I asked: > >So the question is why can't the z boxes use ntp to stay accurate?
On Thu, Oct 27, 2005 at 01:26:19AM -0500, Bruce Hewson ([EMAIL PROTECTED]) wrote: > I always understood it was because the z box is smart and doesnt allow it's > time to go backwards. Which is why IBM charge so much for the sysplex > timers. > > And the "other" boxes dont care if the time jumps backwards or forwards, > but I imagine that there could be problems if they had to forward recover a > database after the box time went backwards. :-) I didn't have DST in mind when I asked "why not ntp?" My question was in response to a gripe about the mainframe clock drift. I'm sure that somebody at IBM could come up with a way to use ntp to adjust the clock speed to keep the clock accurate without requiring any backward adjustments. I don't really know anything about the sysplex timer, but doesn't it do essentially that? Also, I have a really vague recollection of some (optional?) setup on some *ix systems for ntp to handle the hour "fall back" by running the clock slow for a few hours. Seems to me that I was once logged on during that period, when the clock value has no relationship whatsoever to any reality. (I found out later about the ntp gimmick.) However, one point is that I was logged and able to work. Another point is that they didn't have to take the system down to deal with the jump backwards. Does anybody reboot their PC due to DST? /Leonard ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
