On Mon, 2003-12-15 at 14:01, Craig Ayliffe wrote: > Any ideas on why the raid doesn't just ignore the failed disk and run > in degraded mode properly?
My guess would be that the failed drive is causing a large number of IDE bus resets, and waiting for those is how the kernel is spending its time. 'dmesg' probably gives clues. Generally, when you spec software RAID with IDE you expect this sort of thing and are really seeking reduced downtime rather than continued operation upon the failure of the drive. After all, it's not as though you can safely remove a failed IDE drive with the system running. What you do get is the ability to shutdown the box, remove the failed drive's power cable and reboot into a running system. Serial ATA might prove to be more robust. We'll see. -- Glen Turner Tel: (08) 8303 3936 or +61 8 8303 3936 Network Engineer Email: [EMAIL PROTECTED] Australian Academic & Research Network www.aarnet.edu.au -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
