| 1 |
Francesco Talamona wrote: |
| 2 |
> On Thursday 06 January 2011, Dale wrote: |
| 3 |
> |
| 4 |
>> I deleted the /etc/adjtime but it has not been recreated. What |
| 5 |
>> creates this file? I deleted it a couple days ago and nothing has |
| 6 |
>> brought it back yet. |
| 7 |
>> |
| 8 |
> /sbin/hwclock --systohc does. |
| 9 |
> |
| 10 |
> What happens in the long run if you don't run any synchronization |
| 11 |
> service at all? Does the time go stray or it just exhibits a systematic |
| 12 |
> error? |
| 13 |
> |
| 14 |
> Ciao |
| 15 |
> Francesco |
| 16 |
> |
| 17 |
|
| 18 |
I been using hwclock and it did generate adjtime file. So far so good. |
| 19 |
I been googling a bit. I ran a command with && and sleep cycles for a |
| 20 |
while to let the adjtime file get a good figure. It changed each time. |
| 21 |
I'm going to let it run for a while more. I ran that while ntp was |
| 22 |
stopped too. I didn't want ntp to get in the way. |
| 23 |
|
| 24 |
I haven't tried long term without ntp running. I did stop the service a |
| 25 |
while and it was off a lot in just a little bit. I think I was |
| 26 |
unmasking the unstable version and compiling it. Since this is a 4 core |
| 27 |
rig, it doesn't take that long. Downloading source, compiling and |
| 28 |
installing couldn't take longer than a few minutes plus me editing |
| 29 |
config files. I figure it was about 20 minutes or so and it was off a |
| 30 |
little less than a minute if I recall correctly. I guess my clock would |
| 31 |
be off hugely if left alone for a day or so. Since I don't reboot very |
| 32 |
often, I would have my days and nights messed up before long. lol |
| 33 |
|
| 34 |
I'm hoping having a good adjtime file will help some at least. Time |
| 35 |
will tell I guess. Pardon that usage of time. ;-) |
| 36 |
|
| 37 |
Dale |
| 38 |
|
| 39 |
:-) :-) |
Archives