So the DELL PowerEdge 400SC (P4, 2.8Ghz, Windows 2003 Small Business Server - 1GB memory) purchased in 2004 is configured RAID-1 with 2, IDE, 80gb HDs. The alarm sounded the other day - indicating possible failure of one of the channels on the controller. Rebooted into controller BIOS, silenced the alarm and learned one of the drive channels had failed. The other read as ONLINE.
No media errors could be discerned on either driver from the BIOS diagnostics, and rebuild of the failed drive channel failed about 1/4 of the way through. Ordered new 80GB IDE HD (still can find some). Booted back into OS using degraded drive channel) and commenced scan disk of the two main partitions (c: = OS and D: = file shares, etc ...) Can't scan in OS while Windows files are in use, so rebooted and the other scan disk commenced after Windows splash screen rendered. Encountering numerous "File Record Segment XXXX is unreadable" messages. Scan is taking a long time, but because I chose to correct system files and bad sectors before launching scan, can I cling to hope that Windows will rewrite to good sectors or is the data on the unreadable sectors potentially lost?
I learned that errors in the good drive could easily have corrupted the other in the RAID array - so I'm not sure replacing the failed channel drive will restore the system to normal - so hoping the scan disk procedure will help.
I backup regularly to NAS and Cloud - so restoration can and will happen. We were scheduled to replace the server this spring.
Any other helpful thoughts? Advice? How long SHOULD a server live these days?
Well 6 years is a pretty impressive figure, but I think it's time to replace your entire array. Does the server need to stay up 24/7? I'd suggest building a new array from new hard disks. In the meantime, you can check if the disks are actually ok by trying them out on another computer.