[Mageia-dev] Why ntpdate still there?

Fri Feb 22 00:04:49 CET 2013

'Twas brillig, and Colin Guthrie at 11/02/13 14:55 did gyre and gimble:
+> 'Twas brillig, and Pierre Jarillon at 11/02/13 14:23 did gyre and gimble:
+>> Le lundi 11 février 2013 13:18:23, Colin Guthrie a écrit :
+>>> So ntpdate as a service is just a one-shot thing, it happens once at
+>>> boot to ensure the clocks are properly set and then ntpd takes over for
+>>> the rest of the time that machine stays up.
+>>>
+>>> As far as I'm aware, there is nothing integrated into crontab regarding
+>>> ntpdate, but please feel free to correct me on that one.
+>>
+>> Yes, this was the old system. I am not an expert in ntp, but I have read 
+>> carefully http://ntp.org few years ago and I return on it now and then.
+>> I can miss something... I am not the truth!
+>>
+>> On the web site http://www.ntp.org/  ->Implementation Documentation
+>> http://www.eecis.udel.edu/~mills/ntp/html/ntpdate.html said:
+>> Disclaimer: The functionality of this program is now available in the ntpd 
+>> program. See the -q command line option in the ntpd - Network Time Protocol 
+>> (NTP) daemon page. After a suitable period of mourning, the ntpdate program is 
+>> to be retired from this distribution
+>>
+>> And http://www.eecis.udel.edu/~mills/ntp/html/ntpd.html and the man tell us:
+>>
+>> -q
+>> Exit the ntpd just after the first time the clock is set. This behavior mimics 
+>> that of the ntpdate program, which is to be retired. The -g and -x options can 
+>> be used with this option. Note: The kernel time discipline is disabled with 
+>> this option.
+>> -g
+>> Normally, ntpd exits with a message to the system log if the offset exceeds the 
+>> panic threshold, which is 1000 s by default. This option allows the time to be 
+>> set to any value without restriction; however, this can happen only once. If 
+>> the threshold is exceeded after that, ntpd will exit with a message to the 
+>> system log. This option can be used with the -q and -x options. See the tinker 
+>> command for other options.
+>>
+>> The same options are set in the man of ntp.
+>>
+>> IMO, ntpdate should be replaced with: ntpd -gq
+> 
+> Yup that would be fine.
+> 
+>> I don't understand why ntpdate is listed as an active daemon in drakxservices.
+> 
+> This is simply because of the fact that drakxservices doesn't really
+> grok the kind of granularity you want here.
+> 
+> See the direct output from e.g. "systemctl status ntpdate.service" vs.
+> "systemctl status ntpd.service" (hint: compare "active (exited)" vs
+> "active (running)")
+> 
+> For the former, it's a "oneshot" and as it has been run and it ran
+> successfully, it's is thus considered "active".
+> 
+> Really we should teach drakxservices to present that properly: e.g. show
+> it as "Completed" or something, rather than "Active"
+> 
+>> In /etc/sysconfig/ntpd :
+>> - Mageia 1 : OPTIONS="-u ntp:ntp -p /var/run/ntpd.pid"
+>> - Mageia 2 : OPTIONS="-g"
+>> - Mageia 3 : OPTIONS="-g"
+>>
+>> Then it seems that ntpdate is no longer useful since Mageia 2.
+> 
+> Yes indeed. It seems the -g option passed there makes the separate
+> ntpdate service obsolete. It basically gives ntpd a one-chance option to
+> do a big jump, which is basically what we were achieving with that
+> double unit setup.
+> 
+> I'll kill off the ntpdate stuff.
+> 
+> Many thanks for poking into this :)
+
+Hmm, actually, I'm not sure the -g argument is sensible to pass to the
+daemon process generally. It seems that if it cannot reach a server
+(i.e. no networking) then the daemon exits.
+
+Certainly that is what I've seen here.
+
+Also, running ntpd -qg here with a large skew seems to not actually work
+here :s
+
+
+[root at jimmy ~]# systemctl stop ntpd.service
+[root at jimmy ~]# date
+Thu 21 Feb 18:02:19 GMT 2013
+[root at jimmy ~]# date
+Thu 21 Feb 18:02:23 GMT 2013
+[root at jimmy ~]# ntpd -qg
+ntpd: time slew +0.000000s
+[root at jimmy ~]# ntpdate pool.ntp.org
+21 Feb 23:02:53 ntpdate[7741]: step time server 149.5.113.103 offset
+18004.588317 sec
+[root at jimmy ~]# ntpd -qg
+ntpd: time slew +0.000000s
+[root at jimmy ~]# systemctl start ntpd.service
+
+
+So perhaps we should restore the previous setup?
+
+Col
+
+-- 
+
+Colin Guthrie
+colin(at)mageia.org
+http://colin.guthr.ie/
+
+Day Job:
+  Tribalogic Limited http://www.tribalogic.net/
+Open Source:
+  Mageia Contributor http://www.mageia.org/
+  PulseAudio Hacker http://www.pulseaudio.org/
+  Trac Hacker http://trac.edgewall.org/
+