Re: Servers reachable, but can't sync

brad@xxxxxxxxxxxxxxxxxxx (Brad Knowles) · Thu, 8 Sep 2005 16:35:19 GMT

At 9:07 AM -0700 2005-09-08, Greg McCann wrote:
 I am stumped.  The communications seem to be perfect, and the -q
 mode functions flawlessly.  Yet in -b mode, after communication
 with the server is complete, ntp decides for unexplained reasons
 that it was "not suitable".
 I don't know what else I can do to troubleshoot this without going
 into the ntp source code, which I am not really qualified to do.

	Reading the official documentation for ntpdate at 
<http://www.eecis.udel.edu/~mills/ntp/html/ntpdate.html>, I see that 
using the "-b" option is supposed to force the system to always step 
the time with the settimeofday() system call, even in those 
circumstances when it would normally just slew the clock with the 
adjtime() system call, such as would happen if your offset was less 
than +/- 0.5 seconds.
	Checking the description for the "-q" option, I see that it is 
used for query only, and doesn't actually set the system clock.

	Are you sure you're running these commands as root?  If you're 
not root, you're not likely to have the privileges required to 
perform the necessary system calls to do the things you want.
	Or, maybe you're running on something like SELinux, and even 
though you're root, the commands are not being given the necessary 
privileges to actually set the system clock.  In that case, you might 
need to turn off SELinux, or disable the control that SELinux has 
over the necessary system calls.

	Moreover, it doesn't seem to be necessary to use the "-b" option, 
since ntpdate would go ahead and set the system clock with the 
settimeofday() system call if you're more than +/- 0.5 seconds off, 
and slewing the system clock with the adjtime() system call for 
offsets of less than +/- 0.5 seconds doesn't seem to be such a bad 
thing.

	What happens when you try to run a real "ntpd -gq" instead of 
ntpdate?  Can you show us what the tcpdump looks like?  Can you run 
the programs with the "-d" option, so that we can see some more 
debugging output, to get a better idea of what's really going on?
	When running ntpdate, have you made sure that you don't already 
have a copy of ntpd running?  If so, then you're either going to want 
to leave ntpd alone and not run ntpdate, or you should stop ntpd 
before running ntpdate.
	Speaking of which, have you looked in your log files to see if 
there are any entries from ntpdate or ntpd that are complaining about 
being unable to set the clock?  If not, are you sure you've got your 
/etc/syslog.conf file set up correctly so that you would see these 
complaints if they were being logged?

	There's lots more questions that I could ask, but these just 
occurred to me off-the-top-of-my-head.
--
Brad Knowles, <brad@xxxxxxxxxxxxxxxxxxx>
"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."
    -- Benjamin Franklin (1706-1790), reply of the Pennsylvania
    Assembly to the Governor, November 11, 1755
  SAGE member since 1995.  See <http://www.sage.org/> for more info.
_______________________________________________
questions mailing list
questions@xxxxxxxxxxxxxxxxx
https://lists.ntp.isc.org/mailman/listinfo/questions
.

Re: Servers reachable, but can't sync

Relevant Pages