X-Git-Url: https://git.exim.org/exim.git/blobdiff_plain/81aad8c9cc6aa1ad9653966ecfea657056111d57..7eb0e5d2b8453f753bd2d8e2e77cf4b7e0b24b1b:/src/src/eximstats.src diff --git a/src/src/eximstats.src b/src/src/eximstats.src index 015fc5806..727ee44b9 100644 --- a/src/src/eximstats.src +++ b/src/src/eximstats.src @@ -1,7 +1,6 @@ -#!PERL_COMMAND -w -# $Cambridge: exim/src/src/eximstats.src,v 1.9 2005/06/30 09:59:34 steve Exp $ +#!PERL_COMMAND -# Copyright (c) 2001 University of Cambridge. +# Copyright (c) 2001-2016 University of Cambridge. # See the file NOTICE for conditions of use and distribution. # Perl script to generate statistics from one or more Exim log files. @@ -74,7 +73,7 @@ # 2001-10-21 Removed -domain flag and added -bydomain, -byhost, and -byemail. # We now generate our main parsing subroutine as an eval statement # which improves performance dramatically when not all the results -# are required. We also cache the last timestamp to time convertion. +# are required. We also cache the last timestamp to time conversion. # # NOTE: 'Top 50 destinations by (message count|volume)' lines are # now 'Top N (host|email|domain) destinations by (message count|volume)' @@ -143,7 +142,7 @@ # in HTML output. Also added code to convert them back with -merge. # Fixed timestamp offsets to convert to seconds rather than minutes. # Updated -merge to work with output files using timezones. -# Added cacheing to speed up the calculation of timezone offsets. +# Added caching to speed up the calculation of timezone offsets. # # 2003-02-07 V1.25 Steve Campbell # Optimised the usage of mktime() in the seconds subroutine. @@ -163,7 +162,7 @@ # Bernard Massot. # # 2003-06-03 V1.28 John Newman -# Added in the ability to skip over the parsing and evaulation of +# Added in the ability to skip over the parsing and evaluation of # specific transports as passed to eximstats via the new "-nt/.../" # command line argument. This new switch allows the viewing of # not more accurate statistics but more applicable statistics when @@ -201,7 +200,7 @@ # Added -xls and the ability to specify output files. # # 2005-04-29 V1.38 Steve Campbell -# Use FileHandles for outputing results. +# Use FileHandles for outputting results. # Allow any combination of xls, txt, and html output. # Fixed display of large numbers with -nvr option # Fixed merging of reports with empty tables. @@ -226,6 +225,68 @@ # 2005-06-30 V1.43 Steve Campbell # Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear. # +# 2005-07-26 V1.44 Steve Campbell +# Use a glob alias rather than an array ref in the generated +# parser. This improves both readability and performance. +# +# 2005-09-30 V1.45 Marco Gaiarin / Steve Campbell +# Collect SpamAssassin and rejection statistics. +# Don't display local sender or destination tables unless +# there is data to show. +# Added average volumes into the top table text output. +# +# 2006-02-07 V1.46 Steve Campbell +# Collect data on the number of addresses (recipients) +# as well as the number of messages. +# +# 2006-05-05 V1.47 Steve Campbell +# Added 'Message too big' to the list of mail rejection +# reasons (thanks to Marco Gaiarin). +# +# 2006-06-05 V1.48 Steve Campbell +# Mainlog lines which have GMT offsets and are too short to +# have a flag are now skipped. +# +# 2006-11-10 V1.49 Alain Williams +# Added the -emptyok flag. +# +# 2006-11-16 V1.50 Steve Campbell +# Fixes for obtaining the IP address from reject messages. +# +# 2006-11-27 V1.51 Steve Campbell +# Another update for obtaining the IP address from reject messages. +# +# 2006-11-27 V1.52 Steve Campbell +# Tally any reject message containing SpamAssassin. +# +# 2007-01-31 V1.53 Philip Hazel +# Allow for [pid] after date in log lines +# +# 2007-02-14 V1.54 Daniel Tiefnig +# Improved the '($parent) =' pattern match. +# +# 2007-03-19 V1.55 Steve Campbell +# Differentiate between permanent and temporary rejects. +# +# 2007-03-29 V1.56 Jez Hancock +# Fixed some broken HTML links and added missing column headers. +# +# 2007-03-30 V1.57 Steve Campbell +# Fixed Grand Total Summary Domains, Edomains, and Email columns +# for Rejects, Temp Rejects, Ham, and Spam rows. +# +# 2007-04-11 V1.58 Steve Campbell +# Fix to get <> and blackhole to show in edomain tables. +# +# 2007-09-20 V1.59 Steve Campbell +# Added the -bylocaldomain option +# +# 2007-09-20 V1.60 Heiko Schlittermann +# Fix for misinterpreted log lines +# +# 2013-01-14 V1.61 Steve Campbell +# Watch out for senders sending "HELO [IpAddr]" +# # # For documentation on the logfile format, see # http://www.exim.org/exim-html-4.50/doc/html/spec_48.html#IX2793 @@ -329,7 +390,7 @@ Useful for finding out which of your mailing lists are receiving mail. Show the delivery times (B
)for all the messages. -Exim must have been configured to use the +delivery_time logging option +Exim must have been configured to use the +deliver_time logging option for this option to work. I is an optional list of times. Eg -show_dt1,2,4,8 will show @@ -449,6 +510,11 @@ Create the charts in the directory Specify the relative directory for the "img src=" tags from where to include the charts +=item B<-emptyok> + +Specify that it's OK to not find any valid log lines. Without this +we will output an error message if we don't find any. + =item B<-d> Debug flag. This outputs the eval()'d parser onto STDOUT which makes it @@ -475,13 +541,15 @@ mailing list exim-users@exim.org. This program does not perfectly handle messages whose received and delivered log lines are in different files, which can happen when you have multiple mail servers and a message cannot be -immeadiately delivered. Fixing this could be tricky... +immediately delivered. Fixing this could be tricky... Merging of xls files is not (yet) possible. Be free to implement :) =cut +use warnings; use integer; +BEGIN { pop @INC if $INC[-1] eq '.' }; use strict; use IO::File; @@ -518,7 +586,7 @@ use vars qw($WEEK $DAY $HOUR $MINUTE); @days_per_month = (0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334); $gig = 1024 * 1024 * 1024; -$VERSION = '1.43'; +$VERSION = '1.61'; # How much space do we allow for the Hosts/Domains/Emails/Edomains column headers? $COLUMN_WIDTHS = 8; @@ -530,20 +598,29 @@ $WEEK = 7 * $DAY; # Declare global variables. use vars qw($total_received_data $total_received_data_gigs $total_received_count); -use vars qw($total_delivered_data $total_delivered_data_gigs $total_delivered_count); +use vars qw($total_delivered_data $total_delivered_data_gigs $total_delivered_messages $total_delivered_addresses); use vars qw(%timestamp2time); #Hash of timestamp => time. -use vars qw($last_timestamp $last_time); #The last time convertion done. -use vars qw($last_date $date_seconds); #The last date convertion done. -use vars qw($last_offset $offset_seconds); #The last time offset convertion done. +use vars qw($last_timestamp $last_time); #The last time conversion done. +use vars qw($last_date $date_seconds); #The last date conversion done. +use vars qw($last_offset $offset_seconds); #The last time offset conversion done. use vars qw($localtime_offset); use vars qw($i); #General loop counter. use vars qw($debug); #Debug mode? use vars qw($ntopchart); #How many entries should make it into the chart? use vars qw($gddirectory); #Where to put files from GD::Graph -use vars qw($workbook $ws_global $ws_relayed $ws_top50 $ws_errors ); #For use in Speadsheed::WriteExcel -use vars qw($row $col $row_hist $col_hist $row_league_table); + +# SpamAssassin variables +use vars qw($spam_score $spam_score_gigs); +use vars qw($ham_score $ham_score_gigs); +use vars qw(%ham_count_by_ip %spam_count_by_ip); +use vars qw(%rejected_count_by_ip %rejected_count_by_reason); +use vars qw(%temporarily_rejected_count_by_ip %temporarily_rejected_count_by_reason); + +#For use in Spreadsheet::WriteExcel +use vars qw($workbook $ws_global $ws_relayed $ws_errors); +use vars qw($row $col $row_hist $col_hist); use vars qw($run_hist); -use vars qw($f_default $f_header1 $f_header2 $f_headertab $f_percent); #Format Header +use vars qw($f_default $f_header1 $f_header2 $f_header2_m $f_headertab $f_percent); #Format Header # Output FileHandles use vars qw($txt_fh $htm_fh $xls_fh); @@ -553,8 +630,8 @@ $ntopchart = 5; # The following are parameters whose values are # set by command line switches: use vars qw($show_errors $show_relay $show_transport $transport_pattern); -use vars qw($topcount $local_league_table $include_remote_users); -use vars qw($hist_opt $hist_interval $hist_number $volume_rounding); +use vars qw($topcount $local_league_table $include_remote_users $do_local_domain); +use vars qw($hist_opt $hist_interval $hist_number $volume_rounding $emptyOK); use vars qw($relay_pattern @queue_times @user_patterns @user_descriptions); use vars qw(@rcpt_times @delivery_times); use vars qw($include_original_destination); @@ -567,11 +644,12 @@ use vars qw($merge_reports); #Merge old reports ? # The following are modified in the parse() routine, and # referred to in the print_*() routines. use vars qw($delayed_count $relayed_unshown $begin $end); -use vars qw(%messages $message_aref); +use vars qw(%messages @message); use vars qw(%received_count %received_data %received_data_gigs); -use vars qw(%delivered_count %delivered_data %delivered_data_gigs); +use vars qw(%delivered_messages %delivered_data %delivered_data_gigs %delivered_addresses); use vars qw(%received_count_user %received_data_user %received_data_gigs_user); -use vars qw(%delivered_count_user %delivered_data_user %delivered_data_gigs_user); +use vars qw(%delivered_messages_user %delivered_addresses_user %delivered_data_user %delivered_data_gigs_user); +use vars qw(%delivered_messages_local_domain %delivered_addresses_local_domain %delivered_data_local_domain %delivered_data_gigs_local_domain); use vars qw(%transported_count %transported_data %transported_data_gigs); use vars qw(%relayed %errors_count $message_errors); use vars qw(@qt_all_bin @qt_remote_bin); @@ -681,8 +759,8 @@ sub volume_rounded { } else { # We don't want any rounding to be done. - # and we don't need broken formated output which on one hand avoids numbers from - # being interpreted as string by Spreadsheed Calculators, on the other hand + # and we don't need broken formatted output which on one hand avoids numbers from + # being interpreted as string by Spreadsheet Calculators, on the other hand # breaks if more than 4 digits! -> flexible length instead of fixed length # Format the return value at the output routine! -fh #$rounded = sprintf("%d", ($g * $gig) + $x); @@ -795,10 +873,10 @@ $p; # Eg 3h20m5s => 12005 ####################################################################### sub unformat_time { - my($formated_time) = pop @_; + my($formatted_time) = pop @_; my $time = 0; - while ($formated_time =~ s/^(\d+)([wdhms]?)//) { + while ($formatted_time =~ s/^(\d+)([wdhms]?)//) { $time += $1 if ($2 eq '' || $2 eq 's'); $time += $1 * 60 if ($2 eq 'm'); $time += $1 * 60 * 60 if ($2 eq 'h'); @@ -818,6 +896,7 @@ sub unformat_time { # POSIX::mktime. We expect the timestamp to be of the form # "$year-$mon-$day $hour:$min:$sec", with month going from 1 to 12, # and the year to be absolute (we do the necessary conversions). The +# seconds value can be followed by decimals, which we ignore. The # timestamp may be followed with an offset from UTC like "+$hh$mm"; if the # offset is not present, and we have not been told that the log is in UTC # (with the -utc option), then we adjust the time by the current local @@ -841,7 +920,7 @@ sub seconds { # Is the timestamp the same as the last one? return $last_time if ($last_timestamp eq $timestamp); - return 0 unless ($timestamp =~ /^((\d{4})\-(\d\d)-(\d\d))\s(\d\d):(\d\d):(\d\d)( ([+-])(\d\d)(\d\d))?/o); + return 0 unless ($timestamp =~ /^((\d{4})\-(\d\d)-(\d\d))\s(\d\d):(\d\d):(\d\d)(?:\.\d+)?( ([+-])(\d\d)(\d\d))?/o); unless ($last_date eq $1) { $last_date = $1; @@ -852,8 +931,8 @@ sub seconds { } my $time = $date_seconds + ($5 * 3600) + ($6 * 60) + $7; - # SC. Use cacheing. Also note we want seconds not minutes. - #my($this_offset) = ($10 * 60 + $11) * ($9 . "1") if defined $8; + # SC. Use caching. Also note we want seconds not minutes. + #my($this_offset) = ($10 * 60 + $12) * ($9 . "1") if defined $8; if (defined $8 && ($8 ne $last_offset)) { $last_offset = $8; $offset_seconds = ($10 * 60 + $11) * 60; @@ -861,7 +940,7 @@ sub seconds { } - if (defined $7) { + if (defined $8) { #$time -= $this_offset; $time -= $offset_seconds; } elsif (defined $localtime_offset) { @@ -1005,14 +1084,10 @@ my $htm_format = "%s %s%d< printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; if ($htm_fh) { print $htm_fh "

