UNIX for Dummies Questions & Answers

I have a backup script that is suppose to run every night of the week, but it never does, it starts to run. The way I can tell is that the dates on dump_log are changing to when the script runs. If I look in the cron log file is shows
! > root 506342 cr /etc/back_up Fri Aug 5 10:30:00 2005
! < root 506342 cr /etc/back_up rc=2 Fri Aug 5 10:30:00 2005

I have not been able to find anything except if cron exits with 0 it was successful and if >0 it was unsuccessful. Also if I just run the script manually ./etc/back_up it works fine. Any thoughts? Thanks

cron scripts do not have much of an environment - PATH etc may not be defined.


Assume that all references to things like "echo", if they are not shell builtins, must have a path - "/usr/bin/echo" or whatever. Plus, you have to shebang (#!/bin/ksh) on the first line of the script to get the shell you want - korn, etc. Otherwise you will get sh, your system's version of the Bourne shell.

I looked at the env. and it seems to me that every thing is in the PATH, This is a new box, and I checked out the old box and everything seems to be the same. Here is the path and the script, It does work fine if you run it manually. It seems like it is choking on the /sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk0a >>&/etc/dump_log because first it never writes to the file and if I do a ps -ef | grep dump after the scheduled time nothing shows up.


PATH=/sbin:/usr/sbin:/usr/bin:/usr/bin/X11:/admin/wp51/wpbin/:/admin/wp51/shbin:/usr/ud/bin:/usr/local:/usr/udms:/usr/udms/etc




#! /bin/csh
# /etc/back_up 7-25-05
# Use restore -ivf /dev/tape/tape0_db0 -s x (x = volume to be restored from)
# for file by file restores.
# Remember to rewind tape between restores when restoring different
# filesystems(volumes).
# Also, check fs using df to make sure tape does not fill up.
# Check /etc/backup_date for start and stop times and
# /etc/dump_log for play by play dump status info
#
echo > /etc/backup_date
echo > /etc/dump_log
#
echo starting full backup at `date` >> /etc/backup_date
#
#
echo starting dump of: / filesystem at `date` >> /etc/backup_date
/sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk0a >>&/etc/dump_log
echo dump of: / finished at `date` >> /etc/backup_date
#
echo starting dump of: /usr filesystem `date` >> /etc/backup_date
/sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk0g >>&/etc/dump_log
echo dump of: /usr finished at `date` >> /etc/backup_date
#
echo starting dump of: /db1 filesystem `date` >> /etc/backup_date
/sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk1b >>&/etc/dump_log
echo dump of: /db1 finished at `date` >> /etc/backup_date
#
echo starting dump of: /db2 filesystem `date` >> /etc/backup_date
/sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk2b >>&/etc/dump_log
echo dump of: /db2 finished at `date` >> /etc/backup_date
#
echo starting dump of: /db3 filesystem `date` >> /etc/backup_date
/sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk3b >>&/etc/dump_log
echo dump of: /db3 finished at `date` >> /etc/backup_date
#
echo starting dump of: /unidata filesystem `date` >> /etc/backup_date
/sbin/dump 0fsu /dev/ntape/tape0 1000000 /dev/disk/dsk4b >>&/etc/dump_log
echo dump of: /unidata finished at `date` >> /etc/backup_date
#
# Notify sytems administrators
mailx -s "dump done: admin4" admin@admin.com < /etc/dumpdates