uptime 2 years!

[reese at adeptscience.com]

Well sometimes you don't need to upgrade and you aren't connected to the 
internet directly.

elephant: {25} uptime
 5:54PM  up 1756 days,  7:07, 2 users, load averages: 1.04, 1.01, 1.00

elephant 4.9-RELEASE FreeBSD 4.9-RELEASE #0: Mon Oct 27 17:51:09 GMT 2003

This machine is semi-retired now but for its first three years it was 
the database server (ads, reg, hit logging etc.) for a large website
(> 300,000 pages/day).  It also handled the queries for a 
monthly reports server that created detailed reports 
for about 5000 companies that had content on 
the site.   I didn't keep track of the connections then but its 
replacement is doing  28,527 conn/hr.

This was on an internal network that was firewalled from everything 
but port 3306 on the webserver IP, and a couple admin IPs.  It was a big exercise to 
replace it as the databases were quite large (48G) and it took a good fraction 
of an hour to make the occasional snapshot for starting a new replicator 
when needed.  

FreeBSD really is one of the most stable OSs even under a pretty 
good load.



Cheers,
Charlie

"One OS to rule them all"  :-)


On 9 Oct 2008 at 3:45, andrew clarke wrote:

> On Wed 2008-10-08 09:21:53 UTC-0700, Jeremy Chadwick (koitsu at FreeBSD.org) wrote:
> 
> > I don't want to rain on your parade, but uptime ultimately means squat.
> 
> Agreed.
> 
> > I can install FreeBSD on a box under my desk at home, on a UPS, and
> > leave it powered on for the next 30 years -- it tells people absolutely
> > nothing about the reliability of the OS, or what kind of stress it's
> > undergone during that time.
> 
> I'd be impressed if an ordinary PC lasted 30 years continuously
> running.  Even if the HDD is solid-state you still have to think about
> other moving parts, particularly the CPU and PSU cooling fans.  I've
> had a bad run with PSU fans recently.
> 
> Is FreeBSD 7.1 2038-proof?  ;-)
> 
> http://en.wikipedia.org/wiki/Year_2038_problem
> 
> (I wonder what version of FreeBSD will be the latest in 2038?)
> 
> > Additionally, long uptimes also reflect directly on sysadmins: I take it
> > to mean "the administrator is very lazy".  There are security holes
> > (kernel or userland/library-level) which are exploitable on boxes which
> > have been up for that kind of time.  I'm also making the assumption that
> > said boxes have Internet connectivity, hence my point.
> 
> Yes, my initial thought was "what, you don't use freebsd-update?".
> 
> Regards
> Andrew
> _______________________________________________
> freebsd-questions at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> To unsubscribe, send any mail to "freebsd-questions-unsubscribe at freebsd.org"
> 
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email 
> ______________________________________________________________________
>