On Mon, Jun 10, 2002 at 10:24:28AM -0700, Kevin Buhr wrote: > Damien <[EMAIL PROTECTED]> writes: > > > > The problem must have occurred often enuff to other people, IMHO, but I > > can't find the solution online. My 'wall-clock' (as xscreensaver calls it > > in the error messages) keeps jumping ahead -- then back, semi-randomly. It > > seems to do so by always the same amount(?): @ 1 hr 11 minutes. > > I assume you aren't running an NTP daemon and that there are no > messages in the logs about the time being stepped forward or back. > > Does "@" mean "approximately"? Is it closer to one hour, eight > minutes, and 16 seconds? That would be 4096 seconds, and a most > suspicious number of seconds to be jumping. > > Does it jump back and forth between two values, or does it jump > several times in the same direction (so it quickly becomes many hours > off)? > > If it's jumping continually back and forth by 4096 seconds, I'd guess > you have a bad SIMM. One bit (which just happens to be where your > kernel is storing the time-of-day clock) isn't being reliably set, and > you see jumps forward and backward every 20 or 30 times (seconds) the > kernel bumps the seconds counter. > > If this is the case, you might try swapping SIMMs around. However, > I'd suggest labelling their original positions very carefully. If the > problem does "disappear", you want to be able to get back to where you > can reliably reproduce it and eliminate the offending SIMM.
I think it's possibly a dodgy RTC on the motherboard. I saw this exact complaint come up on linux-kernel a while ago, and someone mentioned that it was specific to a certain brand of mobo. No URLs for you, I'm afraid, but if you google a bit, I'm sure it'll appear. As I recall, there was no solution or workaround - the board simply reports the wrong time every so often. It's a bug. jc -- It may stop, it may not. And stop calling me "dj". -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
