On Jan 25, 2010, at 11:49 AM, Dave McGuire <mcguire at neurotica.com>
wrote:
On Jan 25, 2010, at 2:40 PM, Ray Arachelian wrote:
Here's your proof, once again:
ns$ uptime
1:17pm up 1062 day(s), 14:23, 1 user, load average: 2.05, 1.65, 1.85
ns$
For home use, especially if well firewalled off, that's just pure
awesomeness.
For production use, not so much, means that you don't know whether
this
box will boot up or not if it goes down hard. It is possible for
blocks
to have gone back, blocks on which the kernel or other startup
related
files may be mapped - unless you were the guy that set it up and
tested
it, you don't know whether it will boot up and bring up services,
whether you need to add routes to make it work, etc. - i.e. the
original
builder left the job for more cash, etc..
Good point, but this isn't my first BBQ my friend...There's a cold
spare in the rack.
It also means that it hasn't been patched in
that many days and is
likely vulnerable to lots of attacks. Despite that, it's quite
impressive. :-) Well played sir.
It's been patched, just nothing that requires a reboot. And as you
know, most UNIX-based OSs rarely require reboots for patching. For
example, I've replaced the Ethernet interface drivers on this
machine, hot, without rebooting.
I can (and do) replace video drivers on my Windows 7 boxes without
rebooting :) same for network and just about anything else.
Now if I could just beat some sense into the windows update guys so
they stop marking fixes to IE as "reboot required" when they're
clearly not...
Josh