06916@squirrel.tgsoft.com>
next in thread | raw e-mail | index | archive | help
I have ntpdate running out of cron every couple of hours. Each time it runs, it adjusts the clock back about .4 seconds... this would be about 5 seconds a day or 2 minutes a month. Mar 14 02:44:33 squirrel ntpdate[6515]: adjust time server 204.123.2.5 offset -0.382174 Mar 14 04:44:02 squirrel ntpdate[6617]: adjust time server 204.123.2.5 offset -0.396060 Mar 14 06:44:02 squirrel ntpdate[6715]: adjust time server 16.1.0.4 offset -0.382739 I am quite willing to believe that the clock in my PC gains 2 minutes a month. What puzzles me is that the comments in ntpdate seem to suggest that *it* believes that the system will compensate the hardware clock to make this error disappear, kinda like those self-adjusting car clocks. Well, that obviously doesn't happen. I looked at the code in the kernel, and have to sadly admit that it baffles me. What's the truth? Is the system clock adjustable, or does ntpdate just slew it into correctness every time it is run? -mark
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199703141817.KAA06916>