ufsdump/ufsrestore Problems

ufsdump/ufsrestore Problems

Post by Claus Oberste-Brandenbur » Tue, 27 Feb 1996 04:00:00

I am using ufsdump/ufsrestore to backup the filesystems
under Solaris 2.4 for SPARC. I use HP-DAT Drives C1533A DDS-2
and DDS-Tapes (90m and 120m). To do a backup over the network,
I use the following command:

rsh tmx ufsdump 0cfsd tmq:/dev/rmt/1n 15000 15000 /export/home/userdir


tmx - System where userdir is physically located
tmq - System with Streamer
/export/home/userdir - Directory to backup

This is done for each user.

I have two problems:

When doing a restore using

ufsrestore ivf /dev/rmt/1n ,

I get the message

Verify volume and initialize maps
Record size (512) is not a multiple of dump block size (1024)

This occurs only on some tapes a does not seem to be related to
the type of tape used.

It was possible to position the tape (mt -f /dev/rmt/1n asf XXX) and
to restore the data by doing:

dd if=/dev/rmt/1n of=tmpfile
ufsrestore ivf tmpfile

Well, finally I was happy...


Doing the backup on one certain host (tmp) , I get the following output
from ufsdump:

  DUMP: Writing 63 Kilobyte records
  DUMP: Date of this level 0 dump: Mon Feb 26 15:17:44 1996
  DUMP: Date of last level 0 dump: the epoch
  DUMP: Dumping /dev/rdsk/c0t1d0s7 (tmp:/export) to tmq:/dev/null.
  DUMP: Mapping (Pass I) [regular files]
  DUMP: Mapping (Pass II) [directories]
  DUMP: Estimated 842018 blocks (411.14MB) on 0.02 tapes.
  DUMP: Dumping (Pass III) [directories]
  DUMP: Dumping (Pass IV) [regular files]
  DUMP: Lost connection to remote host.

The backup-command nerver returns and therefore the other userareas
are not saved.

I hope somebody can help

  Claus Oberste-Brandenburg


www  : http://www.tm.bi.ruhr-uni-bochum.de/personal/cob.html


1. ufsdump, ufsrestore problem

I use the command " ufsdump 0nuf /dev/rmt/0n $i" where $i is a list of
all my ufs file systems; this works fine and I can restore using the
command " ufsrestore -ivfs /dev/rmt/0 $i" where $i is the integer tape
file number.

The problem occurs when i change the ufsdump command to add a block size
ufsdump 0nufb /dev/rmt/0n 512 $i.  This works also; however I cannot
recover files from the tape dump

I try using ufsrestore and will get the following error message
media read error: not enough space

I am backing up to a 4mm tape.

the changing to a block size of 512 instead of using the default speeds
up the dump by 20-25% so this could be useful.

Thanks for any help

Sent via Deja.com http://www.deja.com/
Before you buy.

2. file name pattern expansion in Korn shell (88)

3. remote ufsdump/ufsrestore problem

4. dhcpd and two network interfaces

5. ufsdump / ufsrestore problem

6. Configuring Apache as a Proxy server

7. Creating mirrored machine with "ufsdump and ufsrestore"

8. scp 3G file on 125MB/s network slow at 25MB/s why?


10. 1 * NAqqIsWqA-UFSDUMP and UFSRESTORE to a new disk

11. ufsdump and ufsrestore to another system

12. logging the contents of a ufsdump with ufsrestore

13. ufsrestore/ufsdump on Solaris Sparc/x86 incompatibility