On Thu, 2003-10-23 at 11:14, Genti Hila wrote: > Hello, > > I have an Linux Box that acts as an NTP (network time protocol) server and i would like to sent the critical logs, especially if the ntpd daemon stops. > > Can anybody give some directions on that ? > > Any idea is appreciated > > Genti this is the only issue I have with ntpd: If the daemon stops or in one case I had a smp machine that would skew the time faster than ntp would keepup with it so I created a script taht would check all the machines listed in a file and do some ugly hacks to figure out if the box was synced or not. I run it via cron periodically: I will put the script at http://www.elevating.com/bret/checktime.pl browser readable : http://www.elevating.com/bret/checktime.pl.txt if you are interested. I have been playing with a better way to do this using ntpdc. For instance ntpdc -c sysinfo |awk '/stratum/ {print $2}' will print the stratum that the peer of the machine is running at. On my machines if it is synced it usually prints 2 or 3 depending on the server it is synced to. If a box is not synced it prints 11 since that is the stratum I set the local fudge to. At least I think that is right. HTH Bret -- redhat-list mailing list unsubscribe mailto:redhat-list-request@xxxxxxxxxx?subject=unsubscribe https://www.redhat.com/mailman/listinfo/redhat-list