$temp

\n"; - print $htm_fh "\n"; - print $htm_fh "
\n"; - print $htm_fh "\n"; + print $htm_fh "
\n"; print $htm_fh "\n"; } -if ($xls_fh) -{ - +if ($xls_fh) { $ws_global->write($row++, $col, "$title: ".$message_type, $f_header2); my @content=("Time", "Messages", "Percentage", "Cumulative Percentage"); &set_worksheet_line($ws_global, $row++, 1, \@content, $f_headertab); @@ -1082,8 +1157,7 @@ push(@chartdatavals, $overflow); #printf("Unknown %6d\n", $queue_unknown) if $queue_unknown > 0; if ($htm_fh) { - print $htm_fh "
TimeMessagesPercentageCumulative Percentage
\n"; - print $htm_fh "
\n"; + print $htm_fh "
"; if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) { my @data = ( @@ -1133,12 +1207,11 @@ my(@chartdatanames); my(@chartdatavals); my($maxd) = 0; -if (!$run_hist) # save first row of print_histogram for xls output -{ +# save first row of print_histogram for xls output +if (!$run_hist) { $row_hist = $row; } -else -{ +else { $row = $row_hist; } @@ -1169,17 +1242,16 @@ if ($htm_fh) { print $htm_fh "
\n";
 }
 
-if ($xls_fh)
-{
+if ($xls_fh) {
   $title =~ s/Messages/Msg/ ;
+  $row += 2;
   $ws_global->write($row++, $col_hist+1, $title, $f_headertab);
 }
 
 
 my $hour = 0;
 my $minutes = 0;
-for ($i = 0; $i < $hist_number; $i++)
-{
+for ($i = 0; $i < $hist_number; $i++) {
   my $c = $interval_count[$i];
 
   # If the interval is an hour (the maximum) print the starting and
@@ -1187,26 +1259,23 @@ for ($i = 0; $i < $hist_number; $i++)
   # minutes, which take up the same space.
 
   my $temp;
-  if ($hist_opt == 1)
-  {
+  if ($hist_opt == 1) {
     $temp = sprintf("%02d-%02d", $hour, $hour + 1);
 
     print $txt_fh $temp if $txt_fh;
     print $htm_fh $temp if $htm_fh;
 
-    if ($xls_fh)
-    {
-      if ($run_hist==0) # only on first run
-      {
-        &set_worksheet_line($ws_global, $row, 0, [$temp], $f_default);
+    if ($xls_fh) {
+      if ($run_hist==0) {
+        # only on first run
+        $ws_global->write($row, 0, [$temp], $f_default);
       }
     }
 
     push(@chartdatanames, $temp);
     $hour++;
   }
-  else
-  {
+  else {
     if ($minutes == 0)
       { $temp = sprintf("%02d:%02d", $hour, $minutes) }
     else
@@ -1214,28 +1283,24 @@ for ($i = 0; $i < $hist_number; $i++)
 
     print $txt_fh $temp if $txt_fh;
     print $htm_fh $temp if $htm_fh;
-    if (($xls_fh) and ($run_hist==0)) # only on first run
-    {
+    if (($xls_fh) and ($run_hist==0)) {
+      # only on first run
       $temp = sprintf("%02d:%02d", $hour, $minutes);
-      &set_worksheet_line($ws_global, $row, 0, [$temp], $f_default);
+      $ws_global->write($row, 0, [$temp], $f_default);
     }
 
     push(@chartdatanames, $temp);
     $minutes += $hist_interval;
-    if ($minutes >= 60)
-      {
+    if ($minutes >= 60) {
       $minutes = 0;
       $hour++;
-      }
+    }
   }
   push(@chartdatavals, $c);
 
   printf $txt_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $txt_fh;
   printf $htm_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $htm_fh;
-  if ($xls_fh)
-  {
-    &set_worksheet_line($ws_global, $row++, $col_hist+1, [$c], $f_default);
-  }
+  $ws_global->write($row++, $col_hist+1, [$c], $f_default) if $xls_fh;
 
 } #end for
 
@@ -1285,234 +1350,245 @@ $run_hist=1; # we have done this once or more
 #######################################################################
 # print_league_table();
 #
-#  print_league_table($league_table_type,\%message_count,\%message_data,\%message_data_gigs);
+#  print_league_table($league_table_type,\%message_count,\%address_count,\%message_data,\%message_data_gigs, $spreadsheet, $row_sref);
 #
-# Given hashes of message count and message data, which are keyed by
-# the table type (eg by the sending host), print a league table
-# showing the top $topcount (defaults to 50).
+# Given hashes of message count, address count, and message data,
+# which are keyed by the table type (eg by the sending host), print a
+# league table showing the top $topcount (defaults to 50).
 #######################################################################
 sub print_league_table {
-my($text,$m_count,$m_data,$m_data_gigs) = @_;
-my($name) = ($topcount == 1)? "$text" : "$topcount ${text}s";
-my($temp) = "Top $name by message count";
-my(@chartdatanames) = ();
-my(@chartdatavals) = ();
-my $chartotherval = 0;
-
-my $htm_format;
-my $txt_format = "%7d %10s   %s\n";
-
-# write header
-printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh;
-if ($htm_fh) {
-  print $htm_fh "

