On  9 Nov, I wrote:
>  I've been doing some Google searches.  Looks like Intel changed the 
>  peripheral chip for the CPU after releasing the spec to the 
>  manufacturers, so it doesn't remember to initialise the IDE PCI 
>  controller properly.  There is some work going on for the 2.4.20 kernel. 

I should mention that I searched using the error message:

PCI: Device 00:1f.1 not available because of resource collisions

from the part of the boot messages that said:

ICH4: IDE controller on PCI bus 00 dev f9
PCI: Device 00:1f.1 not available because of resource collisions
ICH4: (ide_setup_pci_device:) Could not enable device.

>  I'll try and figure out how to patch a standard 2.4.19 kernel to 
>  2.4.20-10ac tomorrow.  My first attempt failed dismally. 

Thanks to Jan Schmidt's explanations of the kernel patch naming
conventions, I've built the 2.4.20-pre10-ac2 kernel, and can happily
report that it fixes the problem.  It's interesting reading the dmesg
messages and seeing how the kernel straightens out the improperly
initialised PCI devices.

So, I now have all drives plugged in and working.  The RAID isn't
functioning again as a RAID yet, since the out of date mirrors have
been noted as such and kicked out.  I'll do some reading on bringing
the RAID back online tomorrow.

I'm truly impressed, again, by how good a job the Linux kernel team do.

Plus, I've proved to myself that having one of your RAID drives out of
action is handled extremely gracefully by Linux's software RAID.  :-)

luke

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Reply via email to