*BSD News Article 82327


Return to BSD News archive

#! rnews 4646 bsd
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!spool.mu.edu!newspump.sol.net!howland.erols.net!math.ohio-state.edu!cs.utexas.edu!atlantis.utmb.edu!news
From: "Ernest Hymel" <ehymel@utmb.edu>
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: URGENT HELP NEEDED! Can't mount disk any longer!
Date: 4 Nov 1996 23:16:19 GMT
Organization: University of Texas Medical Branch
Lines: 96
Message-ID: <01bbcaa6$222d8110$27366d81@masscomp>
References: <01bbca72$45b912e0$27366d81@masscomp>
NNTP-Posting-Host: masscomp.utmb.edu
X-Newsreader: Microsoft Internet News 4.70.1155

OK, some more info on the problem I detailed below. Again, what I need to
know is whether there is anyway of correcting the problem (pfdisk??)
without losing everything on the disk.

I think I've narrowed the problem down to my partition table getting hosed.
I *have* run pfdisk on the disk in question before it was 'broken', and it
complained about some settings needing to be different, but I ignored this
since #1 the disk has been working for me for about a year now, and #2 I
think you can ignore these messages as long as FreeBSD is the only os on
the disk (it is).

Anyway, I ran pfdisk again, and now it looks *different* than it did
before. Here I see where the number '50000' is coming from-- the partition
table. I don't know how it got changed/hosed from its previous settings.
 Sorry, I don't have the old output, but I do have what it says now:

----------------------------------------------------------------------------
------------------
#Partition tabel on device: 1
geometry 1020 67 63 (cyls heads sectors)
#   ID   First(cyl)     Last(cyl)   Name    # start, length (sectors)
1    0         0                    0            empty    # 0 , 0
2    0         0                    0            empty    # 0 , 0
3    0         0                    0            empty    # 0 , 0
4  165       0                    0            unkno    # 0 , 50000
#note:    last(4): phys=(1023,255,63) logical=(11,56,41)
#note:   first(4): phys=(0,0,1) should be (0,1,1)
#note:   last(4): phys=(1023,255,63) should be (1023,66,63)
active: 0(none)
----------------------------------------------------------------------------
-------
I also have info from disklabel -r from the disk:


bytes/sector 512
sectors/track 32
tracks/cylinder 64
sectors/cylinder 2048
cylinders 2103
sectors/unit 4308352
rpm 5400

8 partitions
	size	offset	fstype	[fsize	bsize	bps/cpg]
a:         65536	         0	4.2BSD	  1024	8192	16	    (cyl 0-31)
b:        131072      65536	swap				    (cyl 32-95)
c:       4308352	         0     4.2BSD   1024	8192	16	    (cyl 0-2031)
d:       4308352	         0	unused	         0           0		    (cyl 0-2031)
g:         327680   196608	4.2BSD   1024	8192	16	    (cyl 96-255)
h:       3784064   524288	4.2BSD   1024	8192	16	    (cyl 256-2031)
----------------------------------------------------------------------------
------------------------------

Finally, FreeBSD 2.1.0 is on the disk in question, and I have FreeBSD 2.1.5
on another disk working. What I ultimately need to do is get the info from
the 2.1.0 disk to the new 2.1.5 disk.

Thanks in advance for any help. If naming my first born after you will
help, I'll do it!


Ernest Hymel <ehymel@utmb.edu> wrote in article
<01bbca72$45b912e0$27366d81@masscomp>...
> HELP!
> 
> What could cause disklabel -r to correctly see the disk geometry, but
when
> I try to boot to the same disk, or mount that disk from another FreeBSD
> disk, I get all kinds of messages like:
> 
> sd1: raw partition size != slice size
> sd1: start 0, end 49999, size 50000
> sd1c: start 0, end 430835, size 4308352
> sd1: truncating raw partition
> sd1: rejecting partition in BSD label: it isn't entirely in the slice
> sd1: start 0, end 49999, size 50000
> sd1a: start 0, end 65535, size 65536
> 
> etc. for my remaining partitions sd1b (swap), sd1c (entire disk), etc.
> FreeBSD occupies the entire disk.
> 
> What's strange in the above message is the second line: there is no
> partition definded by disklabel of size 50000. So for some reason the
> kernel copy of the disklabel is not matching the disk copy of the
> disklabel. How could this be?? Please tell me how to correct this error
> *without* losing the information on that disk! I have backups of the
> critical data, but rebuilding the entire disk will be a logistical
> nightmare! This problem has come on suddenly, where I've been
> booting/mounting this disk for several weeks now.
> 
> Please email if you post. Thank you thank you thank you if you can help.
> I'm deparate!
> -- 
> Ernest Hymel
> ehymel@utmb.edu
>