Ok - I'm a dummy. After a good night's sleep, I was able to bring the failed reiser system back up. 1) The / partition was actually on /dev/hda4, not hda2 2) The segmentation fault was ocurring in the /sbin/init.d/boot file, around line 456. Looking at the problem area, this had to do with booting systems in the fstab. 3) I mounted /dev/hda4 from the other system, and commented out /dev/hdg entries in fstab. 4) Rebooted, and everything came up. My hdg is hosed, though. Question: On an ext2 system, if I have problems mounting a filesystem, I have the option of remounting, or going into maintenance mode. If this was an ext2, I could have mounted and edited the fstab. THIS situation only left my machine frozen. How can I prevent this? Here is the stack from the original crash - I sent this to the reiser folks: _______________________________________________________ Aug 27 13:44:07 heronet kernel: hdg: drive not ready for command Aug 27 13:44:37 heronet kernel: hdg: status error: status=0x58 { DriveReady SeekComplete DataRequest } Aug 27 13:44:37 heronet kernel: hdg: drive not ready for command Aug 27 13:47:53 heronet kernel: Unable to handle kernel paging request at virtual address 249de234 Aug 27 13:47:53 heronet kernel: current->tss.cr3 = 0426e000, %cr3 = 0426e000 Aug 27 13:47:53 heronet kernel: *pde = 00000000 Aug 27 13:47:53 heronet kernel: Oops: 0002 Aug 27 13:47:53 heronet kernel: CPU: 1 Aug 27 13:47:53 heronet kernel: EIP: 0010:[search_by_key+140/1080] Aug 27 13:47:53 heronet kernel: EFLAGS: 00010296 Aug 27 13:47:53 heronet kernel: eax: cafc5060 ebx: 00000000 ecx: 249de234 edx: cafc5060 Aug 27 13:47:53 heronet kernel: esi: 00000003 edi: ffffffff ebp: cafc5061 esp: ccbb5e3c Aug 27 13:47:53 heronet kernel: ds: 0018 es: 0018 ss: 0018 Aug 27 13:47:53 heronet kernel: Process smbd (pid: 1054, process nr: 66, stackpage=ccbb5000) Aug 27 13:47:53 heronet kernel: Stack: 00001000 ccbb5e80 ccbb5f24 ccbb5f2c ccbb5f6c ccbb5f7c c02c2ec8 ccbb5e80 Aug 27 13:47:54 heronet kernel: 00000000 ccbb5f68 cbc4f018 00000010 00000000 249de234 cafc5060 00001000 Aug 27 13:47:54 heronet kernel: 22010000 00000000 cffeade0 00000086 c02dd180 c0171d02 c056f400 ccbb5f6c Aug 27 13:47:54 heronet kernel: Call Trace: [search_by_entry_key+42/136] [write_intr+0/312] [reiserfs_readdir+187/556] [do_edge_ioapic_IRQ+120/168]Aug 27 13:47:54 heronet kernel: [sys_getpid+0/20] [system_call+52/64] Aug 27 13:47:54 heronet kernel: Code: 89 01 89 44 24 40 83 c4 10 85 c0 75 7c 8b 5c 24 58 83 c4 f4 --- Ron Heron <heroron@yahoo.com> wrote:

Hello,

I recently did a fresh install on another system, but went with the reiserfs. All seemed to be going well until I ran into a slight problem.

Here's the skinny on the problem:

2.2.14-SMP kernel, on a dual celeron BP6

3 disks total, OS installed on hda with reiser, 2 data disks on UDMA ports (hde, hdg) of the abit bp6 (high point controllers) hda= ext2 on /boot, reiser on /home and /, hde = ext2, hdg =reiser.

Samba shares connected to hdg.

I was copying image files from hde to hda (large files) using kde cut and paste (first time using it - nice). At the same time, one of my lan users started a copy to the samba share on hdg.

That's when I got a kernel oops, segmentation fault.

The kfm crashed, and samba (smbd) was unkillable. I (stupid me) did a restart of the system, and now it is unbootable. When I try and reboot, it gives me a segment fault while trying to mount the root fs.

Since I have another system installed on /dev/hde, I booted into that one (ext2) and was able to mount the /boot (/dev/hda1 - ext2) partition, and the /home (dev/hda3 - reiser) partition, but gave a bad superblock message when trying to mount the root partition (/dev/hda2)

I did a reiserfsk on /dev/hda2, and it failed the --check. So, I did a rebuild, and it was able to fix itself, but I still cannot mount.

#>mount -t reiserfs /dev/hda2 /mnt mount: wrong fs type, bad option, bad superblock on /dev/hda2, or too many mounted file systems

I know the options are good, cause I can mount /dev/hda3 ok.

I want to get back into this system so I can send in the stack dump to the reiser folks - any ideas?

Also, having read the reiser FAQ, it should have hit me then, but there are some issues with nfs. Obviously, samba falls into this? If this is the case, then perhaps reiser isn't a viable option for samba-centric systems?

Any help greatly appreciated.

Ron

__________________________________________________ Do You Yahoo!? Yahoo! Mail - Free email you can access from anywhere! http://mail.yahoo.com/

-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq

__________________________________________________ Do You Yahoo!? Yahoo! Mail - Free email you can access from anywhere! http://mail.yahoo.com/ -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq