Hi, thanks for your reply. I am ok with this font size. i.e - $graph->set_values_font($font, 8). My graph is PERFECT now. I thank you again for your wonderful support. Anyway, I would like to learn below codes since I want to draw in that way. because It's a very perl way. Could you pls explain below codes.

my %count=();  # A hash -it is bcos of of key/value pairs?

while (<$fh>){
  if (/amavis.*(Passed|Blocked) (CLEAN|SPAM|INFECTED|BANNED)/){
    ++$count{$2}; # Why ++$count? why {$2} here? is it bcos of (Passed
+|Blocked) ?
  } elsif (/amavis.*(BAD-HEADER)-/) {
    ++$count{$1}; # why {$1} here? is it bcos of (BAD-HEADER)
  } elsif (/reject/){
    ++$count{'REJECTED'}; #Hmm, Why {'REJECTED'}?
  }
}
[download]

Now I want to learn next code

my @data = map{ $count{$_} || 0 } qw(CLEAN SPAM INFECTED BANNED BAD-HE
+ADER REJECTED);

Why map{ $count{$_} || 0 } ? 

qw(CLEAN SPAM INFECTED BANNED BAD-HEADER REJECTED) # here you create a
+n array for count values?
[download]

my $data = GD::Graph::Data->new([
    ["Cleaned","Spam","Virus","Banned","BadHeader","Rejected"],
    [@data],
]) or die GD::Graph::Data->error; 

[@data] has count values coming from  line qw(CLEAN SPAM INFECTED BANN
+ED BAD-HEADER REJECTED)
[download]

Could you pls explain me these, Since I PREFER this method and I can learn a lot of perl as well.

my %count=();  # A hash -it is bcos of of key/value pairs?

Yes, this declared a hash to store the results of counting the lines. The key will be one of either 'CLEAN','SPAM','INFECTED','BANNED','BAD_HEADER','REJECTED' and the value is the count.

while (<$fh>){

Read each line of the file into $_

  if (/amavis.*(Passed|Blocked) (CLEAN|SPAM|INFECTED|BANNED)/){

Match the line read (in $_) against the regex pattern like grep. The first word captured (Passed or Block) will be in $1 and the second (CLEAN or SPAM or INFECTED or BANNED) will be in $2

    ++$count{$2}; # Why ++$count? why {$2} here? is it bcos of (Passed
+|Blocked) ?
[download]

For example, if $2 is the word CLEAN increment the hash $count{'CLEAN'}

  } elsif (/amavis.*(BAD-HEADER)-/) {
    ++$count{$1}; # why {$1} here? is it bcos of (BAD-HEADER)
[download]

Yes, this regex has only 1 set of brackets so if it matches the word BAD-HEADER will be in $1

  } elsif (/reject/){
    ++$count{'REJECTED'}; #Hmm, Why {'REJECTED'}?
  }
}
[download]

I chose REJECTED because 'rejected' was the label in your graph and all the other keys are upper case

my @data = map{ $count{$_} || 0 } qw(CLEAN SPAM INFECTED BANNED BAD-HEADER REJECTED);

map is a convenient way of creating an array from another another array. It is the same as

my @data=();
$data[0] = $count{'CLEAN'} || 0; # use 0 if $count{'CLEAN'} is undefin
+ed 
$data[1] = $count{'SPAM'} || 0;
$data[2] = $count{'INFECTED'} || 0;
$data[3] = $count{'BANNED'} || 0;
$data[4] = $count{'BAD-HEADER'} || 0;
$data[5] = $count{'REJECTED'} || 0;
[download]

poj

Thanks for your reply. I realized the code due to your kind help. my last code was Global mail stats. It has BOTH Inbound and Outbound mails. Now I want to write separate codes for Inbound and Outbound mails. I wrote below code for Inbound mails. It works. I hope it is OK. If there are any problems with my code, Pls enlighten me. Below code with grep Is what I want.

