I've got a machine running iptables and xntpd. For some reason, ntp quit
working after a power failure. The ntp.conf file looks fine, just lists a
few public servers. Here's what happens, though.
1) If I start xntpd (via the rc scripts), it will spawn two processes--
one of them the parent of the other. Both processes hang there forever.
2) If I put a local/loopback entry as the only 'server' (127.127.1.0), then
it starts up as an undisciplined clock. Only one process is running at
3) If I strace the process (with either config file), I see a process start
up, and then fork off a second process. It seems then that the single
process I see in the second case is a forked process.
4) Now here's the weirdest part: When I 'strace -f' the startup with the
normal ntp.conf file (i.e. the one that has external servers in it),
then my entire session locks up! Furthermore, I can no longer ssh into
the box, necessitating a hard power-cycle! HOWEVER, iptables continues
to work fine, passing packets back and forth exactly like it's supposed
Does anyone know what's going on? It's possible (but rather unlikely) that
this machine has been hacked, but I can't see any exploits that would
indicate this problem.
Oh, and as far as preliminary diagnostics have gone, I checked out all of
the servers listed. I can 'ntpdate' from all of them successfully.