[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
| Title: | DIGITAL UNIX (FORMERLY KNOWN AS DEC OSF/1) |
| Notice: | Welcome to the Digital UNIX Conference |
| Moderator: | SMURF::DENHAM |
|
| Created: | Thu Mar 16 1995 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 10068 |
| Total number of notes: | 35879 |
Customer made a backup of directories using cpio on 3.2g
Then upgraded to 4.0.
He is claiming corruption on restores,
I asked him to use the /usr/opt/obsolete/sbin/cpio
which appears to not worked since he had already restored..
but wanted to see if this likely will resolve what he is
seeing... planing on recommending to try -c and also, to
remove a file so the opbsolte version will replace...
sidjohnson
======================
I attempted to restore the files
using the command:
cpio -iBvda -I/dev/rmt0h
My idea was that, without the u option, it would restore files that were
deleted, but leave others that had not been deleted alone. Now, some of
the files that had not been deleted are scrambled with what would seem to
be a random bit error. For example, in a fortran program, I have a line
that read:
ENDIF
in the original file. After the above cpio was executed the first time,
the line read:
ENTIF
After the line was executed another time, it was back to reading correctly,
but the file had another, similar error in it. Also, executable files now
caused core dumps. This did not happen to all files, so it is difficult to
reproduce. It happened only to some of the files. Hence, the backup did
not work as expected.
/usr/opt/obsolete/sbin/cpio -iBvad -I/dev/rmt0h gives:
current </users/mshouche/paper2/envprog/fig4.hpg> newer or same age
.......repeats this type message for several files.....and ends with...
..
current </users/mshouche/optmd/dataint.f> newer or same age
Out of phase--get help
#
current </users/mshouche/optmd/dataint.f> newer or same age
Out of phase--get help
#
[Posted by WWW Notes gateway]
| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 9625.1 | corruption | RUSURE::KATZ | | Wed Apr 30 1997 00:22 | 13 |
| If I understand correctly, the repeated use of cpio
can cause different results ENDIF vs ENTIF. This sounds
like disk corruption or memory corruption (between
the time it is written into the memory buffer, and
the time it is written out to disk). I wouldn't expect
another version of cpio to do any better. Instead I'd
suspect that more programs than just cpio will malfunction.
Check your error log and if nothing else, take some downtime
and start checking the drive that has been getting corrupted.
(drive,adapter,bus connections all should be suspected).
Hope this helps,
|