Subject: dump_lfs problems
To: NetBSD Users <netbsd-users@netbsd.org>
From: Louis Guillaume <lguillaume@berklee.edu>
List: netbsd-users
Date: 04/10/2007 01:30:58
I recently set up 2 LFS filesystems for /home and /var/Media. The latter
stores music, pictures etc. Here's the environment:

NetBSD 3.1_STABLE (GENERIC) #2: Sun Jan 14 16:48:08 EST 2007

st0 at scsibus0 target 1 lun 0: <IBM, ULT3580-TD2, 38D0> tape removable
st0: density code 66, variable blocks, write-enabled
st0: sync (50.00ns offset 8), 16-bit (40.000MB/s) transfers


LFS Filesystems:

/dev/raid1a   9.4G     3.5G      5.0G    40%    /home
/dev/raid1b    95G      15G       71G    17%    /var/Media


Now I'm trying to back them up by first taking a snapshot using
fssconfig, then backing up the snapshot.

The first time through it worked, but there was some funny reporting of
of dump's progress...

# fssconfig fss0 /home /tmp/fss-home
# dump_lfs -L maat-home -0eautf /dev/nrst0 /dev/fss0
  [00:48:29 EDT] DUMP: Date of this level 0 dump: Tue Apr 10 00:48:29 2007
  [00:48:29 EDT] DUMP: Date of last level 0 dump: the epoch
  [00:48:29 EDT] DUMP: Dumping /dev/fss0 (an unlisted file system) to
/dev/nrst0
  [00:48:29 EDT] DUMP: Label: maat-home
  [00:48:29 EDT] DUMP: mapping (Pass I) [regular files]
  [00:49:51 EDT] DUMP: mapping (Pass II) [directories]
  [00:49:51 EDT] DUMP: estimated 1899382 tape blocks.
  [00:49:51 EDT] DUMP: Volume 1 started at: Tue Apr 10 00:49:51 2007
  [00:49:51 EDT] DUMP: dumping (Pass III) [directories]
  [00:50:05 EDT] DUMP: dumping (Pass IV) [regular files]

  [00:54:51 EDT] DUMP: 83.86% done, finished in 0:00 (at 00:55:48 EDT)
  [00:59:51 EDT] DUMP: 174.63% done, finished in 0:-4 (at 00:55:35 EDT)
  [01:01:17 EDT] DUMP: 3700275 tape blocks on 1 volume
  [01:01:17 EDT] DUMP: Volume 1 completed at: Tue Apr 10 01:01:17 2007
  [01:01:17 EDT] DUMP: Volume 1 took 0:11:26
  [01:01:17 EDT] DUMP: Volume 1 transfer rate: 5393 KB/s
  [01:01:17 EDT] DUMP: Date of this level 0 dump: Tue Apr 10 00:48:29 2007
  [01:01:17 EDT] DUMP: Date this dump completed:  Tue Apr 10 01:01:17 2007
  [01:01:17 EDT] DUMP: Average transfer rate: 5393 KB/s
  [01:01:17 EDT] DUMP: level 0 dump on Tue Apr 10 00:48:29 2007
  [01:01:17 EDT] DUMP: Closing /dev/nrst0
  [01:01:20 EDT] DUMP: DUMP IS DONE



Doesn't look terrible, even though it looks like it went over 100%.

But now, for the next filesystem...

# fssconfig -u fss0
# fssconfig fss0 /var/Media /tmp/maat-varmedia
# dump_lfs -L maat-media -0eautf /dev/nrst0 /dev/fss0
Memory fault (core dumped)


This may be useless, but...

# gdb dump_lfs dump_lfs.core
GNU gdb 5.3nb1

This GDB was configured as "i386--netbsdelf"...
Core was generated by `dump_lfs'.
Program terminated with signal 11, Segmentation fault.

warning: current_sos: Can't read pathname for load map: Input/output error

Reading symbols from /lib/libc.so.12...done.
Loaded symbols for /lib/libc.so.12
#0  0x0804cde5 in rawread (blkno=-1980760064,
    buf=0xbfbfc110
"d\rxpeG/`eíZpg'\021`gÍ<pi\006ó`i­\036pjæÕ`k\226:ðlÏñàmv\034ðn¯ÓàoUþðp\217µàq5àðro\227às\025ÂðtOyàtþßpv8\226`vÞÁpx\030x`x¾£pyøZ`z\236\205p{Ø<`|~gp}¸\036`~^Ip\177\230",
size=8192) at /usr/src/sbin/dump/rcache.c:205
205             if (blkno + (size >> dev_bshift) > ufsib->ufs_dsize) {
(gdb) bt
#0  0x0804cde5 in rawread (blkno=-1980760064,
    buf=0xbfbfc110
"d\rxpeG/`eíZpg'\021`gÍ<pi\006ó`i­\036pjæÕ`k\226:ðlÏñàmv\034ðn¯ÓàoUþðp\217µàq5àðro\227às\025ÂðtOyàtþßpv8\226`vÞÁpx\030x`x¾£pyøZ`z\236\205p{Ø<`|~gp}¸\036`~^Ip\177\230",
size=8192) at /usr/src/sbin/dump/rcache.c:205
#1  0x08051c40 in fs_read_sblock (superblock=0x8076720 "b\001\a")
    at /usr/src/sbin/dump_lfs/lfs_inode.c:107
#2  0x0804aa67 in main (argc=1, argv=0xbfbfeb58)
    at /usr/src/sbin/dump/main.c:449
#3  0x08049a16 in ___start ()
(gdb) quit





Any ideas???

Thanks

Louis