[ale] Reiserfs problem solved

Dow Hurst dhurst at kennesaw.edu
Wed Dec 5 12:02:39 EST 2001


My root filesystem not being found turned out to be a problem I've never
encountered before.  I had originally created and formatted three
partitions, /dev/hda1-3, where 1 was /boot, 2 was swap, and 3 was /. 
These were created in the process of an ftp based SuSE 7.3 install.  The
install hung during DSL IP changes so I had to kill Yast1.  Linuxrc
recovered after Yast1 was killed so I restarted the install without
rebooting.  Package selections were kept so I didn't have to reselect
packages.  At this point I elected to *not* reformat the partitions but
just told the install what the partitions were formatted as and the
mount points.  After the install, 1.36Gb download total, Yast1/linuxrc
seemed to reboot and initialize the system.  The only problem during
initialization was the installation of lilo couldn't find boot.b in
/boot.  I realized, using a separate terminal, that boot.b wasn't in
/boot and that my /dev/hda1 wasn't mounted as /boot during the install
like I thought during the beginning.  I did some fast footwork at the
shell and mounted /dev/hda1 and moved files around appropriately.  The
original boot.b was still in /dev/hda1 so the installation of lilo then
completed with no errors.  The reboot by linuxrc/Yast1 seemed to work. 
I configured X and starting playing with the install while logged in as
root.  I ran Opera (which I liked), downloaded files off the net, and in
general just used the system like normal.  All filesystems were mounted
and working fine (I thought).  When I decided to quit, I thought I'd
sync and reboot just to see how quickly the startup might be and to look
for differences from SuSE 7.1 which was the previous OS.  That's when on
reboot that the kernel couldn't find my /dev/hda3, exiting with some
error about reiserfs superblocks not findable.  I need to reiterate that
the kernel was found and no LILO errors were produced.  I was into the
boot sequence where the / partition is to be mounted at the failure
point.

Well last night I worked on the problem by booting from the SuSE 7.1 DVD
to the rescue system (acts like a miniroot).  I ran reiserfsck --check,
reiserfsck --rebuild-tree, debugfs, fdisk, and mkreiserfs just to see
what info or fixes I might could produce.  Well, there was no
superblocks available at all.  It was like the filesystem had never been
completed or the superblock info had been kept on a ramdisk location. 
Anyway, I decided to reinstall and elect to format the partitions. 
Lot's of bandwidth and time wasted on this go'round.

I wish I could get more info on Yast 1 and 2 centered around ftp based
installs.  There just isn't alot of info on how it works.  Also, last
night the gatech ftp server was not available to me and I couldn't find
a fast mirror.  Anyone know of a fast mirror of SuSE that is reliable? 
Tux.org and rpmfind.net both crashed at the end of the initial ramdisk
download for setup of the ftp installation process.  Twoguys.org is gone
until new sponsors arrive.  So what can we do?
Dow

-- 
__________________________________________________________
Dow Hurst                   Office: 770-499-3428
Systems Support Specialist  Fax:    770-423-6744
1000 Chastain Rd.
Chemistry Department SC428  Email:dhurst at kennesaw.edu
Kennesaw State University         Dow.Hurst at mindspring.com
Kennesaw, GA 30144
*********************************
*Computational Chemistry is fun!*
*********************************

---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.






More information about the Ale mailing list