On Thu, 09 Oct 2008 17:19:46 +1300 (NZDT) Derek Smithies <[EMAIL PROTECTED]> wrote:
> > I've got debian servers with uptimes measured in years ( well, except > > for the single reboot when they moved data centres about a year ago ), > > and I've got CentOS servers in the same category. > Long uptimes are misleading. > All a long uptime reports is the length of time you have between kernel > upgrades. Which "suggests" you are running old kernels.. Sigh - probably > not an issue for you, but..... Indeed, but they're all 2.6.x ( 2.6.9 is the oldest I *think* ). 2.4 (or older!) and I would upgrade. Running the latest, shiniest kernel is ok, once all your software is tested and proved stable on it. However, in most (server) situations it's that stability that's paramount, and installing newer kernels will usually either increase performance or functionality... and if you don't need either, then the years of uptime have proved the stability many times over! IMO it's all down to managing risk, and as we all perceive risk differently we'll make varying decisions. But that's another matter altogether (: I'm a great believer in "if it ain't broke..." (: ...but my lappy's running intrepid, just for fun. Cheers, Steve -- Steve Holdoway <[EMAIL PROTECTED]>
