[ntp:questions] Regular spike_detect on syslog

Sean Austin Critica sean.critica at gmail.com
Sun Sep 16 12:15:03 UTC 2018


> I assume that yours is a Class A device

I think this particular hardware model is shipped as either A or B. In this case it is probably class A.

> Whereas these options do modulate the clock frequency, they do so with
> the knowledge of the operating system. The OS is therefore able to take

Just wondering if there is a reliable test to differentiate if this is a hardware/BIOS configuration issue or an NTP configuration issue?
Trying to toggle each BIOS setting on/off and observing would cause a lot of downtime.

> that into account in its timekeeping (for example by selecting a
> timekeeping clock source that remains constant notwithstanding Turbo
> Boost and all the rest of it).

Can I directly observe these sources and see which ones are stable (maybe by dumping them periodically, remotely from a machine with a known stable clock)? The OS in this case is RedHat EL 7.


More information about the questions mailing list