#Passed CLEAN - Clean Mails

my $clean_total = qx(sudo grep -Ec 'amavis.*Passed CLEAN.*RelayedInbou
+nd' /var/log/maillog);
chomp($clean_total);
#$clean_total =~ s/^\s+|\s+$//g;

#Blocked SPAM - Spam Mails

my $spam_total = qx(sudo grep -Ec 'amavis.*Blocked SPAM.*DiscardedInbo
+und' /var/log/maillog);
chomp($spam_total);
#$spam_total =~ s/^\s+|\s+$//g;

#Blocked INFECTED - Virus Mails

my $virus_total = qx(sudo grep -Ec 'amavis.*Blocked INFECTED.*Discarde
+dInbound' /var/log/maillog);
chomp($virus_total);
#$virus_total =~ s/^\s+|\s+$//g;

#Blocked BANNED - Banned Mails exe,scr etc.

my $banned_total = qx(sudo grep -Ec 'amavis.*Blocked BANNED.*Discarded
+Inbound' /var/log/maillog); 
chomp($banned_total);
#$banned_total =~ s/^\s+|\s+$//g;

#BAD-HEADER- - Bad Header Mails

my $badheader_total = qx(sudo grep -Ec 'amavis.*BAD-HEADER-.*RelayedIn
+bound' /var/log/maillog);
chomp($badheader_total);
#$badheader_total =~ s/^\s+|\s+$//g;

#REJECT - REJECT Mails

my $reject_total = qx(sudo grep -c 'smtpd.*reject' /var/log/maillog);
chomp($reject_total);
#$reject_total =~ s/^\s+|\s+$//g;
[download]

But, Since it is NOT very professional perl method, I wrote below code. It works well. pls see it and give your comments

#!/usr/bin/perl
# inbound.cgi

use CGI qw(:standard);
#use CGI ':standard';
use strict;
use warnings;
use GD::Graph;
use GD::Graph::bars;
use CGI::Carp 'fatalsToBrowser'; # use only for testing

my $logfile = '/var/log/maillog';
open my $fh, '<', $logfile or die "Could not open $logfile : $!";
my %count=();

while (<$fh>){
  if (/amavis.*(Passed|Blocked) (CLEAN|SPAM|INFECTED|BANNED).*(Relayed
+Inbound|DiscardedInbound)/){
    ++$count{$2};
  } elsif (/amavis.*(BAD-HEADER)-.*RelayedInbound/) {
    ++$count{$1};
  } elsif (/smtpd.*reject/){
    ++$count{'REJECTED'};
  }
}
close $fh;  
my @data = map{ $count{$_} || 0 } qw(CLEAN SPAM INFECTED BANNED BAD-HE
+ADER REJECTED);

my $data = GD::Graph::Data->new([
    ["Cleaned","Spam","Virus","Banned","BadHeader","Rejected"],
    [@data],
]) or die GD::Graph::Data->error; 

my $graph = GD::Graph::bars->new(600,300);
$graph->set( 
    x_label       => 'Category',
    y_label       => 'Count',
    title         => 'Inbound Mail Statistics '.localtime,
    show_values   => 1,
    y_tick_number => 8,
    bar_spacing   => 10,
    shadow_depth  => -4,
    shadowclr     => 'dred',
    transparent   => 0,
) or die $graph->error;

my $font = '/usr/share/fonts/dejavu/DejaVuSans-Bold.ttf';
$graph->set_title_font($font,12);
$graph->set_values_font($font, 8);
$graph->set_x_axis_font($font,8);
$graph->set_y_axis_font($font,8);
$graph->set_x_label_font($font,12);
$graph->set_y_label_font($font,12);

print header('image/png');
$graph->plot($data) or die $graph->error;
binmode STDOUT;
print $graph->gd->png;
[download]

I want to generate both graphs withing page.cgi. So I rearranged it.

#!/usr/bin/perl
# page.cgi
use strict;
use CGI ':standard';

print header();
print start_html(-title  =>'Global Mail Statistics',
                 -bgcolor=>'#95B8DB');
print qq!<br/><center>
<h1>Mail Statistics</h1>
<img src = "bargraph.cgi"></center>!;
print qq!<br/><br/><center>
<img src = "inbound.cgi"></center>!;
print end_html;
[download]

pls advice me if it is OK? Next, I am going to write outbound.cgi. I will let you know if I I have any issues? Sir, Thanks you very much for your meritorious deed. Hope to hear from you.