Tim Kientzle wrote:
> M. Warner Losh wrote:
>> In message: <B3BCAF4246A8A84983A80DAB50FE72424C6970@secnap2.secnap.com>
>> "Michael Scheidell" <scheidell@secnap.net> writes:
>> : > I presume the servers are all stable (ie not stepping) and : >
>> have a reasonably low delay. If so, I suspect your ntpd PLL : > has
>> locked up. I've seen problems with some versions of ntpd : : 20
>> different machines?
>>
>> That would strongly imply a poor choice of upstream server.
>
> It might also help to set up 1 machine as
> your NTP master. Have it sync to the upstream
> servers and have the rest of your machines
> sync to it. This won't solve the problem,
> but might make it easier to solve (once your
> NTP master is stable, the rest should follow).
>
Except that all 20 are on different physical networks, one in a
different content.
I went back and looked at out 4.11 systems, and didn't see any 'time
reset' errors.
Internally, here at this physical location, we do have one master ntp
server, pulling from (3) servers, not two.
All internal systems then pull from this one master.
> You can experiment with different upstream
> servers, different network settings, different
> NTP settings, etc.
>
only ntp setting that helped was the -x option (adjust once, slew 128m
max each other time)
One external box is now 30 seconds out of sync (after 24 hours)
Again, ntp problem is one thing, but what about the nanosleep() issue?
5.4 worked fine, 5.5 has problems.
Did something change in 5.5 with clock() or nanosleep()?
Its not documented in the readme for 5.5.
--
Michael Scheidell, CTO
SECNAP Network Security / www.secnap.com
scheidell@secnap.net / 1+561-999-5000, x 1131