can't get NTP server 'time

[lit]

When i config NTP server by SAM, there is a warning like this:

Connectivity is NOT configured for the host "10.243.1.193." Use the ???
Internet Connectivity" area to add connectivity to "10.243.1.193." ???
Continue entering the rest of the information on the screen if you are sure this is what you want to do. See the online help for more information.

after running NTP server ,if I use ntpq -p :

KKK:/#ntpq -p
remote refid st t when poll reach delay offset disp
==============================================================================
10.243.1.193 0.0.0.0 16 u 23 64 0 0.00 0.000 16000.0
10.243.1.197 0.0.0.0 16 u 22 64 0 0.00 0.000 16000.0


It looks that i can't get the ntp server's time.

Who can give me a hand? Thanks in advance.

br,litian

[Pete Randall]

Litian,

Is your server configured to access the "outside world"? Default routes in place and all that stuff? DNS?


Pete

[Jeff Schussele]

Hi,

Do you have connectivity to that IP?
Try a traceoute to it.
May be a firewall issue or you may need a static route to that host.
Remember that if this goes through a firewall you'll need port 123/UDP open in both directions to reach it with NTP requests.

Rgds,
Jeff

[Elmar P. Kolkman]

What you can do is:
1) try to ping the ntp host you are trying to use, in your case 10.243.1.193 and 197.
2) ntpq -p 10.243.1.193 or ntpq -p 10.243.1.197.
3) find out if those hosts are not by any chance using ntp keys (check out their ntp.conf)
4) check out log files after starting xntpd with debugging on (/var/adm/syslog/syslog.log on server and client).

Hope this helps you on your way to solve the issue.

[lit]

anyway, i can ping the 10.243.1.193,
and there is other host's ntp status like this:
kk1:/#ntpq -p
remote refid st t when poll reach delay offset disp
==============================================================================
10.243.1.193 211.136.70.97 16 u 24 64 0 2.44 51.478 16000.0
10.243.1.197 211.136.70.97 16 u 56 64 0 2.50 64.027 16000.0
,

and i had checked the firewall's policy:
ntp is allowed pass.

how about ur think? the ntp server need add
this new host ip ?

br,
litian

[Jeff Schussele]

Well, if *both* the client & server are in DNS then there's no need to add the client to the server's hosts file.

Try to telnet to the server on port 123

telnet 10.243.1.193 123

Jeff

[lit]

Hi,Jeff
The new host need added to DNS ?bcs i ntp it just by ip address.

br,litian

[jerry1]

Try stopping and starting /sbin/init.d/xntpd.
Check output messages from command line
and in /var/adm/syslog.log unless other logfile is specified. Look for xntpd in
log file.

Check your config files:

/etc/rc.config.d/netdaemons
/etc/ntp.conf
/etc/ntp.drift - optional performance file.

Check for other files also.
/etc/ntp.keys
/etc/ntp.conf.example

Let me know what you find.

[lit]

Hi, Jery
Tks, I will check it net Monday.

br,litian

[lit]

today if i use ntpdate -d 10.243.1.193,following result:

/var/adm#ntpdate -d 10.243.1.193
transmit(10.243.1.193)
receive(10.243.1.193)
transmit(10.243.1.193)
receive(10.243.1.193)
transmit(10.243.1.193)
receive(10.243.1.193)
transmit(10.243.1.193)
receive(10.243.1.193)
transmit(10.243.1.193)
server 10.243.1.193, port 123
stratum 16, precision -18, leap 00, trust 000
refid [211.136.70.97], delay 0.02763, dispersion 0.00002
transmitted 4, in filter 4
reference time: c2d3c786.e0901000 Fri, Aug 1 2003 0:51:18.877
originate timestamp: c390e8e4.deb04000 Mon, Dec 22 2003 11:52:04.869
transmit timestamp: c390e8d9.2c8e0000 Mon, Dec 22 2003 11:51:53.174
filter delay: 0.02809 0.02765 0.02763 0.02769
0.00000 0.00000 0.00000 0.00000
filter offset: 11.69494 11.69479 11.69478 11.69475
0.000000 0.000000 0.000000 0.000000
delay 0.02763, dispersion 0.00002
offset 11.694781

22 Dec 11:51:53 ntpdate[7424]: no server suitable for synchronization found

ntptrace command:
kkk:/#ntptrace 10.243.1.193
10.243.1.193: stratum 16, offset 11.672237, synch distance 142.07112
211.136.70.97: *Timeout*

so, here i have a question:if the reference host can't work normal, can i get the time from the ntp server which ntp.conf
point to?

[Bryan D. Quinn]

Hello,

What is 211.136.70.97, is this a router or something?

Also, is this the only client that is synching to your ntp server or do you have another?

If this is the only client, can you provide the ntp.conf file for this client and the server?

-Bryan

[lit]

hi,Bryan


211.136.70.97 is a ntp server, and now it can't work. my question is why my host can't synchronize with 10.243.1.193?


br,litian

[Bryan D. Quinn]

Hello,

If your host can ping 10.243.1.193 and you make the neccessary changes to the /etc/ntp.conf file for both your host and the 10.243.1.193 machine, you should be able to synchronize. Here is what you need to do:

1. Make 10.243.1.193 your NTP server. Your /etc/ntp.conf file on that server should look something like this:

server 127.127.1.1
fudge 127.127.1.1 stratum 8

2. Modify you /etc/ntp.conf file on your host to look like this:

server 10.243.1.193
driftfile /var/adm/ntp.drift

NOTE the driftfile is optional.

3. Stop and start NTP:

/sbin/init.d/xntpd stop
/sbin/init.d/xntpd start

Please note that these boxes need to be fairly close in time. If the time difference between them is too great, the host will not synch to the NTP server's time.

Hope this helps!
-Bryan

[Bryan D. Quinn]

Sorry, I did not clarify something. If you modify the server and client(host)'s ntp.conf file, make sure you stop and start xntpd on both machines.

Also, you need to make sure you have the following in your /etc/rc.config.d/netdaemons file:

export XNTPD=1

on both the host(NTP client) and the NTP server.

-Bryan

[lit]

hi,Byran
I had done u say many times except the server settings which belong to other Dept. and had run long time. now the outer ntp server is not ok and i want create new ntp client .
why my computer not syn. with 10.243.1.193?

br,tianlit

[Bryan D. Quinn]

Hello,

Are you only using SAM to try to configure NTP? If so, I would recommend manually trying to configure it.

-Bryan

[Bill Hassell]

NTP is pretty simple to get configured. You just need an entry in /etc/ntp.conf as in:

server tick.jrc.us

(or an IP address if DNS doesn't resolve the hostname). Then use ntpq to talk to the server:

ntpq -p tick.jrc.us

and you should see how the NTP is configured. If this fails, your firewall is blocking port 123 or the server is no longer providing NTP services. ping lets you know that there is a computer there but ntpq verifies that NTP services are available. No need to go any farther in configuring NTP until ntpq -p starts working. You'll need your network folks to trace the failure.

[Bill Thorsteinson]

Try runnign the ntpdate command with the debug option (-d) and you will
get a fair bit of information which will
indicate whether the server is resposnding.
Try
ntpdate -d localhost 10.243.1.193 10.243.1.197

There are security options that the other
department may set that can block access
from other servers. They may have done
a security review and locked down there servers.
Also as noted above an interveening firewall
may be blocking 123.
You may try using your DNS servers as your
NTP servers as they usually run NTP.