I am trying to setup a 9.0Pro machine using 4 scsi discs on a RAID5 (software based) but I am having troubles with it. If I boot on runlevel 1 the system does not crash, but if I boot over another runlevel, the system will freeze after a while with a Oops message, related to ipv6 driver, raid5.o, ssh, or other. Here are some of the Oops, This is a brand new installation with almost everything default!!! Thanks for any help you can give! David ############################################################################ # Jan 15 22:27:39 linux kernel: md: md0: sync done. Jan 15 22:27:39 linux kernel: raid5: resync finished. Jan 15 22:47:40 linux kernel: Unable to handle kernel paging request at virtual address 000011e6 Jan 15 22:47:40 linux kernel: printing eip: Jan 15 22:47:40 linux kernel: c7c13ddc Jan 15 22:47:40 linux kernel: *pde = 00000000 Jan 15 22:47:40 linux kernel: Oops: 0002 2.4.21-166-athlon #1 Thu Dec 18 18:24:05 UTC 2003 Jan 15 22:47:40 linux kernel: CPU: 0 Jan 15 22:47:40 linux kernel: EIP: 0010:[ipv6:mipv6_functions_R874e5304+51758556/804608079] Tainted: P Jan 15 22:47:40 linux kernel: EIP: 0010:[<c7c13ddc>] Tainted: P Jan 15 22:47:40 linux kernel: EFLAGS: 00010296 Jan 15 22:47:40 linux kernel: eax: 00000000 ebx: 08073fc0 ecx: 00000000 edx: 08073fd0 Jan 15 22:47:40 linux kernel: esi: 00000030 edi: c7c13eba ebp: c7c13fa4 esp: c7c13dc4 Jan 15 22:47:40 linux kernel: ds: 0018 es: 0018 ss: 0018 Jan 15 22:47:40 linux kernel: Process man (pid: 3459, stackpage=c7c13000) Jan 15 22:47:40 linux kernel: Stack: 08073fc8 c7c13ddc 00000008 08073fd4 00033f8d 00000000 11e68900 00000000 Jan 15 22:47:40 linux kernel: 00000028 00000000 c8765848 c7c13f24 00000016 c875cb26 c7c13eba c875c3f0 Jan 15 22:47:40 linux kernel: c328e470 c7c13fa4 c7c13ea4 00000016 11e68900 00000000 0002c39a 00000000 Jan 15 22:47:40 linux kernel: Call Trace: [keybdev:__insmod_keybdev_O/lib/modules/2.4.21-166-athlon/kernel/dri+4265276 528/96] (216) [cached_lookup+24/112] (120) [vfs_readdir+169/176] (12) Jan 15 22:47:40 linux kernel: Call Trace: [<c328e470>] (216) [<c0154798>] (120) [<c0159369>] (12) Jan 15 22:47:40 linux kernel: [filldir64+0/288] (20) [sys_getdents64+91/181] (08) [filldir64+0/288] (32) [sys_fcntl64+93/192] (36) Jan 15 22:47:40 linux kernel: [<c01599f0>] (20) [<c0159b6b>] (08) [<c01599f0>] (32) [<c0158acd>] (36) Jan 15 22:47:40 linux kernel: [system_call+51/56] (60) ############################################################################

On Monday 02 February 2004 22:01, Alexandr Malusek wrote:

Rikard Johnels writes:

...

...

...

I am trying to back up a 6 gig harddrive with "dd" ... dd: writing to '/mnt/image.dd': File to large 4195234+0 records in 4195234+0 records out

NFS from the standard instaled 8.2 release. (Probably NFS v2)

You can avoid using NFS via:

dd if=/dev/hda | ssh -c blowfish hostname "dd of=image.dd"

where hostname is the other computer.

-- A.M.

Do i really need the -c blowfish? Its on my own three computer network Also, will it ask ask for password? I havent set up password less logins -- /Rikard ------------------------------------------------------------------------------------ Rikard Johnels email : rikjoh@norweb.se Web : http://www.rikjoh.com Mob : +46 (0)735 05 51 01 ------------------------ Public PGP fingerprint ---------------------------- < 15 28 DF 78 67 98 B2 16 1F D3 FD C5 59 D4 B6 78 46 1C EE 56 >

On Mon, Feb 02, 2004 at 09:21:10PM +0100, Rikard Johnels wrote:

On Monday 02 February 2004 20:54, Alexandr Malusek wrote:

...

Rikard Johnels writes:

...

I am trying to back up a 6 gig harddrive with "dd" ... dd: writing to '/mnt/image.dd': File to large 4195234+0 records in 4195234+0 records out

Is there a easy way to get around the problem?

Use the skip and count options of dd to save the partition in several files.

...

How do i fix the system for larger files?

Which network FS do you use?

...

The source FS is NTFS The targer FS is ReiserFS (on the network disk)

Both can handle files larger than 2 GiB, see the Large File Support chapter in the Admin Guide.

--

Would using DD be a good way to backup/restore a system from bare metal? For example. I have a system running and configured with a few years of configuration changes etc. If I lose a drive and have to rebuild from scratch after intalling the base system from CD's it would be darn near impossible ( in a small amout of time ) to get the system fully configured as it was. What if I did a DD of my entire system ( all partitions ) and then reinstalled from the SuSE CD's and simply wrote out the dd image files appropriately? Is this a good idea? If it could work, what would be some pointers to doing it? I really don't use DD much ( only in the old days to create bootable linux floppies ). -- Jim Norton - http://www.jimnorton.org 2 + 2 = 5 for large values of 2

The Monday 2004-02-02 at 16:43 -0500, Ken Schneider wrote:

Another old time solution that you could use is dump/restore. This would work well if changing partition sizes.

Unless you have a reiserfs: | dump - ext2/3 filesystem backup -- Cheers, Carlos Robinson