Re: uptime

[Alan Pope <alan.pope@gmail.com>]

On 09/11/05, Joseph Manning <manning@cs.ucc.ie> wrote:
> On 2005-Nov-09 (Wed) at 07:54 (-0600), Willie McKemie wrote:
>
> >> This is not laptop specific, but.....
> >>
> >> --
> >> Willie, ONWARD!  Through the fog!
> >> http://counter.li.org Linux registered user #228836 since 1995
> >> FOUL! My uptime counter "rolled over" to 0 at about 497 days 11/6/2005
> >> Actual uptime is about 497 days plus amount listed below
> >> Debian3.0/GNU/Linux system uptime  2 days 18 hours 34 minutes
>
> Willie,
>
>    Yes, this has happened to me also.
>
>    Note that 2^32 hundredths-of-a-second = 497.1027 days,
>    which seems like a plausible explanation of what happened.
>
>    It's also the case that the CPU time of a process rolls over
>    after the same period of time (497+ days).  Yes, this has happened
>    to me.  Really.
>
>    Hah! -- whoever designed such short time limits?
>    O they of little faith!
>

It does actually say on the site..

http://counter.li.org/reports/uptimestats.php

"NOTE: The Linux kernel (at least up to 2.4.2) has a flaw: It computes
the result of the "uptime" based on the internal "jiffies" counter,
which counts the time since boot, in units of 10 milliseconds. This is
typecast as an "unsigned long" - on the Intel boxes, that's an
unsigned 32-bit number. "

Cheers,
Al.