Your script can be done much simpler if you generalize it.

use File::Copy;
use File::Find;

use strict;
use warnings;

### Parse numerical characters and trailing white-space from rmtdb.lrl
+ for matching.
copy("rmtdb.lrl" , "rmtdb.tmp")
    or die "rmtdb.lrl file cannot be copied: $!\n";
system "cat rmtdb.tmp | cut -d ' ' -f1 > rmtdb.tmp1";

my %lines;

for my $file (qw(dblist.comdbg dblist.varldb rmtdb.tmp1)) {
    open my $fh, $file or die "Can't open $file: $!";
    while (<$fh>) {
        chomp;

        # Special case of rmtdb
        $_ = lc $_ if $file eq 'rmtdb.tmp1';
        
        $lines{$_}{$file}++;
    }
    close $fh;
}

foreach my $line (keys %lines) {
    my @files = sort keys %{$lines{$line}});
    next if @files < 2;
    print "$line is duplicated in: " . join(' ', @files) . "\n";
}

### Cleanup
unlink "rmtdb.tmp", "rmtdb.tmp1";
[download]

Then if you want to process 20 files, you can just add them to the loop

Thank you for the response. I haven't tried this yet but I will shortley. The problem I'm having is first I have to search and grep for each rmtdb.lrl file because each server is different. This is what I'm working with for that purpose.

### Collect and parse rmt(*)db.lrl files.
my @dirs=qw(/bb/bin/jwarfel);
find(\&parse, @dirs);
   sub parse {
      if (/^rmt[1-9]|[1-9][1-9]|db.lrl$/) {
         copy("$_" , "$_.tmp");
         system "cat $_.tmp | cut -d ' ' -f1 > $_.tmp1";
         open $_, "< $_.tmp1";
         @$_ = <$_>; 
         close $_;
         unlink "$_.tmp", "$_.tmp1";
         foreach (@$_) {s/$_/\L$_/gi;}
         $rmtdb="$_";
         push(@rmtdbs , $rmtdb);
      }
   }
[download]

Can you be more specific as to what you're having trouble with? You know how to find duplicates between files. You found the files of interest. I don't understand "matching afterwards".

Repeating the code three times for each pair of files is bad to begin with, and clearly will not scale to larger groups of files! Use a loop. You don't need separate hashes for everything, and I don't see the point of temp files.

Here's how I would do it: For each file, read each line and hash it. Store the hash (not the whole line) as the key to a master hash, with the value being a list of file names it was seen in. So, for each line, push the current file name onto the value of that line's key.

After going through all the files, iterate through each hash entry and note which ones have more than one item in the value.

system "cat rmtdb.tmp | cut -d ' ' -f1 > rmtdb.tmp1";

system "cut -d ' ' -f1 rmtdb.tmp > rmtdb.tmp1";
[download]

Thanks for everyone's advice and help on this. Here was my final code source which works fantastic in DEV so far.

#!/usr/bin/perl

use File::Find;

use strict;
use warnings;

my @dbs;

my @dirs=qw(/Users/jwarfel/Documents/perl);
find(\&parse, @dirs);
   sub parse {
      if (/^rmt[0-9]|[0-9][0-9]|db.lrl$/) {
         my $rmtdb="$_";
         push(@dbs , $rmtdb);
      }
   }
 
foreach (qw(dblist.comdbg dblist.varldb)) {push(@dbs , $_);}

my %lines;

for my $file (@dbs) {
    open my $fh, $file or die "Can't open $file: $!";
    while (<$fh>) {
        chomp;

        # Special case of rmtdb
        $_ = lc $_;
        if ($_ =~ /^(\w+)\s+/) {
            $_ = $1;
        }
        
        $lines{$_}{$file}++;
    }
    close $fh;
}

foreach my $line (keys %lines) {
    my @files = sort keys %{$lines{$line}};
    next if @files < 2;
    print "$line is duplicated in: " . join(' ', @files) . "\n";
}

exit 0;
[download]