If its two servers doing the same, then I guess it's not likely they both have the same hardware problem. The thing is, that's not something centos is going to do on its own, so it's some program that's been added, or some common bios setting that's wrong. Do they connect to a UPS with a serial/usb cable? >>Also run memtest86 on them overnight (getting at least one complete iteration). I've seen one memtest iteration pass, but 2 or 3 were needed before a failure showed up. That's not usually the case, though... _______________________________________________ CentOS mailing list CentOS@xxxxxxxxxx http://lists.centos.org/mailman/listinfo/centos