Re: Minor NTP question ....

[mlelstv%serpens.de@localhost (Michael van Elst)]

wam%hiwaay.net@localhost ("William A. Mahaffey III") writes:

>[wam@fly, ~/Suspect, 4:58:50pm] 334 % ntpq -pn
>      remote           refid      st t when poll reach   delay offset  
>jitter
>==============================================================================
>-216.180.99.1    128.59.59.177    3 u  947 1024  377    0.340 10.937   0.434
>-216.180.122.1   198.82.247.51    3 u  358 1024  377    2.706 12.187   1.078
>+208.75.88.4     69.36.224.15     2 u  384 1024  377   79.207 -5.931   0.584
>+98.143.24.53    64.250.229.100   2 u  472 1024  377   47.575 6.415   0.006
>*204.2.134.163   187.253.153.32   2 u  969 1024  377   77.611 2.938   0.155
>[wam@fly, ~/Suspect, 4:58:53pm] 335 % cat /var/lib/ntp/drift
>-28.010
>[wam@fly, ~/Suspect, DING!] 336 %

>& we have a winner !!!! That -28-ish isn't so bad, but with so much 
>difference between its 3 external sources,

Hmm,

the offset values are in milliseconds, your external sources are pretty
close.

I was refering to the drift value, it says that your system clock
has a deviation of 28ppm (about 2.4 seconds per day). But that's
an average, the real deviation could be different, in particular,
when the system is busy or if it is booting.
The RPI has no clock chip, every reboot resets the clock to the
timestamp of the root filesystem.

-- 
-- 
                                Michael van Elst
Internet: mlelstv%serpens.de@localhost
                                "A potential Snark may lurk in every tree."