> Is 60+ days of uptime unstable?? I have 3 5.0 boxes here and have never
> had them crash for any reason except a power failure that lasts longer than
> the ups batteries. Loads vary from light to very heavy. What is the beef?
my experience has been that 5.0 boxes are stable to a point.
We use 4.2 as servers, but I have a 5.0 box on my desk, for use as an X
workstation -- and it works..it only crashes on exceedingly rare occasions
(maybe once or twice a month, and it sees ten hour a day use). It has
locked solid thanks to large textfiles, thanks to the root partition
filling up (which should make it unhappy, but having it freeze solid (to
the point that I had to boot off floppy to fix the problem) is not a good
trait for a production server) and because of something netscape did (I
never did figure out what, but that was the only app open at the time, and
it opened a billion or so error windows, then the machine locked solid --
not even ctrl-alt-delete would reboot it, nor would ctrl-alt-back space
take it out of x windows). There is also some funkyness with the keyboard
routines (the box will lock up waiting for the keyboard under certain
conditions...like when you're typing on a remote machine via ssh and your
network connection goes out from under you). I guess the best way to put
it would be that in normal conditions, it is stable, but the farther away
from normal you go, the less likely it will stay that way.
For a production server, I want something that is rock solid, 4.2 is
usually that way.
The problems largely seem to be related to glibc, btw.
And none of this goes for alpha-linux..for which 4.2 is stable and 5.0
sucks large rocks through a straw :) but that's a discussion for the AXP
list.
Vinnie
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
