NTP

[Daniel Fourie]

I get the following error when I start ntp:

Jun 9 09:19:43 nnoc-f4 xntpd[23347]: tickadj = 625, tick = 10000, tvu_maxslew = 61875
Jun 9 09:19:43 nnoc-f4 xntpd[23347]: precision = 27 usec
Jun 9 09:19:43 nnoc-f4 xntpd[23347]: bind() fd 11, family 2, port 123, addr c40150ff, in_classd=0 flags=0 fails: Can't assign requested
address


Please help!!!!

[Robert-Jan Goossens · Accepted solution]

Hi,

Are you running 10.20 ?

Robert-Jan.

[Daniel Fourie]

No, this is HP-UX 11i in a mc-service guard cluster

[Robert-Jan Goossens]

Hi,

What is the time diff between your server and the time server ?

Robert-Jan.

[Daniel Fourie]

The diff between the two is
(offset -0.000704 sec)

[Robert-Jan Goossens]

That's oke,

Check if xntp is still running,

# ps -ef | grep ntp
kill the deamon and try again.

There is a patch for 11.11
JAGac20260 / SR 8606113990:
Description: "/sbin/init.d/xntpd stop" command does not terminate xntpd daemon.

http://www5.itrc.hp.com/service/patch/patchDetail.do?patchid=PHNE_27223&context=hpux:800:11:11

Robert-Jan.

[Massimo Bianchi]

Hi,
can you ping correctly the ntp server ? It looks like yoyr sever is no longer reachable.

Massimo

[Robert-Jan Goossens]

Correction to the patch I gave you,

http://www5.itrc.hp.com/service/cki/patchDocDisplay.do?patchId=PHNE_24512

This is the latest.

[Daniel Fourie]

I have the patch, but still the same problem

[Robert-Jan Goossens]

Just a wildshot, check your /etc/services file if you got port 123 enabled.

ntp 123/udp #ntp Network Time Protocol

[Massimo Bianchi]

Can you please post your /etc/ntpconf ?
May be there is a mis-typing.

Massimo

[Bill Hassell]

Run ntpq -p to see the status of all your NTP servers (you do have several, right?) Then run ntpq -p for each of the NTP servers to see if they respond. And make sure that nothing else is using port 123 by looking for multiple xntpd processes as in:

UNIX95= ps -fC xntpd

and also check the port after you shutdown xntpd as in:

/sbin/init.d/xntpd stop
lsof -i :ntp

If you don't have lsof (get a copy from http://hpux.connect.org.uk/) then use netstat -a | grep ntp and look for any processes that are holding the port open. This could be due to using kill -9 in the past.