$temp

\n"; - print $htm_fh "\n"; - print $htm_fh ""; $sender_txt_header .= " " x ($COLUMN_WIDTHS - length($_)) . $_ . 's'; - $sender_html_format .= ""; - $sender_txt_format .= " " x ($COLUMN_WIDTHS - 5) . "%6d"; + $sender_html_format .= ""; + $sender_txt_format .= " " x ($COLUMN_WIDTHS - 5) . "%6s"; + push(@col_headers,"${_}s"); } - my $txt_format1 = " %-16s %9s %6d $sender_txt_format"; + my $txt_format1 = " %-16s %9s %6d %6s $sender_txt_format"; my $txt_format2 = " %6d %4.1f%% %6d %4.1f%%", - my $htm_format1 = "$sender_html_format"; + my $htm_format1 = "$sender_html_format"; my $htm_format2 = ""; if ($txt_fh) { @@ -2296,43 +2642,24 @@ sub print_grandtotals { print $txt_fh "\n"; print $txt_fh "Grand total summary\n"; print $txt_fh "-------------------\n"; - print $txt_fh " $sender_spaces At least one address\n"; - print $txt_fh " TOTAL Volume Messages $sender_txt_header Delayed Failed\n"; + print $txt_fh " $sender_spaces At least one address\n"; + print $txt_fh " TOTAL Volume Messages Addresses $sender_txt_header Delayed Failed\n"; } if ($htm_fh) { - print $htm_fh "\n"; + print $htm_fh "\n"; print $htm_fh "

Grand total summary

\n"; print $htm_fh "
\n"; - print $htm_fh "\n"; - print $htm_fh "\n"; + my($text,$m_count,$a_count,$m_data,$m_data_gigs,$spreadsheet, $row_sref) = @_; + my($name) = ($topcount == 1)? "$text" : "$topcount ${text}s"; + my($title) = "Top $name by message count"; + my(@chartdatanames) = (); + my(@chartdatavals) = (); + my $chartotherval = 0; + $text = ucfirst($text); # Align non-local addresses to the right (so all the .com's line up). # Local addresses are aligned on the left as they are userids. my $align = ($text !~ /local/i) ? 'right' : 'left'; - $htm_format = "\n"; -} -if ($xls_fh) -{ - $ws_top50->write($row_league_table++, 0, $temp, $f_header2); - &set_worksheet_line($ws_top50, $row_league_table++, 0, ["Messages", "Bytes", "Average", $text], $f_headertab ); -} -# write content -my($key,$htmlkey,$rounded_volume,$rounded_average,$count,$data,$gigs); -foreach $key (top_n_sort($topcount,$m_count,$m_data_gigs,$m_data)) { + ################################################ + # Generate the printf formats and table headers. + ################################################ + my(@headers) = ('Messages'); + #push(@headers,'Addresses') if defined $a_count; + push(@headers,'Addresses') if defined $a_count && %$a_count; + push(@headers,'Bytes','Average') if defined $m_data; - # When displaying the average figures, we calculate the average of - # the rounded data, as the user would calculate it. This reduces - # the accuracy slightly, but we have to do it this way otherwise - # when using -merge to convert results from text to HTML and - # vice-versa discrepencies would occur. - $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); - $data = $gigs = 0; - un_round($rounded_volume,\$data,\$gigs); - $count = $$m_count{$key}; - $rounded_average = volume_rounded($data/$count,$gigs/$count); - my @content=( $count, $rounded_volume, $rounded_average); + my $txt_format = "%10s " x @headers . " %s\n"; + my $txt_col_headers = sprintf $txt_format, @headers, $text; + my $htm_format = "" . ''x@headers . "\n"; + my $htm_col_headers = sprintf $htm_format, @headers, $text; + $htm_col_headers =~ s/(<\/?)td/$1th/g; #Convert
MessagesBytesAverage\u$text
%d%s%s%s
%s%s
's to 's for the header. - # write content - # any reason not to include rounded_average in txt-output? -fh - printf $txt_fh ($txt_format, $count, $rounded_volume, $key) if $txt_fh; + + ################################################ + # Write the table headers + ################################################ + printf $txt_fh ("%s\n%s\n%s", $title, "-" x length($title),$txt_col_headers) if $txt_fh; if ($htm_fh) { - $htmlkey = $key; - $htmlkey =~ s/>/\>\;/g; - $htmlkey =~ s/

$title

+ +
+ +EoText + print $htm_fh $htm_col_headers } - if (scalar @chartdatanames < $ntopchart) - { - push(@chartdatanames, $key); - push(@chartdatavals, $$m_count{$key}); - } - else - { - $chartotherval += $$m_count{$key}; + if ($xls_fh) { + $spreadsheet->write(${$row_sref}++, 0, $title, $f_header2); + $spreadsheet->write(${$row_sref}++, 0, [@headers, $text], $f_headertab); } -} -push(@chartdatanames, "Other"); -push(@chartdatavals, $chartotherval); -print $txt_fh "\n" if $txt_fh; -if ($htm_fh) -{ - print $htm_fh "
\n"; - print $htm_fh "
\n"; - if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) - { - # calculate the graph - my @data = ( - \@chartdatanames, - \@chartdatavals - ); - my $graph = GD::Graph::pie->new(300, 300); - $graph->set( - x_label => 'Name', - y_label => 'Amount', - title => 'By count', - ); - my $gd = $graph->plot(\@data) or warn($graph->error); - if ($gd) { - my $temp = $text; - $temp =~ s/ /_/g; - open(IMG, ">$chartdir/${temp}_count.png") or die "Could not write $chartdir/${temp}_count.png: $!\n"; - binmode IMG; - print IMG $gd->png; - close IMG; - print $htm_fh ""; + # write content + foreach my $key (top_n_sort($topcount,$m_count,$m_data_gigs,$m_data)) { + + # When displaying the average figures, we calculate the average of + # the rounded data, as the user would calculate it. This reduces + # the accuracy slightly, but we have to do it this way otherwise + # when using -merge to convert results from text to HTML and + # vice-versa discrepencies would occur. + my $messages = $$m_count{$key}; + my @content = ($messages); + push(@content, $$a_count{$key}) if defined $a_count; + if (defined $m_data) { + my $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); + my($data,$gigs) = (0,0); + un_round($rounded_volume,\$data,\$gigs); + my $rounded_average = volume_rounded($data/$messages,$gigs/$messages); + push(@content, $rounded_volume, $rounded_average); } - } - print $htm_fh "\n"; - print $htm_fh "
\n\n"; -} -if ($xls_fh) -{ - $row_league_table++; -} + # write content + printf $txt_fh ($txt_format, @content, $key) if $txt_fh; -# write header - -$temp = "Top $name by volume"; + if ($htm_fh) { + my $htmlkey = $key; + $htmlkey =~ s/>/\>\;/g; + $htmlkey =~ s/write(${$row_sref}++, 0, [@content, $key], $f_default) if $xls_fh; -printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; -if ($htm_fh) { - print $htm_fh "

$temp

\n"; - print $htm_fh "\n"; - print $htm_fh "
\n"; - print $htm_fh "\n"; - print $htm_fh "\n"; -} -if ($xls_fh) -{ - $ws_top50->write($row_league_table++, 0, $temp, $f_header2); - &set_worksheet_line($ws_top50, $row_league_table++, 0, ["Messages", "Bytes", "Average", $text], $f_headertab); -} + if (scalar @chartdatanames < $ntopchart) { + push(@chartdatanames, $key); + push(@chartdatavals, $$m_count{$key}); + } + else { + $chartotherval += $$m_count{$key}; + } + } -@chartdatanames = (); -@chartdatavals = (); -$chartotherval = 0; -my $use_gig = 0; -foreach $key (top_n_sort($topcount,$m_data_gigs,$m_data,$m_count)) { - # The largest volume will be the first (top of the list). - # If it has at least 1 gig, then just use gigabytes to avoid - # risking an integer overflow when generating the pie charts. - if ($$m_data_gigs{$key}) { - $use_gig = 1; - } - - $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); - $data = $gigs = 0; - un_round($rounded_volume,\$data,\$gigs); - $count = $$m_count{$key}; - $rounded_average = volume_rounded($data/$count,$gigs/$count); - my @content=($count, $rounded_volume, $rounded_average ); + push(@chartdatanames, "Other"); + push(@chartdatavals, $chartotherval); - # write content - # any reasons for not including rounded_average in the txt-version?? -fh - printf $txt_fh ($txt_format, $count, $rounded_volume, $key) if $txt_fh; + print $txt_fh "\n" if $txt_fh; if ($htm_fh) { - $htmlkey = $key; - $htmlkey =~ s/>/\>\;/g; - $htmlkey =~ s/\n"; + print $htm_fh "
MessagesBytesAverage\u$text\n"; + if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) + { + # calculate the graph + my @data = ( + \@chartdatanames, + \@chartdatavals + ); + my $graph = GD::Graph::pie->new(300, 300); + $graph->set( + x_label => 'Name', + y_label => 'Amount', + title => 'By count', + ); + my $gd = $graph->plot(\@data) or warn($graph->error); + if ($gd) { + my $temp = $text; + $temp =~ s/ /_/g; + open(IMG, ">$chartdir/${temp}_count.png") or die "Could not write $chartdir/${temp}_count.png: $!\n"; + binmode IMG; + print IMG $gd->png; + close IMG; + print $htm_fh ""; + } + } + print $htm_fh "\n"; + print $htm_fh "
\n\n"; } + ++${$row_sref} if $xls_fh; + + if (defined $m_data) { + # write header - if (scalar @chartdatanames < $ntopchart) { - if ($use_gig) { + $title = "Top $name by volume"; + + printf $txt_fh ("%s\n%s\n%s", $title, "-" x length($title),$txt_col_headers) if $txt_fh; + + if ($htm_fh) { + print $htm_fh <

