[BBLISA] ntp question

[John P. Rouillard]

In message <001301c9a746$ae0d5b30$0a281190$@com>,
Edward Ned Harvey writes:

>> of systems.  One problem I've encountered is that many of our "systems"
>> are actually VMWare based virtual machines and VMWare seems to have a
>> borderline ridiculously bad clock drift.  From what I've read on the
>
>Many of my systems are also virtual, but perhaps the usage pattern is making
>our clocks behave differently from each other.  
>
>IMHO, even if you cron'd the ntpdate command for every 5 minutes, or every
>minute, that would still incur less network traffic than ntpd (if you're
>worried about ntp traffic clogging your network.)

Actually not if the clock is able to keep time. Expect one ntp sync to
each higher staratum server every 1024 seconds (17 minutes) on any
system where ntpd is working.

However if the host clock is so crappy that changing the tick interval
in unable to keep it stable, ntpd can't correct that. It expects it
oscillators to be stable (not accurate, but relatively stable
(i.e. reproducable)), if they aren't then there is no way to maintain
a monotonically increasing time without jumps.

(Well, changing some of the rules of physics about how time works
would help but...)

--
				-- rouilj
John Rouillard
===========================================================================
My employers don't acknowledge my existence much less my opinions.