When RAID arrays go bad
April 5, 2005 7:58 PM Subscribe
The 2x250Gb RAID1 array on my desktop system (WinXP) started throwing up 'Delayed write failure' errors and forcing mirroring on reboot. I unplugged the HDs and reinstalled Windows, but suspect a hardware failure. What's the best way to run diagnostics while minimising the risk of corrupted data being mirrored across? Plug in each drive separately, detach the array, or re-mirror and hope for the best?
This thread is closed to new comments.
As a stopgap, I might suggest switching from write-back to write-thru caching. How this is done varies according to your hardware -- sometimes it's not even an option for desktop-oriented (as opposed to server-oriented) hardware.
With write-back caching, data is written to the cache immediately but to the disk later (either when cache memory needs to be recycled or when some transaction counter / timer expires.) This is the "delayed write" that your system is complaining about. In write-through caching data is written to both cache and disk immediately -- safer but slower.
Did I mention backing up your system? If the data is important enough to warrant mirroring, then you should already have been following a regular backup procedure.
posted by randomstriker at 12:40 AM on April 6, 2005