$title

+ +
+ +EoText + print $htm_fh $htm_col_headers; + } + if ($xls_fh) { + $spreadsheet->write(${$row_sref}++, 0, $title, $f_header2); + $spreadsheet->write(${$row_sref}++, 0, [@headers, $text], $f_headertab); + } + + @chartdatanames = (); + @chartdatavals = (); + $chartotherval = 0; + my $use_gig = 0; + foreach my $key (top_n_sort($topcount,$m_data_gigs,$m_data,$m_count)) { + # The largest volume will be the first (top of the list). + # If it has at least 1 gig, then just use gigabytes to avoid + # risking an integer overflow when generating the pie charts. if ($$m_data_gigs{$key}) { - push(@chartdatanames, $key); - push(@chartdatavals, $$m_data_gigs{$key}); + $use_gig = 1; + } + + my $messages = $$m_count{$key}; + my @content = ($messages); + push(@content, $$a_count{$key}) if defined $a_count; + my $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); + my($data ,$gigs) = (0,0); + un_round($rounded_volume,\$data,\$gigs); + my $rounded_average = volume_rounded($data/$messages,$gigs/$messages); + push(@content, $rounded_volume, $rounded_average ); + + # write content + printf $txt_fh ($txt_format, @content, $key) if $txt_fh; + if ($htm_fh) { + my $htmlkey = $key; + $htmlkey =~ s/>/\>\;/g; + $htmlkey =~ s/write(${$row_sref}++, 0, [@content, $key], $f_default) if $xls_fh; + + + if (scalar @chartdatanames < $ntopchart) { + if ($use_gig) { + if ($$m_data_gigs{$key}) { + push(@chartdatanames, $key); + push(@chartdatavals, $$m_data_gigs{$key}); + } + } + else { + push(@chartdatanames, $key); + push(@chartdatavals, $$m_data{$key}); + } + } + else { + $chartotherval += ($use_gig) ? $$m_data_gigs{$key} : $$m_data{$key}; } } - else { - push(@chartdatanames, $key); - push(@chartdatavals, $$m_data{$key}); - } - } - else { - $chartotherval += ($use_gig) ? $$m_data_gigs{$key} : $$m_data{$key}; - } -} -push(@chartdatanames, "Other"); -push(@chartdatavals, $chartotherval); + push(@chartdatanames, "Other"); + push(@chartdatavals, $chartotherval); -print $txt_fh "\n" if $txt_fh; -if ($htm_fh) { - print $htm_fh "
\n"; - print $htm_fh "
\n"; - if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) { - # calculate the graph - my @data = ( - \@chartdatanames, - \@chartdatavals - ); - my $graph = GD::Graph::pie->new(300, 300); - $graph->set( - x_label => 'Name', - y_label => 'Volume' , - title => 'By Volume', - ); - my $gd = $graph->plot(\@data) or warn($graph->error); - if ($gd) { - $temp = $text; - $temp =~ s/ /_/g; - open(IMG, ">$chartdir/${temp}_volume.png") or die "Could not write $chartdir/${temp}_volume.png: $!\n"; - binmode IMG; - print IMG $gd->png; - close IMG; - print $htm_fh ""; + print $txt_fh "\n" if $txt_fh; + if ($htm_fh) { + print $htm_fh "
\n"; + print $htm_fh "
\n"; + if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) { + # calculate the graph + my @data = ( + \@chartdatanames, + \@chartdatavals + ); + my $graph = GD::Graph::pie->new(300, 300); + $graph->set( + x_label => 'Name', + y_label => 'Volume' , + title => 'By Volume', + ); + my $gd = $graph->plot(\@data) or warn($graph->error); + if ($gd) { + my $temp = $text; + $temp =~ s/ /_/g; + open(IMG, ">$chartdir/${temp}_volume.png") or die "Could not write $chartdir/${temp}_volume.png: $!\n"; + binmode IMG; + print IMG $gd->png; + close IMG; + print $htm_fh ""; + } + } + print $htm_fh "\n"; + print $htm_fh "
\n\n"; } + + ++${$row_sref} if $xls_fh; } - print $htm_fh "
\n"; - print $htm_fh "
\n\n"; -} -if ($xls_fh) -{ - ++$row_league_table; } -} ####################################################################### # top_n_sort(); @@ -1557,6 +1633,12 @@ sub top_n_sort { my $n_minus_1 = $n - 1; my $n_minus_2 = $n - 2; + # Create a dummy hash incase the user has not provided us with + # tiebreaker hashes. + my(%dummy_hash); + $href2 = \%dummy_hash unless defined $href2; + $href3 = \%dummy_hash unless defined $href3; + # Pick out the top $n keys. my($key,$value1,$value2,$value3,$i,$comparison,$insert_position); while (($key,$value1) = each %$href1) { @@ -1564,7 +1646,18 @@ sub top_n_sort { #print STDERR "key $key ($value1,",$href2->{$key},",",$href3->{$key},") <=> ($minimum_value1,$minimum_value2,$minimum_value3)\n"; # Check to see that the new value is bigger than the lowest of the - # top n keys that we're keeping. + # top n keys that we're keeping. We test the main key first, because + # for the majority of cases we can skip creating dummy hash values + # should the user have not provided real tie-breaking hashes. + next unless $value1 >= $minimum_value1; + + # Create a dummy hash entry for the key if required. + # Note that setting the dummy_hash value sets it for both href2 & + # href3. Also note that currently we are guaranteed to have a real + # value for href3 if a real value for href2 exists so don't need to + # test for it as well. + $dummy_hash{$key} = 0 unless exists $href2->{$key}; + $comparison = $value1 <=> $minimum_value1 || $href2->{$key} <=> $minimum_value2 || $href3->{$key} <=> $minimum_value3 || @@ -1622,6 +1715,7 @@ sub top_n_sort { } + ####################################################################### # html_header(); # @@ -1703,6 +1797,7 @@ Valid options are: -bydomain show results by sending domain. -byemail show results by sender's email address -byedomain show results by sender's email domain +-bylocaldomain show results by local domain -pattern "Description" /pattern/ Count lines matching specified patterns and show them in @@ -1718,6 +1813,8 @@ Valid options are: from where to include the charts in the html file -chartdir and -chartrel default to '.' +-emptyok It is OK if there is no valid input, don't print an error. + -d Debug mode - dump the eval'ed parser onto STDERR. EoText @@ -1745,8 +1842,9 @@ EoText sub generate_parser { my $parser = ' my($ip,$host,$email,$edomain,$domain,$thissize,$size,$old,$new); - my($tod,$m_hour,$m_min,$id,$flag); - my($seconds,$queued,$rcpt_time); + my($tod,$m_hour,$m_min,$id,$flag,$extra,$length); + my($seconds,$queued,$rcpt_time,$local_domain); + my $rej_id = 0; while (<$fh>) { # Convert syslog lines to mainlog format. @@ -1754,41 +1852,81 @@ sub generate_parser { next unless s/^.*? exim\\b.*?: //; } - next if length($_) < 38; - next unless /^(\\d{4}\\-\\d\\d-\\d\\d\\s(\\d\\d):(\\d\\d):\\d\\d( [-+]\\d\\d\\d\\d)?)/o; + $length = length($_); + next if ($length < 38); + next unless /^ + (\\d{4}\\-\\d\\d-\\d\\d\\s # 1: YYYYMMDD HHMMSS + (\\d\\d) # 2: HH + : + (\\d\\d) # 3: MM + :\\d\\d + ) + (\\.\\d+)? # 4: subseconds + (\s[-+]\\d\\d\\d\\d)? # 5: tz-offset + (\s\\[\\d+\\])? # 6: pid + /ox; + + $tod = defined($5) ? $1 . $5 : $1; + ($m_hour,$m_min) = ($2,$3); + + # PH - watch for GMT offsets in the timestamp. + if (defined($5)) { + $extra = 6; + next if ($length < 44); + } + else { + $extra = 0; + } - ($tod,$m_hour,$m_min) = ($1,$2,$3); + # watch for subsecond precision + if (defined($4)) { + $extra += length($4); + next if ($length < 38 + $extra); + } + + # PH - watch for PID added after the timestamp. + if (defined($6)) { + $extra += length($6); + next if ($length < 38 + $extra); + } - # PH - my($extra) = defined($4)? 6 : 0; $id = substr($_, 20 + $extra, 16); $flag = substr($_, 37 + $extra, 2); + + if ($flag !~ /^([<>=*-]+|SA)$/ && /rejected|refused|dropped/) { + $flag = "Re"; + $extra -= 3; + } + + # Rejects can have no MSGID... + if ($flag eq "Re" && $id !~ /^[-0-9a-zA-Z]+$/) { + $id = "reject:" . ++$rej_id; + $extra -= 17; + } '; # Watch for user specified patterns. my $user_pattern_index = 0; foreach (@user_patterns) { $user_pattern_totals[$user_pattern_index] = 0; - $parser .= " if ($_) {\n"; - $parser .= " \$user_pattern_totals[$user_pattern_index]++ if $_;\n"; - $parser .= " \$user_pattern_interval_count[$user_pattern_index][(\$m_hour*60 + \$m_min)/$hist_interval]++;\n" if ($hist_opt > 0); - $parser .= " }\n"; + $parser .= " if ($_) {\n"; + $parser .= " \$user_pattern_totals[$user_pattern_index]++;\n"; + $parser .