I'm passing along a question from the Lawndale network.. they have a few
nodes that have "resisted" upgrades in the past and as a consequence are
running older versions. I'm not sure if an interrupted upgrade or
incorrect disk geometry or something else is to blame. These nodes are
not trivial to access to just re-flash (net4526 boards).
For example, on one node running CUWiN rev 3664:
# mount
/dev/wd0a on / type ffs (read-only, noatime, local)
tmpfs on /dev type tmpfs (local)
tmpfs on /mfs type tmpfs (local)
/etc on /permanent/etc type null (local)
/home on /permanent/home type null (local)
/tmp on /permanent/tmp type null (local)
/var on /permanent/var type null (local)
/mfs/etc on /etc type null (local)
/mfs/home on /home type null (local)
/mfs/tmp on /tmp type null (local)
/mfs/var on /var type null (local)
# upgrade -C ...
upgrade: upgrading the boot partition is dangerous; use -f
I was told '-f' had already been tried and failed so I threw caution to
the wind and:
# upgrade -f -C ...
Disk: /dev/rwd0d
NetBSD disklabel disk geometry:
cylinders: 977, heads: 4, sectors/track: 32 (128 sectors/cylinder)
total sectors: 125056
BIOS disk geometry:
cylinders: 977, heads: 4, sectors/track: 32 (128 sectors/cylinder)
total sectors: 727412841289204
Partition 0:
<UNUSED>
Making partition 0 active.
Preparing for upgrade on /dev/wd0e.
newfs: /dev/rwd0e: open for read: Device not configured
Upgrade failed on newfs /dev/rwd0e [1]
/sbin/upgrade: tmpdir: parameter not set
And this can't be good:
# fdisk -vv
Disk: /dev/rwd0d
NetBSD disklabel disk geometry:
cylinders: 977, heads: 4, sectors/track: 32 (128 sectors/cylinder)
total sectors: 125056
BIOS disk geometry:
cylinders: 977, heads: 4, sectors/track: 32 (128 sectors/cylinder)
total sectors: 727412841289204
Partition table:
0: <UNUSED> (sysid 0)
start 0, size 0, Active
beg: cylinder 0, head 0, sector 0
end: cylinder 0, head 0, sector 0
1: <UNUSED> (sysid 0)
start 0, size 0
beg: cylinder 0, head 0, sector 0
end: cylinder 0, head 0, sector 0
2: <UNUSED> (sysid 0)
start 0, size 0
beg: cylinder 0, head 0, sector 0
end: cylinder 0, head 0, sector 0
3: <UNUSED> (sysid 0)
start 0, size 0
beg: cylinder 0, head 0, sector 0
end: cylinder 0, head 0, sector 0
So the partition table looks hosed. Is there a recipe to follow that
could repair this node without having to haul out the big ladder?
Bill
--
Bill Comisky
bcomisky at pobox.com