[foxboro] Fw: ufsdump problem

Thanks Stan,This is the script.I have cut short the header details.let me 
know if i have missed out or there is a mistake in our script.Not sure why 
/usr files are not backed.The problem started when we need some latest 
/usr/a1 (display) files,but tried to get the files from the backup tape 
but couldn't locate.
Let me know what is the difference between OPTIONS='0ucbdf 96 54000' and 
OPTION='0ucbdf 20 54000 13000'
#!/bin/sh
# Program name : /opt/foxind/tools/backup/backup 
# Function     : To perform a backup of all critical user files on all
#                ap's and aw's to a single tape on CCAW12.  Each file 
system is 
#                saved as a separate volume on the same tape
DMP=/usr/sbin/ufsdump
REST=/usr/sbin/ufsrestore
OPTIONS='0ucbdf 96 54000'
REMDEV='ccaw12le1:/dev/nrst29'
BUHOME=/opt/foxind/tools/backup
VIEWDIR=/opt/foxind/Viewer/exe

cd $BUHOME
rm $BUHOME/*dumpdate
rm -r $BUHOME/foxind
/usr/bin/mt -f /dev/rst29 rewind
echo "USER FILE SYSTEM BACKUP STARTED AT `date`" > $BUHOME/bu.log
echo "" >> $BUHOME/bu.log

#do ufsdumps for each machine, create .dmpdate file to do a readback 
#comparison from tape to verify a good copy.
for ap in CCAW11 CCAW12 A1AW11 B1AW11 A2AW11 OMAW11
do
   rsh $ap echo "$ap User files backed up at `date`" ">" 
/opt/foxind/tools/"$ap".dumpdate
   rsh $ap $DMP $OPTIONS $REMDEV /opt/foxind /opt/fox/ciocfg
   rcp $ap:/opt/foxind/tools/"$ap".dumpdate $BUHOME
done

#Append dump of display files from CCAW11
rsh CCAW11 $DMP $OPTIONS $REMDEV /usr/a1 /usr/a2 /usr/b1 /usr/b2

#Append dump of display files from CCAW12
rsh CCAW12 $DMP $OPTIONS $REMDEV /usr/a1 /usr/a2 /usr/b1 /usr/b2

#Append dmc files from A1AW11, A2AW11, B1AW11
for ap in A1AW11 A2AW11 B1AW11
do
        rsh $ap $DMP $OPTIONS $REMDEV /opt/aos /opt/dmcplus /opt/cimio
done

#Append dump of tis and blender files from OMAW11
rsh OMAW11 $DMP $OPTIONS $REMDEV /opt/tis /opt/gen_libs /opt/ras 
/opt/usrtxt /opt/meslog 

#rewind the tape for the readback check
/usr/bin/mt -f /dev/rst29 rewind

#read back the dumpdate file from all volumes
space_ctr=1 #next volume counter
for ap in CCAW11 CCAW12 A1AW11 B1AW11 A2AW11 OMAW11
do
   echo "1\ny\n" | $REST xfs /dev/rst29 $space_ctr 
./foxind/tools/"$ap".dumpdate 
   space_ctr=`expr $space_ctr + 1`
   echo "READ BACK FILE FOR $ap : \c" >> $BUHOME/bu.log
   cmp $BUHOME/"$ap".dumpdate $BUHOME/foxind/tools/"$ap".dumpdate
   if [ $? -ne 0 ]
   then
      echo FAILED >> $BUHOME/bu.log
   else
      echo OK >> $BUHOME/bu.log
      cat $BUHOME/"$ap".dumpdate >> $BUHOME/bu.log
   fi
done


$VIEWDIR/Viewer $BUHOME/bu.log -Wx large -Ws 800 850 -WP 5 760 -title 
"SYSTEM BACKUP REPORT"

Thanks and best Regards 
Ajit 
----- Forwarded by Ajit S Kumar/Technical/SAMREF/SA on 31/03/2008 07:02 PM 
-----

stan <stanb@xxxxxxxxx> 
Sent by: foxboro-bounce@xxxxxxxxxxxxx
31/03/2008 06:53 PM
Please respond to
foxboro@xxxxxxxxxxxxx


To
Ajit.S.Kumar@xxxxxxxxxxxxx
cc
foxboro@xxxxxxxxxxxxx
Subject
Re: [foxboro] ufsdump problem






On Mon, Mar 31, 2008 at 06:23:24PM +0300, Ajit.S.Kumar@xxxxxxxxxxxxx 
wrote:
> Hi List,
> We have a problem with ufsdump script. We take backup of some critical 
> files weekly thru cron.
> But lately seen not all the files are backed up as specified in the 
> script.
> Can anybody tell me whats the difference between ufsdump options 
> /arguments.
> Earlier we used ufsdump 0ucbdf 20 54000 13000 
> and took backup on 5GB tape drive with 2GB Tape and it was working fine.
> Problem started  when using 12GB Tapedrive with 12GBTape.
> The ufsdump option was changed to 0ucbdf 96 54000 and now it's taking 
only 
> /opt and /foxind files only.
> All the additional files in /usr are skipped.
> Please assist and thanks in advance. if required I can send the complete 

> script.

I don't lnow what your script looks like, but i would geuss that someone
made an error in editing it. ufsdump does _filesystems_ so I would suppose
you invoke it once for each filesystem you want to backup. I will be that
you will find that it is not being invoked corectly on the /isr partion
(filesystem). Run it from the command line and llok for error messages.
Cron is likely set up to discrad its standard out, and error out, thus
masking the problem.


-- 
One of the main causes of the fall of the roman empire was that, lacking
zero, they had no way to indicate successful termination of their C
programs.
 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
 


 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts: