LinuxSA Mailing list archives

Index: [thread] [date] [subject] [author]
  From: Alan Kennington <akenning@dog.topology.org>
  To  : newton@atdot.dotat.org
  Date: Thu, 2 Sep 1999 10:05:51 +0930

Re: SuSE

Topic: NTP functionality on linux boxes.

===================================================================
Mark,

To confirm what you said about xntpd replacing rdate, I've just had
the response back from the SuSE support people (in the US).

------------------------------------------------------------------
SuSE comes with `ntpdate' (package `xntp') that will allow you to query
NTP-timeservers in the Internet. This is considered to be the standard
solution for time synchronization.
------------------------------------------------------------------

I knew about NTP before, and I've read up on the RFC(s) for this,
but I had thought of this as a protocol for big organisational
computers rather than PCs on desktops.
Obviously once again, I'm being dragged into the modern world.

It seems credible then, that redhat are supporting "rdate"
for backwards compatibility or something.
That's what used to burden (and still greatly burdens)
the likes of microsoft - the need to keep stone-age users happy.
(The fact that the first ISA cards still work on modern
mainboards is a similar phenomenon.)

This makes me even more happy that I've made the change out of RH.

But now how do I find a suitable NTP client out there in the
real world to sync to?
With "rdate", I was running a cron job once every 24 hours
(or 6 hours for my least accurate PC) out to augean with:

-------------------------------------------------------------------
25 2,8,14,20 * * * /usr/bin/rdate -s augean.eleceng.adelaide.edu.au
-------------------------------------------------------------------

So now:

1.	Where do I get a list of the accurate NTP servers
	with which I can sync?
	[I'm aware of the list at
	http://www.eecis.udel.edu/~mills/ntp/servers.htm
	which gives primary and secondary NTP server lists.
	But I suspect these lists may be incomplete.]
	(Since my bigpond link emerges into the real world
	in Melbourne somewhere, I guess I need to use a
	Melbourne machine. traceroute gives:

--------------------------------------------------------------------
[akenning@dog]$ traceroute augean.eleceng.adelaide.edu.au
traceroute: Warning: Multiple interfaces found; using 203.38.148.51 @ eth0
traceroute to augean.eleceng.adelaide.edu.au (129.127.28.4), 30 hops max, 40 byt
e packets
 1  gw.way7.Adelaide.telstra.net (139.130.140.1)  238.774 ms  131.105 ms  118.21
9 ms
 2  Ethernet1-0.way1.Adelaide.telstra.net (139.130.237.65)  127.082 ms  141.073 
ms  118.189 ms
 3  Fddi0-0.way-core2.Adelaide.telstra.net (139.130.237.229)  127.023 ms  121.41
1 ms  128.133 ms
 4  Atm1-0-1.lon-core1.Melbourne.telstra.net (203.50.6.5)  197.034 ms  189.798 m
s  358.207 ms
 5  FastEthernet0-0-0.lon14.Melbourne.telstra.net (139.130.239.235)  367.103 ms 
 201.083 ms  198.241 ms
 6  optvm.lnk.telstra.net (139.130.55.2)  197.039 ms  201.397 ms  188.125 ms
 7  atm2-0-19.mb1.optus.net.au (192.65.89.138)  187.786 ms  191.863 ms  188.110 
ms
 8  aarnet-sa.mb1.optus.net.au (192.65.88.226)  367.084 ms  371.820 ms  378.066 
ms
 9  lis255.atm2-0.ntce.adelaide.gw.saard.net (203.21.37.3)  277.225 ms  211.627 
ms  248.063 ms
10  * * *
11  augean.eleceng.adelaide.edu.au (129.127.28.4)  348.718 ms  361.561 ms *
------------------------------------------------------------------------------

	I.e. it goes out over some pretty wierd looking subnets
	to Melbourne and back.
	I realise that the difference in accuracy is purely
	academic in my case!)

2.	Do you need to get some sort of permission and/or
	configuration to do an NTP sync to someone's
	NTP host? 

Cheers,
Alan Kennington.

=======================================================================
PS. Traceroute chose the wrong interface eth0!

PPS. I've just answered some of my query.
Running "ntpdate" on my new machine correctly modifies the
date like "rdate", and also gives this output:

-----------------------------------------------------------------
emu /var/spool# ntpdate augean.eleceng.adelaide.edu.au
 2 Sep 09:55:38 ntpdate[31306]: step time server 129.127.28.4 offset 51.116448 sec
-----------------------------------------------------------------

This means I can stay back in the stone age by just
running "ntpdate" once every 24 hours on all 4 machines.
But my questions above refer to the _real_ NTP.
I.e. the 24 hr/day daemon which keeps permanent sync going.

-- 
Check out the LinuxSA web pages at http://www.linuxsa.org.au/
To unsubscribe from the LinuxSA list:
  mail linuxsa-request@linuxsa.org.au with "unsubscribe" as the subject


Index: [thread] [date] [subject] [author]
Return to the LinuxSA Mailing List Information Page