A good idea is to copy the whole corpus to hard disk -- this also simplifies the following checks. If that is not possible (because of the size of the corpus), try to include all the following tests into one script that will then be run over all media of the corpus (e.g. by mounting one CDROM after the other). That way you minimize the handling of CDROMs and tapes to a minimum.
#
# Frame to check a large number of CDROMs directly
# In this case 32 volumes of a speech corpus
#
set LOGFILE = Logfile.txt
set VOLCNT = 32
set volnr = 1
if ( ! -e $LOGFILE ) touch $LOGFILE
echo "" >> $LOGFILE
echo "Start validation script at: >> $LOGFILE
date >> $LOGFILE
echo "" >> $LOGFILE
umount /cdrom
while ( $volnr <= $VOLCNT )
  echo ""
  echo "Insert the next CDROM number $volnr and hit RETURN"
  set inp = $<
  mount /cdrom
  if ( $status != 0 ) then 
    echo "ERROR: cannot mount CDROM number $volnr - skipping \
        checks" >> $LOGFILE
  else
    echo "CDROM $volnr mounted successfully" >> $LOGFILE  
    #
    # Add the checks per volume here
    #
    (
       ...
       ...
    ) >> $LOGFILE
    #
    #
    #
  endif
  @ volnr ++
end
umount /cdrom
echo "" >> $LOGFILE
echo "End validation script at: >> $LOGFILE
date >> $LOGFILE
echo "" >> $LOGFILE