1 |
On 18/03/16 05:14, Alan McKinnon wrote: |
2 |
> On 17/03/2016 22:02, Håkon Alstadheim wrote: |
3 |
>> On 03/17/2016 02:03 PM, Bill Kenworthy wrote: |
4 |
>>> On 17/03/16 20:26, Alan McKinnon wrote: |
5 |
>>>> On 17/03/2016 08:50, Håkon Alstadheim wrote: |
6 |
>>>>> I have a server SUPPOSED to be running 24/7, but every once in a while |
7 |
>>>>> during a prolonged absence the box will go down. The Real Time Clock |
8 |
>>>>> will drift, and in the rush to get the box up again I let everything |
9 |
>>>>> boot up automatically and get both wrong time on the main systems, and |
10 |
>>>>> different times on the various systems. |
11 |
>>>>> |
12 |
>>>>> My setup has a main server which does NTP, but with no direct link to |
13 |
>>>>> the outside. Router&firewall /have/ to be booted booted later (dumb |
14 |
>>>>> setup, don't ask), after which I can finally get correct time from NTP. |
15 |
>>>>> |
16 |
>>>>> NTP initiates "11 minute mode", which makes /etc/adjtime useless as far |
17 |
>>>>> as I understand. Anybody have a /correct/ way to account for RTC drift |
18 |
>>>>> on a box running ntpd? Right now I have a ---file in |
19 |
>>>>> /etc/cron.d/time-bad like so: |
20 |
>>>>> * * * * * root adjtimex -S 5 >/dev/null 2>&1 </dev/null |
21 |
>>>>> --- |
22 |
>>>>> |
23 |
>>>>> Combined with an old-fashioned setup for hwclock during boot and |
24 |
>>>>> shutdown. This feels really wrong, and I have no idea what I am doing. |
25 |
>>>>> |
26 |
>>>>> TLDR: Anybody have a /correct/ way to account for RTC drift on a box |
27 |
>>>>> running ntpd? |
28 |
>>>>> |
29 |
>>>>> |
30 |
>>>> |
31 |
|
32 |
Have you looked at adjtimex ... its in portage |
33 |
|
34 |
|
35 |
From the man page ... |
36 |
"For a standalone or intermittently connected machine, where it’s not |
37 |
ossible to run ntpd, you may use adjtimex instead to correct the sys-tem |
38 |
clock for systematic drift. |
39 |
|
40 |
There are several ways to estimate the drift rate. If your |
41 |
computer can be connected to the net, you might run ntpd for at least |
42 |
several hours and run "adjtimex --print" to learn what values of tick |
43 |
and freq it settled on. Alternately, you could estimate values using as |
44 |
a reference the CMOS clock (see the --compare and --adjust switches), |
45 |
another host (see --host and --review), or some other source of time |
46 |
(see --watch and --review). You could then add a line to rc.local |
47 |
invoking adjtimex, or configure /etc/init.d/adjtimex or |
48 |
/etc/default/adjtimex, to set those parameters each time you reboot." |
49 |
|
50 |
Used it at one time for dialup which approximates your condition. |
51 |
|
52 |
BillK |