I have been trying to chase down and eliminate this error for a while now.
The cvm leader only wants to use itself as the time source - when I run this command (allssh ntpq -pn) it shows
remote refid st t when poll reach delay offset jitter
x.x.x.x y.y.y.y 2 u 603 1024 377 0.337 55090.5 419.722
x.x.x.x 126.96.36.199 3 u 200 1024 4 0.358 75064.8 0.000
*127.127.1.0 .LOCL. 10 l 587 1024 377 0.000 0.000 0.000
(The asterisk indicates that it is using itself as a time source, is that correct?)
I can run ntpdate successfully to the configured ntp servers and the cvm can connect to them. (I can even run ntpdate to an external time server that is not configured for that matter)
How do I get the cvm leader to use the configured ntp servers as the source (and not use itself)?
Best answer by maazView original
@Mike2.0, can you please open a case with Nutanix Support as we will need to look at this in detail.
Thanks for looking at this.
I’ve been hacking away (maybe wrong choice of words) at this and finally made headway today.
We had some issues with our local NTP servers (high values for offset and jitter), and that probably had some kind of ripple effect that caused the problem.
Anyway, all good now.