Han Boetes wrote:

quoted text
> Gilles Chehade wrote:
> > [snip]
> > ...and in OpenBSD land you better have a GOOD reason to have an
> > uptime longer than two releases.
> > [snip]
> > Gilles,
> > proud to never exceed 200 days of uptime
>
> Speaking of which... I think I have a pretty good reason to bring
> down my uptime very shortly. YAY!
>
> ~% uptime
> 10:39AM  up 82 days, 14:06, 1 user, load averages: 0.22, 0.20, 0.17

IMO, indvidual machines' uptime doesn't matter at all.

Overall system uptime does matter (this is why we have
carp, pfsync and relayd).

Think Google. They have their cheap machines dying
(or rebooting after patching) all the time, but overall
system stability is very high.

Alexey