= " \$user_pattern_interval_count[$user_pattern_index][(\$m_hour*60 + \$m_min)/$hist_interval]++;\n" if ($hist_opt > 0); + $parser .= " }\n"; $user_pattern_index++; } $parser .= ' - next unless ($flag =~ /<=|=>|->|==|\\*\\*|Co/); - - #Strip away the timestamp, ID and flag (which could be "Com" for completed) - #This speeds up the later pattern matches. - # $_ = substr($_, 40); + next unless ($flag =~ /<=|=>|->|==|\\*\\*|Co|SA|Re/); + #Strip away the timestamp, ID and flag to speed up later pattern matches. + #The flags include Co (Completed), Re (Rejected), and SA (SpamAssassin). $_ = substr($_, 40 + $extra); # PH - # Get a pointer to an array of information about the message. + # Alias @message to the array of information about the message. # This minimises the number of calls to hash functions. $messages{$id} = [] unless exists $messages{$id}; - $message_aref = $messages{$id}; + *message = $messages{$id}; # JN - Skip over certain transports as specified via the "-nt/.../" command @@ -1802,33 +1940,35 @@ sub generate_parser { #ENDIF ($transport_pattern) - $host = "local"; #Host is local unless otherwise specified. - $domain = "localdomain"; #Domain is localdomain unless otherwise specified. - # Do some pattern matches to get the host and IP address. # We expect lines to be of the form "H=[IpAddr]" or "H=Host [IpAddr]" or # "H=Host (UnverifiedHost) [IpAddr]" or "H=(UnverifiedHost) [IpAddr]". # We do 2 separate matches to keep the matches simple and fast. - if (/\\sH=(\\S+)/) { - $host = $1; - - ($ip) = /\\sH=.*?(\\s\\[[^]]+\\])/; - # If there is only an IP address, it will be in $host and $ip will be - # unset. That is OK, because we only use $ip in conjunction with $host - # below. But make it empty to avoid warning messages. - $ip = "" if !defined $ip; + # Host is local unless otherwise specified. + # Watch out for "H=([IpAddr])" in case they send "[IpAddr]" as their HELO! + $ip = (/\\bH=(?:|.*? )(\\[[^]]+\\])/) ? $1 + # 2008-03-31 06:25:22 Connection from [213.246.33.217]:39456 refused: too many connections from that IP address // .hs + : (/Connection from (\[\S+\])/) ? $1 + # 2008-03-31 06:52:40 SMTP call from mail.cacoshrf.com (ccsd02.ccsd.local) [69.24.118.229]:4511 dropped: too many nonmail commands (last was "RSET") // .hs + : (/SMTP call from .*?(\[\S+\])/) ? $1 + : "local"; + $host = (/\\bH=(\\S+)/) ? $1 : "local"; - #IFDEF ($do_sender{Domain}) - if ($host !~ /^\\[/ && $host =~ /^(\\(?)[^\\.]+\\.([^\\.]+\\..*)/) { - # Remove the host portion from the DNS name. We ensure that we end up - # with at least xxx.yyy. $host can be "(x.y.z)" or "x.y.z". - $domain = lc("$1.$2"); - $domain =~ s/^\\.//; #Remove preceding dot. - } - #ENDIF ($do_sender{Domain}) + $domain = "localdomain"; #Domain is localdomain unless otherwise specified. + #IFDEF ($do_sender{Domain}) + if ($host =~ /^\\[/ || $host =~ /^[\\d\\.]+$/) { + # Host is just an IP address. + $domain = $host; } + elsif ($host =~ /^(\\(?)[^\\.]+\\.([^\\.]+\\..*)/) { + # Remove the host portion from the DNS name. We ensure that we end up + # with at least xxx.yyy. $host can be "(x.y.z)" or "x.y.z". + $domain = lc("$1.$2"); + $domain =~ s/^\\.//; #Remove preceding dot. + } + #ENDIF ($do_sender{Domain}) #IFDEF ($do_sender{Email}) #IFDEF ($include_original_destination) @@ -1844,16 +1984,25 @@ sub generate_parser { #ENDIF ($do_sender{Email}) #IFDEF ($do_sender{Edomain}) + if (/^(<>|blackhole)/) { + $edomain = $1; + } #IFDEF ($include_original_destination) - #$edomain = (/^(\S+) (<\S*?\\@(\S+)>)?/) ? $3 || $1 : ""; - $edomain = (/^(\S+ (<\S*?\\@(\S+?)>)?)/) ? $1 : ""; - chomp($edomain); - lc($edomain); + elsif (/^(\S+ (<\S*?\\@(\S+?)>)?)/) { + $edomain = $1; + chomp($edomain); + $edomain =~ s/@(\S+?)>/"@" . lc($1) . ">"/e; + } #ENDIF ($include_original_destination) - #IFNDEF ($include_original_destination) - $edomain = (/^\S*?\\@(\S+)/) ? lc($1) : ""; + elsif (/^\S*?\\@(\S+)/) { + $edomain = lc($1); + } #ENDIF ($include_original_destination) + else { + $edomain = ""; + } + #ENDIF ($do_sender{Edomain}) if ($tod lt $begin) { @@ -1866,16 +2015,16 @@ sub generate_parser { if ($flag eq "<=") { $thissize = (/\\sS=(\\d+)( |$)/) ? $1 : 0; - $message_aref->[$SIZE] = $thissize; - $message_aref->[$PROTOCOL] = (/ P=(\S+)/) ? $1 : undef; + $message[$SIZE] = $thissize; + $message[$PROTOCOL] = (/ P=(\S+)/) ? $1 : undef; #IFDEF ($show_relay) if ($host ne "local") { # Save incoming information in case it becomes interesting # later, when delivery lines are read. my($from) = /^(\\S+)/; - $message_aref->[$FROM_HOST] = "$host$ip"; - $message_aref->[$FROM_ADDRESS] = $from; + $message[$FROM_HOST] = "$host$ip"; + $message[$FROM_ADDRESS] = $from; } #ENDIF ($show_relay) @@ -1927,7 +2076,7 @@ sub generate_parser { add_volume(\\$total_received_data,\\$total_received_data_gigs,$thissize); #IFDEF ($#queue_times >= 0 || $#rcpt_times >= 0) - $message_aref->[$ARRIVAL_TIME] = $tod; + $message[$ARRIVAL_TIME] = $tod; #ENDIF ($#queue_times >= 0 || $#rcpt_times >= 0) #IFDEF ($hist_opt > 0) @@ -1936,9 +2085,9 @@ sub generate_parser { } elsif ($flag eq "=>") { - $size = $message_aref->[$SIZE] || 0; + $size = $message[$SIZE] || 0; if ($host ne "local") { - $message_aref->[$REMOTE_DELIVERED] = 1; + $message[$REMOTE_DELIVERED] = 1; #IFDEF ($show_relay) @@ -1948,7 +2097,7 @@ sub generate_parser { # addresses, there may be a further address between the first # and last. - if (defined $message_aref->[$FROM_HOST]) { + if (defined $message[$FROM_HOST]) { if (/^(\\S+)(?:\\s+\\([^)]\\))?\\s+<([^>]+)>/) { ($old,$new) = ($1,$2); } @@ -1958,7 +2107,7 @@ sub generate_parser { if ("\\L$new" eq "\\L$old") { ($old) = /^(\\S+)/ if $old eq ""; - my $key = "H=\\L$message_aref->[$FROM_HOST]\\E A=\\L$message_aref->[$FROM_ADDRESS]\\E => " . + my $key = "H=\\L$message[$FROM_HOST]\\E A=\\L$message[$FROM_ADDRESS]\\E => " . "H=\\L$host\\E$ip A=\\L$old\\E"; if (!defined $relay_pattern || $key !~ /$relay_pattern/o) { $relayed{$key} = 0 if !defined $relayed{$key}; @@ -1993,35 +2142,52 @@ sub generate_parser { #IFNDEF ($include_original_destination) if ($user =~ /^[\\/|]/) { #ENDIF ($include_original_destination) - my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/; - $user = "$user $parent" if defined $parent; + #my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/; + my($parent) = $_ =~ / (<.+?>) /; #DT 1.54 + if (defined $parent) { + $user = "$user $parent"; + #IFDEF ($do_local_domain) + if ($parent =~ /\\@(.+)>/) { + $local_domain = lc($1); + ++$delivered_messages_local_domain{$local_domain}; + ++$delivered_addresses_local_domain{$local_domain}; + add_volume(\\$delivered_data_local_domain{$local_domain},\\$delivered_data_gigs_local_domain{$local_domain},$size); + } + #ENDIF ($do_local_domain) + } } - ++$delivered_count_user{$user}; + ++$delivered_messages_user{$user}; + ++$delivered_addresses_user{$user}; add_volume(\\$delivered_data_user{$user},\\$delivered_data_gigs_user{$user},$size); } } #ENDIF ($local_league_table || $include_remote_users) #IFDEF ($do_sender{Host}) - $delivered_count{Host}{$host}++; + $delivered_messages{Host}{$host}++; + $delivered_addresses{Host}{$host}++; add_volume(\\$delivered_data{Host}{$host},\\$delivered_data_gigs{Host}{$host},$size); #ENDIF ($do_sender{Host}) #IFDEF ($do_sender{Domain}) if ($domain) { - ++$delivered_count{Domain}{$domain}; + ++$delivered_messages{Domain}{$domain}; + ++$delivered_addresses{Domain}{$domain}; add_volume(\\$delivered_data{Domain}{$domain},\\$delivered_data_gigs{Domain}{$domain},$size); } #ENDIF ($do_sender{Domain}) #IFDEF ($do_sender{Email}) - ++$delivered_count{Email}{$email}; + ++$delivered_messages{Email}{$email}; + ++$delivered_addresses{Email}{$email}; add_volume(\\$delivered_data{Email}{$email},\\$delivered_data_gigs{Email}{$email},$size); #ENDIF ($do_sender{Email}) #IFDEF ($do_sender{Edomain}) - ++$delivered_count{Edomain}{$edomain}; + ++$delivered_messages{Edomain}{$edomain}; + ++$delivered_addresses{Edomain}{$edomain}; add_volume(\\$delivered_data{Edomain}{$edomain},\\$delivered_data_gigs{Edomain}{$edomain},$size); #ENDIF ($do_sender{Edomain}) - ++$total_delivered_count; + ++$total_delivered_messages; + ++$total_delivered_addresses; add_volume(\\$total_delivered_data,\\$total_delivered_data_gigs,$size); #IFDEF ($show_transport) @@ -2040,29 +2206,78 @@ sub generate_parser { for ($i = 0; $i <= $#delivery_times; $i++) { if ($seconds < $delivery_times[$i]) { ++$dt_all_bin[$i]; - ++$dt_remote_bin[$i] if $message_aref->[$REMOTE_DELIVERED]; + ++$dt_remote_bin[$i] if $message[$REMOTE_DELIVERED]; last; } } if ($i > $#delivery_times) { ++$dt_all_overflow; - ++$dt_remote_overflow if $message_aref->[$REMOTE_DELIVERED]; + ++$dt_remote_overflow if $message[$REMOTE_DELIVERED]; } } #ENDIF ($#delivery_times > 0) } - elsif ($flag eq "==" && defined($message_aref->[$SIZE]) && !defined($message_aref->[$DELAYED])) { + elsif ($flag eq "->") { + + #IFDEF ($local_league_table || $include_remote_users) + #IFDEF ($local_league_table && $include_remote_users) + { #Store both local and remote users. + #ENDIF ($local_league_table && $include_remote_users) + + #IFDEF ($local_league_table && ! $include_remote_users) + if ($host eq "local") { #Store local users only. + #ENDIF ($local_league_table && ! $include_remote_users) + + #IFDEF ($include_remote_users && ! $local_league_table) + if ($host ne "local") { #Store remote users only. + #ENDIF ($include_remote_users && ! $local_league_table) + + if (my($user) = split((/\\s]*>)/; + my($parent) = $_ =~ / (<.+?>) /; #DT 1.54 + $user = "$user $parent" if defined $parent; + } + ++$delivered_addresses_user{$user}; + } + } + #ENDIF ($local_league_table || $include_remote_users) + + #IFDEF ($do_sender{Host}) + $delivered_addresses{Host}{$host}++; + #ENDIF ($do_sender{Host}) + #IFDEF ($do_sender{Domain}) + if ($domain) { + ++$delivered_addresses{Domain}{$domain}; + } + #ENDIF ($do_sender{Domain}) + #IFDEF ($do_sender{Email}) + ++$delivered_addresses{Email}{$email}; + #ENDIF ($do_sender{Email}) + #IFDEF ($do_sender{Edomain}) + ++$delivered_addresses{Edomain}{$edomain}; + #ENDIF ($do_sender{Edomain}) + + ++$total_delivered_addresses; + } + + elsif ($flag eq "==" && defined($message[$SIZE]) && !defined($message[$DELAYED])) { ++$delayed_count; - $message_aref->[$DELAYED] = 1; + $message[$DELAYED] = 1; } elsif ($flag eq "**") { - if (defined ($message_aref->[$SIZE])) { - unless (defined $message_aref->[$HAD_ERROR]) { + if (defined ($message[$SIZE])) { + unless (defined $message[$HAD_ERROR]) { ++$message_errors; - $message_aref->[$HAD_ERROR] = 1; + $message[$HAD_ERROR] = 1; } } @@ -2075,18 +2290,18 @@ sub generate_parser { elsif ($flag eq "Co") { #Completed? #IFDEF ($#queue_times >= 0) - $queued = queue_time($tod, $message_aref->[$ARRIVAL_TIME], $id); + $queued = queue_time($tod, $message[$ARRIVAL_TIME], $id); for ($i = 0; $i <= $#queue_times; $i++) { if ($queued < $queue_times[$i]) { ++$qt_all_bin[$i]; - ++$qt_remote_bin[$i] if $message_aref->[$REMOTE_DELIVERED]; + ++$qt_remote_bin[$i] if $message[$REMOTE_DELIVERED]; last; } } if ($i > $#queue_times) { ++$qt_all_overflow; - ++$qt_remote_overflow if $message_aref->[$REMOTE_DELIVERED]; + ++$qt_remote_overflow if $message[$REMOTE_DELIVERED]; } #ENDIF ($#queue_times >= 0) @@ -2095,13 +2310,13 @@ sub generate_parser { $seconds = wdhms_seconds($1); #Calculate $queued if not previously calculated above. #IFNDEF ($#queue_times >= 0) - $queued = queue_time($tod, $message_aref->[$ARRIVAL_TIME], $id); + $queued = queue_time($tod, $message[$ARRIVAL_TIME], $id); #ENDIF ($#queue_times >= 0) $rcpt_time = $seconds - $queued; my($protocol); - if (defined $message_aref->[$PROTOCOL]) { - $protocol = $message_aref->[$PROTOCOL]; + if (defined $message[$PROTOCOL]) { + $protocol = $message[$PROTOCOL]; # Create the bin if its not already defined. unless (exists $rcpt_times_bin{$protocol}) { @@ -2127,6 +2342,126 @@ sub generate_parser { delete($messages{$id}); } + elsif ($flag eq "SA") { + $ip = (/From.*?(\\[[^]]+\\])/ || /\\((local)\\)/) ? $1 : ""; + #SpamAssassin message + if (/Action: ((permanently|temporarily) rejected message|flagged as Spam but accepted): score=(\d+\.\d)/) { + #add_volume(\\$spam_score,\\$spam_score_gigs,$3); + ++$spam_count_by_ip{$ip}; + } elsif (/Action: scanned but message isn\'t spam: score=(-?\d+\.\d)/) { + #add_volume(\\$ham_score,\\$ham_score_gigs,$1); + ++$ham_count_by_ip{$ip}; + } elsif (/(Not running SA because SAEximRunCond expanded to false|check skipped due to message size)/) { + ++$ham_count_by_ip{$ip}; + } + } + + # Look for Reject messages or blackholed messages (deliveries + # without a transport) + if ($flag eq "Re" || ($flag eq "=>" && ! /\\sT=\\S+/)) { + # Correct the IP address for rejects: + # rejected EHLO from my.test.net [10.0.0.5]: syntactically invalid argument(s): + # rejected EHLO from [10.0.0.6]: syntactically invalid argument(s): + $ip = $1 if ($ip eq "local" && /^rejected [HE][HE]LO from .*?(\[.+?\]):/); + if (/SpamAssassin/) { + ++$rejected_count_by_reason{"Rejected by SpamAssassin"}; + ++$rejected_count_by_ip{$ip}; + } + elsif ( + /(temporarily rejected [A-Z]*) .*?(: .*?)(:|\s*$)/ + ) { + ++$temporarily_rejected_count_by_reason{"\u$1$2"}; + ++$temporarily_rejected_count_by_ip{$ip}; + } + elsif ( + /(temporarily refused connection)/ + ) { + ++$temporarily_rejected_count_by_reason{"\u$1"}; + ++$temporarily_rejected_count_by_ip{$ip}; + } + elsif ( + /(listed at [^ ]+)/ || + /(Forged IP detected in HELO)/ || + /(Invalid domain or IP given in HELO\/EHLO)/ || + /(unqualified recipient rejected)/ || + /(closed connection (after|in response) .*?)\s*$/ || + /(sender rejected)/ || + # 2005-09-23 15:07:49 1EInHJ-0007Ex-Au H=(a.b.c) [10.0.0.1] F=<> rejected after DATA: This message contains a virus: (Eicar-Test-Signature) please scan your system. + # 2005-10-06 10:50:07 1ENRS3-0000Nr-Kt => blackhole (DATA ACL discarded recipients): This message contains a virus: (Worm.SomeFool.P) please scan your system. + / rejected after DATA: (.*)/ || + / (rejected DATA: .*)/ || + /.DATA ACL discarded recipients.: (.*)/ || + /rejected after DATA: (unqualified address not permitted)/ || + /(VRFY rejected)/ || +# /(sender verify (defer|fail))/i || + /(too many recipients)/ || + /(refused relay.*?) to/ || + /(rejected by non-SMTP ACL: .*)/ || + /(rejected by local_scan.*)/ || + # SMTP call from %s dropped: too many syntax or protocol errors (last command was "%s" + # SMTP call from %s dropped: too many nonmail commands + /(dropped: too many ((nonmail|unrecognized) commands|syntax or protocol errors))/ || + + # local_scan() function crashed with signal %d - message temporarily rejected + # local_scan() function timed out - message temporarily rejected + /(local_scan.. function .* - message temporarily rejected)/ || + # SMTP protocol synchronization error (input sent without waiting for greeting): rejected connection from %s + /(SMTP protocol .*?(error|violation))/ || + /(message too big)/ + ) { + ++$rejected_count_by_reason{"\u$1"}; + ++$rejected_count_by_ip{$ip}; + } + elsif (/rejected [HE][HE]LO from [^:]*: syntactically invalid argument/) { + ++$rejected_count_by_reason{"Rejected HELO/EHLO: syntactically invalid argument"}; + ++$rejected_count_by_ip{$ip}; + } + elsif (/response to "RCPT TO.*? was: (.*)/) { + ++$rejected_count_by_reason{"Response to RCPT TO was: $1"}; + ++$rejected_count_by_ip{$ip}; + } + elsif ( + /(lookup of host )\S+ (failed)/ || + + # rejected from <%s>%s%s%s%s: message too big: + /(rejected [A-Z]*) .*?(: .*?)(:|\s*$)/ || + # refused connection from %s (host_reject_connection) + # refused connection from %s (tcp wrappers) + /(refused connection )from.*? (\(.*)/ || + + # error from remote mailer after RCPT TO:: host a.b.c [10.0.0.1]: 450 : Recipient address rejected: Greylisted for 60 seconds + # error from remote mailer after MAIL FROM:<> SIZE=3468: host a.b.c [10.0.0.1]: 421 a.b.c has refused your connection because your server did not have a PTR record. + /(error from remote mailer after .