mdmfs: newfs exited with error code 36

[Deiter Alexandr Valerievich]

mdmfs don't work on CURRENT diskless sparc64:

Loading configuration files.
Entropy harvesting: interrupts ethernet point_to_point.
Starting file system checks:
nfs: can't update /var/db/mounttab for 1.1.1.1:/sparc64/netboot
cp: utmp: Read-only file system
/etc/rc.d/cleanvar: cannot create /var/run/clean_var: Read-only file system
+++ mount_md of /var
mdmfs: newfs exited with error code 36
nfs: can't update /var/db/mounttab for 1.1.1.1:/sparc64/netboot
.....
dropping into shell, ^D to continue anyway.

# mdmfs -i 4096 -s 32m -M md /var
mdmfs: newfs exited with error code 36
# mdmfs -i 4096 -s 16m -M md /var
# df -h
Filesystem                             Size   Used  Avail Capacity  Mounted
on
1.1.1.1:/sparc64/netboot    16G   3.0G    11G    21%    /
devfs                                  1.0K   1.0K     0B   100%    /dev
/dev/md2                                15M   2.0K    14M     0%    /var
# top
last pid:   124;  load averages:  0.01,  0.05,  0.03  up 0+00:09:58
23:30:01
4 processes:   1 running, 3 sleeping
Mem: 2864K Active, 1544K Inact, 10M Wired, 40K Cache, 3392K Buf, 472M Free
Swap:
  PID USERNAME PRI NICE   SIZE    RES STATE    TIME   WCPU    CPU COMMAND
   93 root       8    0  5440K  1448K wait     0:00  0.00%  0.00% sh
   35 root       8    0  5408K  1408K wait     0:00  0.00%  0.00% sh
  109 root       8    0  5216K  1320K wait     0:00  0.00%  0.00% sh
  124 root      96    0  6808K  1480K RUN      0:00  0.00%  0.00% top

P.S. On current 20030825 diskless operations works fine.

Thanks!