Tor Harald Thorland was having trouble with a USB disk.
I suggested:
Can you try booting with acpi=off (at the end of the kernel comand line: edit it in grub or in /boot/grub/grub.conf)? I know that it's not really a proper solution on a laptop, but it might help you find one.
I don't really know enough about the ins and outs of acpi to tell, but if acpi=off cures the problem, then acpi=noirq might be worth trying.
He replied:
The messages file now contains the following:
Sep 27 20:11:00 dsl-69-134 kernel: ohci_hcd 0000:00:02.0: wakeup Sep 27 20:11:00 dsl-69-134 kernel: usb 1-1: new full speed USB device using address 2 Sep 27 20:11:00 dsl-69-134 kernel: Initializing USB Mass Storage driver... Sep 27 20:11:00 dsl-69-134 kernel: scsi0 : SCSI emulation for USB Mass Storage devices Sep 27 20:11:00 dsl-69-134 kernel: Vendor: USB 2.0 Model: Mobile Disk N4M Rev: 1.00 Sep 27 20:11:00 dsl-69-134 kernel: Type: Direct-Access ANSI SCSI revision: 02 Sep 27 20:11:00 dsl-69-134 kernel: Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0, type 0 Sep 27 20:11:00 dsl-69-134 kernel: usbcore: registered new driver usb-storage Sep 27 20:11:00 dsl-69-134 kernel: USB Mass Storage support registered. Sep 27 20:11:00 dsl-69-134 kernel: SCSI device sda: 999424 512-byte hdwr sectors (512 MB) Sep 27 20:11:01 dsl-69-134 kernel: sda: Write Protect is off Sep 27 20:11:01 dsl-69-134 kernel: sda: assuming drive cache: write through Sep 27 20:11:01 dsl-69-134 kernel: sda: sda1 Sep 27 20:11:01 dsl-69-134 kernel: Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0 Sep 27 20:11:01 dsl-69-134 kernel: updfstab: Using deprecated /dev/sg mechanism instead of SG_IO on the actual device Sep 27 20:11:01 dsl-69-134 scsi.agent[2425]: disk at /devices/pci0000:00/0000:00:02.0/usb1/1-1/1-1:1.0/host0/0:0:0:0
Sounds like things are getting further, then. What happens if you try mounting /dev/sda1 somewhere?
mount /dev/sda1 /misc
What does the acpi do?
Too much, and not well enough. The Advanced Configuration and Power Interface is the subject of many a rant by kernel developers. It's supposed to be a standard way for hardware to tell software the details of what's there, how to configure it, and how to put it into power- saving mode. In practice, this means that the kernel has to rely on data from the BIOS, whose writers often consider it debugged if Windows boots.
It also affects the way that IRQs are detected and handled.
James.