*?:).*(: .*?)(:|\s*$)/ || + + # a.b.c F= rejected after DATA: "@" or "." expected after "Undisclosed-Recipient": failing address in "To" header is: + /rejected after DATA: ("." or "." expected).*?(: failing address in .*? header)/ || + + # connection from %s refused load average = %.2f + /(Connection )from.*? (refused: load average)/ || + # connection from %s refused (IP options) + # Connection from %s refused: too many connections + # connection from %s refused + /([Cc]onnection )from.*? (refused.*)/ || + # [10.0.0.1]: connection refused + /: (Connection refused)()/ + ) { + ++$rejected_count_by_reason{"\u$1$2"}; + ++$rejected_count_by_ip{$ip}; + } + elsif ( + # 2008-03-31 06:25:22 H=mail.densitron.com [216.70.140.224]:45386 temporarily rejected connection in "connect" ACL: too fast reconnects // .hs + # 2008-03-31 06:25:22 H=mail.densitron.com [216.70.140.224]:45386 temporarily rejected connection in "connect" ACL // .hs + /(temporarily rejected connection in .*?ACL:?.*)/ + ) { + ++$temporarily_rejected_count_by_ip{$ip}; + ++$temporarily_rejected_count_by_reason{"\u$1"}; + } + else { + ++$rejected_count_by_reason{Unknown}; + ++$rejected_count_by_ip{$ip}; + print STDERR "Unknown rejection: $_" if $debug; + } + } }'; # We now do a 'C preprocessor style operation on our parser @@ -2200,9 +2535,9 @@ sub print_header { if ($htm_fh) { print $htm_fh html_header($title); print $htm_fh "\n
\n"; @@ -2266,9 +2611,10 @@ sub print_grandtotals { # Get the sender by headings and results. This is complicated as we can have # different numbers of columns. - my($sender_txt_header,$sender_html_header,$sender_txt_format,$sender_html_format); + my($sender_txt_header,$sender_txt_format,$sender_html_format); my(@received_totals,@delivered_totals); my($row_tablehead, $row_max); + my(@col_headers) = ('TOTAL', 'Volume', 'Messages', 'Addresses'); foreach ('Host','Domain','Email','Edomain') { next unless $do_sender{$_}; @@ -2280,15 +2626,15 @@ sub print_grandtotals { push(@received_totals,scalar(keys %{$received_data{$_}})); push(@delivered_totals,scalar(keys %{$delivered_data{$_}})); } - $sender_html_header .= "
${_}s%d%s
%s%s%d
%s%s%s%s%d%4.1f%%%d%4.1f%%
\n"; - print $htm_fh "$sender_html_header\n"; + print $htm_fh "\n"; } - if ($xls_fh) - { - $ws_global->write($row++, $col, "Grand total summary", $f_header2); - - $row_tablehead = $row+1; # header-row of TOTALS table - - &set_worksheet_line($ws_global, $row_tablehead, 0, ['Received', 'Delivered', 'TOTAL'], $f_headertab); - - my @content= ( - "Volume", - "Messages", - $sender_txt_header, - "At least one address Delayed (Total)", - "At least one address Delayed (Percent)", - "At least one address Failed (Total)", - "At least one address Failed (Percent)" - ); - - for (my $i=0; $i < scalar(@content); $i++) - { - $ws_global->write($row_tablehead+$i+1, 2, $content[$i], $f_default); - $row++; - } - $row_max = $row_tablehead+scalar(@content)+2; # continue from this row + if ($xls_fh) { + $ws_global->write($row++, 0, "Grand total summary", $f_header2); + $ws_global->write($row, 0, \@col_headers, $f_header2); + $ws_global->merge_range($row, scalar(@col_headers), $row, scalar(@col_headers)+1, "At least one addr Delayed", $f_header2_m); + $ws_global->merge_range($row, scalar(@col_headers)+2, $row, scalar(@col_headers)+3, "At least one addr Failed", $f_header2_m); + #$ws_global->write(++$row, scalar(@col_headers), ['Total','Percent','Total','Percent'], $f_header2); } - my($volume,$failed_count); if ($merge_reports) { $volume = volume_rounded($report_totals{Received}{Volume}, $report_totals{Received}{'Volume-gigs'}); @@ -2349,7 +2676,7 @@ sub print_grandtotals { no integer; my @content=( - $volume,$total_received_count, + $volume,$total_received_count,'', @received_totals, $delayed_count, ($total_received_count) ? ($delayed_count*100/$total_received_count) : 0, @@ -2359,42 +2686,87 @@ sub print_grandtotals { printf $txt_fh ("$txt_format1$txt_format2\n", 'Received', @content) if $txt_fh; printf $htm_fh ("$htm_format1$htm_format2\n", 'Received', @content) if $htm_fh; - if ($xls_fh) - { - $row = $row_tablehead+1; - for (my $i=0; $i < scalar(@content); $i++) - { + if ($xls_fh) { + $ws_global->write(++$row, 0, 'Received', $f_default); + for (my $i=0; $i < scalar(@content); $i++) { if ($i == 4 || $i == 6) { - $ws_global->write($row+$i, 0, $content[$i]/100, $f_percent); + $ws_global->write($row, $i+1, $content[$i]/100, $f_percent); } else { - $ws_global->write($row+$i, 0, $content[$i], $f_default); + $ws_global->write($row, $i+1, $content[$i], $f_default); } } } } + if ($merge_reports) { $volume = volume_rounded($report_totals{Delivered}{Volume}, $report_totals{Delivered}{'Volume-gigs'}); - $total_delivered_count = get_report_total($report_totals{Delivered},'Messages'); + $total_delivered_messages = get_report_total($report_totals{Delivered},'Messages'); + $total_delivered_addresses = get_report_total($report_totals{Delivered},'Addresses'); } else { $volume = volume_rounded($total_delivered_data, $total_delivered_data_gigs); } - my @content=($volume, $total_delivered_count, @delivered_totals); - printf $txt_fh ("$txt_format1\n\n", 'Delivered', @content) if $txt_fh; - printf $htm_fh ("$htm_format1\n\n", 'Delivered', @content) if $htm_fh; - printf $htm_fh "
TOTALVolumeMessagesAt least one addr
Delayed
At least one addr
Failed
" . join('',@col_headers) . "At least one addr
Delayed
At least one addr
Failed
\n" if $htm_fh; - if ($xls_fh) - { + my @content=($volume, $total_delivered_messages, $total_delivered_addresses, @delivered_totals); + printf $txt_fh ("$txt_format1\n", 'Delivered', @content) if $txt_fh; + printf $htm_fh ("$htm_format1\n", 'Delivered', @content) if $htm_fh; - $row = $row_tablehead+1; - for (my $i=0; $i < scalar(@content); $i++) - { - $ws_global->write($row+$i, 1, $content[$i], $f_default); + if ($xls_fh) { + $ws_global->write(++$row, 0, 'Delivered', $f_default); + for (my $i=0; $i < scalar(@content); $i++) { + $ws_global->write($row, $i+1, $content[$i], $f_default); + } + } + + if ($merge_reports) { + foreach ('Rejects', 'Temp Rejects', 'Ham', 'Spam') { + my $messages = get_report_total($report_totals{$_},'Messages'); + my $addresses = get_report_total($report_totals{$_},'Addresses'); + if ($messages) { + @content = ($_, '', $messages, ''); + push(@content,get_report_total($report_totals{$_},'Hosts')) if $do_sender{Host}; + #These rows do not have entries for the following columns (if specified) + foreach ('Domain','Email','Edomain') { + push(@content,'') if $do_sender{$_}; + } + + printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; + printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; + $ws_global->write(++$row, 0, \@content) if $xls_fh; + } + } + } + else { + foreach my $total_aref (['Rejects',\%rejected_count_by_ip], + ['Temp Rejects',\%temporarily_rejected_count_by_ip], + ['Ham',\%ham_count_by_ip], + ['Spam',\%spam_count_by_ip]) { + #Count the number of messages of this type. + my $messages = 0; + map {$messages += $_} values %{$total_aref->[1]}; + + if ($messages > 0) { + @content = ($total_aref->[0], '', $messages, ''); + + #Count the number of distinct IPs for the Hosts column. + push(@content,scalar(keys %{$total_aref->[1]})) if $do_sender{Host}; + + #These rows do not have entries for the following columns (if specified) + foreach ('Domain','Email','Edomain') { + push(@content,'') if $do_sender{$_}; + } + + printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; + printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; + $ws_global->write(++$row, 0, \@content) if $xls_fh; } - $row = $row_max; + } } + + printf $txt_fh "\n" if $txt_fh; + printf $htm_fh "
\n" if $htm_fh; + ++$row; } @@ -2415,7 +2787,7 @@ sub print_user_patterns { print $txt_fh "\n Total\n"; } if ($htm_fh) { - print $htm_fh "

