- Fishstick
- Jul 9, 2005
-
Does not require preheating
|
I need some help trying to mount a UFS disk in linux. The disk is sata and was in a freebsd (5.x) system in a raid 1. It's attached to the linux machine currently with a usb -> sata cable.
I'm using a 2.6.9.x linux kernel (centosplus), with ufs compiled as a module. So far so good, but none of the ufs types are recognized:
code:# mount /dev/sdc /mnt/bsd/ -r -t ufs -o ufstype=5xbsd
mount: wrong fs type, bad option, bad superblock on /dev/sdc,
ufs_read_super: bad magic number
code:(parted) select /dev/sdc
Using /dev/sdc
(parted) check 1
Error: Unable to open /dev/sdc - unrecognised disk label.
(parted) print all
Error: Unable to open /dev/sdc - unrecognised disk label.
code:SCSI device sdc: 625142448 512-byte hdwr sectors (320073 MB)
sdc: Write Protect is off
sdc: Mode Sense: 00 38 00 00
sdc: assuming drive cache: write through
SCSI device sdc: 625142448 512-byte hdwr sectors (320073 MB)
sdc: Write Protect is off
sdc: Mode Sense: 00 38 00 00
sdc: assuming drive cache: write through
sdc: unknown partition table
sd 2:0:0:0: Attached scsi disk sdc
sd 2:0:0:0: Attached scsi generic sg2 type 0
--
# cat /proc/scsi/usb-storage/2
Host scsi2: usb-storage
Vendor: JMicron
Product: USB to ATA/ATAPI Bridge
Serial Number: D57CA0D48104
Protocol: Transparent SCSI
Transport: Bulk
--
# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: ATA Model: ST3250310AS Rev: 3.AA
Type: Direct-Access ANSI SCSI revision: 05
Host: scsi0 Channel: 00 Id: 01 Lun: 00
Vendor: ATA Model: ST3250310AS Rev: 3.AA
Type: Direct-Access ANSI SCSI revision: 05
Host: scsi2 Channel: 00 Id: 00 Lun: 00
Vendor: WDC WD32 Model: 00YS-01PGB0 Rev: 0M21
Type: Direct-Access ANSI SCSI revision: 02
Entertaining suggestions!
Fishstick fucked around with this message at 11:07 on Apr 24, 2008
|
#
¿
Apr 24, 2008 11:04
|
|
- Adbot
-
ADBOT LOVES YOU
|
|
|
#
¿
Apr 19, 2024 23:58
|
|
- Fishstick
- Jul 9, 2005
-
Does not require preheating
|
H110Hawk posted:
Are you sure the UFS and BSD partition modules are loaded? Is the raid1 completely transparent? Can you make a raid1 of a linux disk, then split it off onto this machine and use that?
I'm going to suggest you upgrade your kernel to something recent. 2.6.9 is pretty old. http://kernel.org/pub/linux/kernel/v2.6/linux-2.6.25.tar.bz2 Give that a whirl and report back.
Unfortunate typo on my part on the kernel version, it's actually a 2.6.18-53.1.14.el5.centos.plus.
The UFS and bsd_label modules are active, before activating them mount will just spit out "unknown filesystem option" if you try to use ufs, now it lists the ufstypes option as well. Modprobe doesn't error either.
I'm not clear on what you mean by this though:
quote:Can you make a raid1 of a linux disk, then split it off onto this machine and use that?
|
|
#
¿
Apr 24, 2008 13:31
|
|
- Fishstick
- Jul 9, 2005
-
Does not require preheating
|
It's the same hardware/setup we use on other machines, just bsd.
It's a 3ware hardware raid, and on our linux machines I've always been able to just remove a drive from the array, replace it with a fresh one, and pull data off of the removed drive just by mounting it as ext2/3. Unless i'm once again misunderstanding what you mean.
|
|
#
¿
Apr 24, 2008 14:41
|
|
- Fishstick
- Jul 9, 2005
-
Does not require preheating
|
Yeah, but machine wasn't recognizing any of the slices, so that was my desperate attempt  .
In the meantime i've been able to get the BSD box running again using a 4month old disk that was taken out of the raid 1 at some point (these guys had an awesome backup scheme!), and booting from there. I've hooked the data disc (with the recent data) on a sata port outside of raid. But it's not looking too good:
code:# fdisk /dev/ad2
******* Working on device /dev/ad2 *******
parameters extracted from in-core disklabel are:
cylinders=155061 heads=16 sectors/track=63 (1008 blks/cyl)
Figures below won't work with BIOS for partitions not in cyl 1
parameters to be used for BIOS calculations are:
cylinders=155061 heads=16 sectors/track=63 (1008 blks/cyl)
fdisk: invalid fdisk partition table found
Media sector size is 512
Warning: BIOS sector numbering starts with sector 1
Information from DOS bootblock is:
The data for partition 1 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 63, size 156301425 (76319 Meg), flag 80 (active)
beg: cyl 0/ head 1/ sector 1;
end: cyl 436/ head 15/ sector 63
The data for partition 2 is:
<UNUSED>
The data for partition 3 is:
<UNUSED>
The data for partition 4 is:
<UNUSED>
If I understand correctly, I should be able to retrieve the disklabels through scan_ffs, match it with the disklabels of the older raid-removed disk - that is now the bootdrive - and apply them to the hosed disk, after making a copy with dd. This way I should hopefully be able to recover the slices and mount it.
However, there's a 2block difference in the labels between the bootdisk and the datadisk, could this be due to disk geometry (bootdisk is 160gb with 80gb partition, data disk is 80gb with 80gb partition) or should I correct the label with what I know is correct, versus what scan_ffs has found on it?
code:#Boot/Old removed disk:
h: 120176370 36119952 4.2BSD 2048 16384 28552
# Disk to be recovered:
X: 120176368 36121039 4.2BSD 2048 16384 0 # /usr
E: Welp, that worked. Too bad it was for nought since half the goddamn folders were busted in the end anyway. Lessons learned.
Fishstick fucked around with this message at 18:52 on Apr 25, 2008
|
|
#
¿
Apr 25, 2008 08:46
|
|
#
¿
Apr 24, 2008 11:04
