On Thu, 27 Jul 2006, Russell Butturini wrote: > Hello all: > > > > My company builds Linux based web content filters using CentOS 4.3 > as the base. We always apply updates to the new machines from the > default repositories before deployment. Recently, the last two > CentOS machines we built on Dell Dimension 1100s have failed, and > the root file system seem to be getting corrupted. An e2fsck tries > to recover the journal, but never repairs the machine enough to > completely boot up. This only seems to be happening on the > Dimension 1100s, as we have several other machines built with the > same kernel version and software running on it that have not failed. > The machines are Celeron-Ds, 2.4-2.8 Ghz, with either 256 or 512 MB > of RAM and IDE hard disks. We've got a handul of 1100s (actually, the 1100n model, mostly because I love getting a printed copy of the GPL from Dell :-), with CentOS 4.3. No troubles to report. Can you get to a recovery shell? If so, does 'smartctl -a /dev/hda' provide any clues? -- Paul Heinlein <heinlein at madboa.com>