=?iso-8859-1?Q?Re:_[ntp:questions]_Advice_appreciated_on_ntp.conf_files, _etc?=

cave.dnb@tiscali.fr cave.dnb at tiscali.fr
Thu Dec 8 21:03:54 UTC 2005

 Apologies for having to resort to webmail again. Brad Knowles has been trying to unblacklist servers being blocked by SORBS, 
so that I can send to the questions list, but SORBS are apparently being very slow in accomodating his request. Personally, I'm 
getting a bit T'd off now. I'm not a spammer, and having to use webmail is an inconvenience, but I'm coming around to the 
conclusion that it might be better for Brad to keep these servers blacklisted for the sake of keeping , or trying to keep the 
questions list spam-free. I'll just put up with using webmail for my odd postings, and waiting for them to receive approval from 
the moderator. Quite honestly I think the Internet is going down the tubes, because of spammers scammers, and all the other 
scum that are trying to make a quick buck at everyone ele's expense. And I also apologise for the ads which I have no control 
On Sunday 04 December 2005 04:11, Danny Mayer wrote: 
> cave.dnb at tiscali.fr wrote: 
> > Hi. At last. Having read multiple times the docs, and viewed all the 
> > postings to the list over the last 2 weeks, I am once again attempting to 
> > post to the list, this time by webmail, as SORBS have DNS servers blocked 
> > te the mailing list.I'd like some help in understanding the various 
> > ntf.conf default entries for 
> > different distros, and have attached the files for ntpq-output, 
> > and /etc/ntp,conf files for the different distro's below. 
> Get rid of all your restrict lines and see if that works. It's dangerous 
> to start with restrict lines unless you know exactly how to configure 
> them. Have you read http://ntp.isc.org/bin/view/Support/ConfiguringNTP 
> and particularly section 6.4? 
> Danny 
Hi Danny. Sorry for the delay in replying. I've been waiting on Brad Knowles  
to make it possible for me to post to list. 
I commented out all restrict lines, read the doc you mentioned, also the  
ntpq.html doc again, and the doc "NTP Debugging Techniques" which answered a  
lot of the stuff I'd asked you.  
All seems to be working ok now. The clocks on the server and client are nicely  
in synch, but as the ntp daemon is running continually on the server, and  
port 123/UDP is in use, I'm no longer getting updates to the time on the  
Smoothwall Express2 firewall. This isn't a big problem as Smoothwall  
Express3, now in version Alpha, has the capabilities to allow clients to  
retrieve time from the Smoothwall. 
Just 2 queries I have. The Server I have running at the moment is FC2. When I  
first set this up with 3 Internet time servers, the refid on ntpq> pe showed  
the IP addresses that were on the timeserver website. Now they have changed,  
but the FC2 client is still showing a refid for the original IP address of  
whichever Internet time server is running as a sys.peer. Again, the time is  
in synch, so there is clearly no problem, but I'd be interested in knowing  
why this is so. ntpq> pe & as outputs below. 
FC2-serverntpq> pe 
     remote           refid      st t when poll reach   delay   offset  jitter 
+lptfpc46.obspm.   2 u  613 1024   37  212.045   29.814  23.999 
*ntp.kamino.fr   2 u  741 1024   37  311.487   31.968   7.659 
+ntp2.belbone.be    2 u  696 1024   37  268.909   97.301  45.988 
ntpq> as 
ind assID status  conf reach auth condition  last_event cnt 
  1   596  b474   yes   yes  none  candidat   reachable  7 
  2   597  b674   yes   yes  none  sys.peer   reachable  7 
  3   598  b474   yes   yes  none  candidat   reachable  7 
ntpq> pe 
     remote           refid      st t when poll reach   delay   offset  jitter 
*     3 u  504  512  377    0.390    1.136  12.721   .INIT.          16 u    - 1024    0    0.000    0.000 4000.00 
ntpq> as 
ind assID status  conf reach auth condition  last_event cnt 
  1 15380  9614   yes   yes  none  sys.peer   reachable  1 
  2 15381  8000   yes   yes  none    reject 
The remote address is for FC1, which I run as an alternative  
timeserver on the same machine as the FC2 server is running. 
I have now reconfigured the /etc/ntp,conf on the FC1 server, removing the  
original default stratum 1 server (clock1.redhat.c) and replacing it with the  
same 3 Internet timeservers that I use on the FC2 server. 
My second. also unimportant query is. The FC2 servers refid shows IPaddresses,  
but the FC1 servers refid shows alternative hostnames. Why might this be?  
Again the FC2 client accessing this server is showing the original IP  
addresses as refid. 
FC2 server is using ntp-4.2.0-7 
FC! server is using ntp-4.1.2-5 
Lastly. When FC1 server was using clock1.redhat.c as a stratum 1 timeserver,  
the refid showed a kiss code .CDMC.  This one isn't on your kiss code list.  
Do you know, just out of interest what this denotes? 
Many thanks for NTP. This is the first time I've had all my distro's clocks in  
sync.  Nigel Henry. 

-------------- ALICE HAUT DEBIT : TRIPLE PLAY A 29,95 EUR/MOIS -------------- 
Découvrez vite ALICEBOX : avec le modem WIFI, profitez de l'ADSL, de la TELEPHONIE et en exclusivité de la TELEVISION ! 
Bénéficiez aussi de la hotline gratuite 24h/24 ! Soumis à conditions. Pour en profiter cliquez ici http://abonnement.aliceadsl.fr

More information about the questions mailing list