Found it, and it's a real doozy. Needless to say the gateway has at least two interfaces all with differing IPs. The client was resolving the hostname to one of the IPs but the daemon was replying with a different IP as the source, hence ntpdate didn't believe what it was getting back and believed that the server was not found.
On Mon, 27 May 2002, Howard Lowndes wrote: > I have an ntpd daemon running on my gateway which syncs to external clocks > OK, but when I try to do an ntpdate from an intranet workstation to that > daemon then it comes back with: > > no server suitable for synchronization found > > tcpdump shows the traffic between the ntp client and the ntp server as > being OK and there is no internal filtering going on. > > I would prefer not to have the intranet workstations syncing with external > clocks, I would prefer that they use my internal clock and only have that > syncing. > > One thing I have noticed is that the ntpd daemon running on the gateway > never gets above the minpoll frequency (minpoll 6 = 64 seconds), whereas > another, totally separate, daemon that I have running elsewhere is > consistently running at maxpoll (maxpoll 16 = 65536 seconds) > > Any clues out there? > > -- Howard. LANNet Computing Associates - Your Linux people Contact detail at http://www.lannetlinux.com "Life is like a hot bath - it's great when you're in it, but the longer you stay the more wrinkled you become." - Garfield -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
