Server locks up or mysteriously reboots during disk I/O over the IDE bus

Problem

The server locks up hard or mysteriously reboots during moderate to heavy disk I/O over the IDE bus. When this occurs, the kernel does not report any error messages in the system log file or on the screen.

Affected Models

Solution

According to the kernel developer, there is a generic bug in the IDE code that results in a race condition. When the condition is triggered, the bus will lock up hard. To workaround the bug, APIC must be disabled by specifying the kernel parameter 'noapic' to the Linux boot loader.

Here is a sample entry for the GRUB boot loader: 

# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE:  You have a /boot partition.  This means that
#          all kernel and initrd paths are relative to /boot/, eg.
#          root (hd0,0)
#          kernel /vmlinuz-version ro root=/dev/hda3
#          initrd /initrd-version.img
#boot=/dev/hda
default=0
timeout=5
splashimage=(hd0,0)/grub/splash.xpm.gz
hiddenmenu
title CentOS-4 x86_64 (2.6.9-22.ELsmp)
        root (hd0,0)
        kernel /vmlinuz-2.6.9-22.ELsmp ro root=LABEL=/ rhgb quiet numa=on noapic
        initrd /initrd-2.6.9-22.ELsmp.img