The main 40 GB hard drive that powers this server has apparently given up the ghost. Yesterday, I woke up to check the server as normal and it had decided to remount the filesystem read-only, which is never a good sign. After rebooting (and losing my >25 days of uptime 😛 ) into my recovery environment, I was able to repair the filesystem, reboot and it worked fine.
Until about 2am this morning, when the same thing apparently happened again.
Unfortunately, that probably means the main drive is dead. There are actually two drives in this server, that 40 GB one and a 160 GB drive which important web server data is mirrored to each night (although which has question marks about stability for running a boot drive).
For now, I’ve copied everything over to the 160 GB drive until I can get a replacement main drive. Thanks to the flexibility of Linux, I simply copied the data over, rebooted off a GRUB CD-ROM, entered the boot commands for the second drive and it booted as if nothing had happened. 😉
Now, I don’t know whether to just get another drive and continue with the current setup or get two SATA drives and try doing some form of RAID. Or do that later. I don’t know.