Emilio Perea wrote:
> I ran into a problem when rebooting to a current kernel (i386 GENERIC)
> due to a secondary disk without an 'a' partition.
I don't think the lack of an 'a' partition is your problem. Goodness
knows, I've got a lot of machines with no 'a' partition on the second
and later disks.
> Disk sd0 checked out
> fine, but all the partitions on sd1 had bad magic numbers and failed
> fsck:
>
> /dev/rsd1d: BAD SUPER BLOCK: MAGIC NUMBER WRONG
> /dev/rsd1d: UNEXPECTED INCONSISTENCY; RUN fsck_ffs MANUALLY.
> ...
> /dev/rsd1n: BAD SUPER BLOCK: MAGIC NUMBER WRONG
> /dev/rsd1n: UNEXPECTED INCONSISTENCY; RUN fsck_ffs MANUALLY.
>
> "Old" disklabel sd1:
>
...
> 15 partitions:
> # size offset fstype [fsize bsize cpg]
> c: 71681967 63 unused 0 0 # Cyl 0*-
> 4461
huh?
c should be the entire disk. There shouldn't be an offset there.
I'm not sure that is your problem, but that doesn't look right at all.
Messing with the 'c' partition is going to break things.
> d: 2104452 63 4.2BSD 2048 16384 132 # Cyl 0*-
> 130
> e: 8385930 2104515 4.2BSD 2048 16384 328 # Cyl 131 -
> 652
> f: 23294250 48387780 4.2BSD 2048 16384 328 # Cyl 3012 -
> 4461
> h: 4112640 15936480 4.2BSD 2048 16384 256 # Cyl 992 -
> 1247
> i: 2104515 40933620 4.2BSD 2048 16384 132 # Cyl 2548 -
> 2678
> j: 18828180 20049120 4.2BSD 2048 16384 328 # Cyl 1248 -
> 2419
> k: 5349645 43038135 4.2BSD 2048 16384 16 # Cyl 2679 -
> 3011
> l: 2056320 38877300 4.2BSD 2048 16384 128 # Cyl 2420 -
> 2547
> m: 2104515 10490445 4.2BSD 2048 16384 132 # Cyl 653 -
> 783
> n: 3341520 12594960 4.2BSD 2048 16384 208 # Cyl 784 -
> 991
>
> "New" disklabel sd1:
new? old? I'm not following that...
...
> 16 partitions:
> # size offset fstype [fsize bsize cpg]
> c: 71687370 0 unused 0 0 # Cyl 0 -
> 19065*
> d: 2097000 34314840 4.2BSD 1024 8192 16 # Cyl 9126*-
> 9683
> e: 1049040 36411840 4.2BSD 1024 8192 16 # Cyl 9684 -
> 9962
> f: 4196160 37460880 4.2BSD 1024 8192 16 # Cyl 9963 -
> 11078
> g: 4196160 41657040 4.2BSD 1024 8192 16 # Cyl 11079 -
> 12194
> h: 8388560 45853200 4.2BSD 1024 8192 16 # Cyl 12195 -
> 14425
> i: 530082 63 ext2fs # Cyl 0*-
> 140*
> j: 1060290 16466625 unknown # Cyl 4379*-
> 4661*
> k: 16787925 17526915 ext2fs # Cyl 4661*-
> 9126*
> l: 15936480 530145 ext2fs # Cyl 140*-
> 4379*
>
That's more like it...except you have a lot of partitions crossing
cylinder boundaries. That's not a problem, but it makes checking for
overlapping partitions more difficult. They may not grossly overlap,
but I didn't look for a "few sector" overlaps...which would really ruin
your day if they were there.
...
>
> I assume this is due to using the new kernel with the old fsck and
> that installing the next snapshot will fix it. If this is unexpected,
> please let me know if you want additional information.
>
> Last dmesg, just in case...
>
> OpenBSD 4.0-current (GENERIC) #1141: Sun Oct 8 13:54:04 MDT 2006
> [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC
> cpu0: Intel(R) Pentium(R) 4 CPU 1500MHz ("GenuineIntel" 686-class) 1.50 GHz
gotta love a dmesg. :)
However, I'm confused by what you are showing me:
a problem including an 'n' partition.
a "old", misconfigured drive with an 'n' partition
a "new", seemingly properly configured drive without the 'n' partition
Looks like your drive geometry changed between "old" and "new". I'm
curious about why. Usually, that means you changed controllers.
So...if the problem is with the first drive configuration, I'd try again
with a proper 'c' partition. Otherwise..I'm confused...which isn't to
say I'm not missing something.
Nick.
