connection refused not understood (ntp)

I'm setting up a new RH9 router (firewall, NTP server, etc).
Initially, I'm trying to get ntp working, but I'm getting
"connection refused" messages from my existing RH6.1 router.
The old one is already running ntpd, and I'm pretty darn
sure other machines can connect to it, but the new one
can't. I think I know how to fix whatever is broken, but
I guess I need a list of things to check. Here's what I
know:
Nothing is getting logged by ipchains (local net rules open)
hosts.deny and hosts.allow open to local, closed to outside.
No firewall on new machine.
NTP is running on old one (netstat shows udp:123)
local machines being NAT'd
ssh, rlogin, telnet, ftp work between the two.
New machine can surf the 'net
Seems like a port specific problem

I know this is really basic, but what else needs to be checked?

Thx, Chris

--
Chris Richmond | I don't speak for Intel & vise versa

Chris Richmond - MD6-FDC ~ <(E-Mail Removed)> wrote:
> NTP is running on old one (netstat shows udp:123)
> local machines being NAT'd
> ssh, rlogin, telnet, ftp work between the two.

If address/port translation is indeed involved (in communication
of NTP client with NTP server), then you should make sure that
UDP#123 is also translated. Other services you've mentioned are
all TCP-based.

--
andrei