Rob Janssen <[EMAIL PROTECTED]> wrote: > Tim Shoppa wrote: > > any sane ntp client would either > > ride these network glitches out or just switch to a reachable server, > > right? > > > As has been discussed, the ntp software we all love (ntpd, previously > known as xntpd) does not do this. > > When a server is configured with a DNS name (like pool.ntp.org) it > performs a DNS lookup exactly once at startup of the daemon, and uses > the IP address to contact the server. When this server fails > (immediately or later), it does not attempt to do the DNS lookup again, > so it remains stuck with the same nonfunctional server. > > The risk of this happening can be decreased a bit by using two or three > servers, but when they all fail there still is a problem. > Simple-NTP clients often do the lookup for every request so they don't > show this behaviour.
Rob - I was talking about minor network glitches, not a "server gone bad". When riding out minor network glitches, I don't see how a set of new glitchily-connected servers is any better than an old set of glitchily-connected servers. It is extremely likely that the glitches are following the client rather than the server (or that they're following a large subset of the servers to only a small subset of clients), anyway. Obviously dealing with servers gone bad is not the same thing, but most servers are up for months or years at a time, right? Don't try to make it sound like they're things that pop up and go away over a few hours in common use! Tim. > > Rob > _______________________________________________ > timekeepers mailing list > [email protected] > https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
