 | | | Raid5 filesystem corruption RH9 | Raid5 filesystem corruption RH9 2005-10-04 - By Lars Bjaerris
Back
Hi everybody,
I have a coworker with a problem, but must admit that I'm not
experienced enough, with this kind of problem to give advice.
He's not on the list, so I thought I'd give it a go here and learn.
So here goes:
The box is running Redhat 9.0
Primarily running anaconda
Raid5 with a HighPoint 374 IDE RAID controller.
During boot fsck reports filesystem corruption on the root partition.
It has 2 partitions / and /boot
Disk drivers were added during anaconda install.
Any help or pointers would be greatly appreciated.
Lars Sorensen
#################### Email sent to me ###################
The problem is that I have a HighPoint 374 IDE RAID controller
in the machine and it isn't possible to access the raid volume
unless I (a) turn off the kernels ability to auto-probe the
controller and just assign each of the six disks as their own
IDE devices (/dev/hd<x> you know), and (b) load the appropriate
kernel module from somewhere.
(a) isn't very difficult to accomplish. It should suffice to
write something like "ide0=noprobe ..." or maybe "hdd=noprobe ..."
at the "boot: " prompt.
(b) is a lot more difficult to make happen during booting.
So, I tried loading the "hpt374.o" kernel module manually
using "insmod". No luck. There are at least three unresolved
symbols that have to do with the scsi driver, I think.
Note: this is the same drivers I installed the stupid machine
with originally! (Only I added the drivers in the installation
phase of "anaconda" where I could use the "Add driver disk"
feature).
You wouldn't happen to know if it is possible to do the
"Add driver disk" *after* installation, do you?
(I.e. use some kind of program that would mimic what "anaconda"
does at installation, but doing it in "run time" so to speak.)
#################### Email sent to me ###################
--
Taroon-list mailing list
Taroon-list@(protected)
https://www.redhat.com/mailman/listinfo/taroon-list
|
|
|