[ntp:questions] Re: NTP socket in use error message

Danny Mayer mayer at gis.net
Thu Jan 8 16:35:07 UTC 2004


"Mike Robinson" <BlackMagic at computer.org> wrote in message news:<mailman.22.1073551334.1757.questions at ntp.org>...
> I've got ntp-4.2.0 running successfully on more than a dozen hosts under Re=
> dhat 8.0.
> 
> Today I restarted ntpd on one of the hosts and got this message:=20
> service ntpd restart
> Shutting down ntpd:                                        [  OK  ]
> ntpd: Synchronizing with time server: Looking for host 132.163.4.102 and se=
> rvice ntp
> host found : time-B.timefreq.bldrdoc.gov
>                                                            [FAILED]
> The system log shows this message:
> ntpdate[13944]: the NTP socket is in use, exiting
> and /var/log/ntp.log shows that ntpd was running before I restarted it:
> 8 Jan 01:34:03 ntpd[30241]: synchronized to 132.163.4.102, stratum=3D1
> 8 Jan 15:11:00 ntpd[30241]: ntpd exiting on signal 15
> 
> Rebooting the machine doesn't fix the problem, and lsof -c ntpd produces no=
> thing. I've deleted ntp* in /usr/local/bin and recompiled the source and re=
> booted the system but ntpd just won't start.
> 
> I've searched the recent archives looking for references to sockets, but fo=
> und only one unrelated message.
> 
> Any ideas anyone?
> 
> MJR
> 

This looks like the script is running both ntpdate and ntpd. The error
message is for ntpdate. There is no reason to use ntpdate if you are running
ntpd. You can use the -g option with ntpd to achieve the same affect as
running ntpdate before running ntpd.

Danny



More information about the questions mailing list