User Specified Patterns

\n"; + print $htm_fh "

User Specified Patterns

\n"; print $htm_fh "\n"; print $htm_fh "
\n"; print $htm_fh "\n"; @@ -2446,10 +2818,7 @@ sub print_user_patterns { foreach $key (@user_descriptions) { printf $txt_fh ("$txt_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $txt_fh; printf $htm_fh ("$htm_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $htm_fh; - if ($xls_fh) - { - &set_worksheet_line($ws_global, $row++, 0, [$key,$user_pattern_totals[$user_pattern_index]]); - } + $ws_global->write($row++, 0, [$key,$user_pattern_totals[$user_pattern_index]]) if $xls_fh; $user_pattern_index++; } } @@ -2469,6 +2838,79 @@ sub print_user_patterns { } } +####################################################################### +# print_rejects() +# +# print_rejects(); +# +# Print statistics about rejected mail. +####################################################################### +sub print_rejects { + my($format1,$reason); + + my $txt_format1 = " %-40s %6d"; + my $htm_format1 = ""; + + if ($txt_fh) { + print $txt_fh "Rejected mail by reason\n"; + print $txt_fh "-----------------------"; + print $txt_fh "\n Total\n"; + } + if ($htm_fh) { + print $htm_fh "

Rejected mail by reason

\n"; + print $htm_fh "
%s%d
\n"; + print $htm_fh "\n"; + } + if ($xls_fh) { + $ws_global->write($row++, $col, "Rejected mail by reason", $f_header2); + &set_worksheet_line($ws_global, $row++, 1, ["Total"], $f_headertab); + } + + + my $href = ($merge_reports) ? $report_totals{rejected_mail_by_reason} : \%rejected_count_by_reason; + my(@chartdatanames, @chartdatavals_count); + + foreach $reason (top_n_sort($topcount, $href, undef, undef)) { + printf $txt_fh ("$txt_format1\n",$reason,$href->{$reason}) if $txt_fh; + printf $htm_fh ("$htm_format1\n",$reason,$href->{$reason}) if $htm_fh; + set_worksheet_line($ws_global, $row++, 0, [$reason,$href->{$reason}], $f_default) if $xls_fh; + push(@chartdatanames, $reason); + push(@chartdatavals_count, $href->{$reason}); + } + + $row++ if $xls_fh; + print $txt_fh "\n" if $txt_fh; + + if ($htm_fh) { + print $htm_fh "
 Total
"; + if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_count > 0)) { + # calculate the graph + my @data = ( + \@chartdatanames, + \@chartdatavals_count + ); + my $graph = GD::Graph::pie->new(200, 200); + $graph->set( + x_label => 'Rejection Reasons', + y_label => 'Messages', + title => 'By count', + ); + my $gd = $graph->plot(\@data) or warn($graph->error); + if ($gd) { + open(IMG, ">$chartdir/rejections_count.png") or die "Could not write $chartdir/rejections_count.png: $!\n"; + binmode IMG; + print IMG $gd->png; + close IMG; + print $htm_fh ""; + } + } + print $htm_fh "
\n\n"; + } +} + + + + ####################################################################### # print_transport(); @@ -2492,15 +2934,13 @@ sub print_transport { print $txt_fh "\n Volume Messages\n"; } if ($htm_fh) { - print $htm_fh "

Deliveries by Transport

\n"; - print $htm_fh "\n"; - print $htm_fh "
\n"; - print $htm_fh "\n"; + print $htm_fh "

Deliveries by Transport

\n"; + print $htm_fh "
\n"; print $htm_fh "\n"; } if ($xls_fh) { - $ws_global->write($row++, $col, "Deliveries by transport", $f_header2); - &set_worksheet_line($ws_global, $row++, 1, ["Volume", "Messages"], $f_headertab); + $ws_global->write(++$row, $col, "Deliveries by transport", $f_header2); + $ws_global->write(++$row, 1, ["Volume", "Messages"], $f_headertab); } my($key); @@ -2515,9 +2955,7 @@ sub print_transport { push(@chartdatavals_vol, $report_totals{transport}{$key}{'Volume-gigs'}*$gig + $report_totals{transport}{$key}{Volume} ); printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; - if ($xls_fh) { - &set_worksheet_line($ws_global, $row++, 0, \@content, $f_default); - } + $ws_global->write(++$row, 0, \@content) if $xls_fh; } } else { @@ -2530,15 +2968,13 @@ sub print_transport { push(@chartdatavals_vol, $transported_data_gigs{$key}*$gig + $transported_data{$key}); printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; - if ($xls_fh) { - &set_worksheet_line($ws_global, $row++, 0, \@content); - } + $ws_global->write(++$row, 0, \@content) if $xls_fh; } } print $txt_fh "\n" if $txt_fh; if ($htm_fh) { - print $htm_fh "
 VolumeMessages
\n"; - print $htm_fh "
\n"; + print $htm_fh "
"; + if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_count > 0)) { # calculate the graph @@ -2561,7 +2997,7 @@ sub print_transport { print $htm_fh ""; } } - print $htm_fh "\n"; + print $htm_fh ""; if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_vol > 0)) { my @data = ( @@ -2574,19 +3010,16 @@ sub print_transport { ); my $gd = $graph->plot(\@data) or warn($graph->error); if ($gd) { - open(IMG, ">$chartdir/transports_vol.png") or die "Could not write $chartdir/transports_count.png: $!\n"; + open(IMG, ">$chartdir/transports_vol.png") or die "Could not write $chartdir/transports_vol.png: $!\n"; binmode IMG; print IMG $gd->png; close IMG; print $htm_fh ""; } } + print $htm_fh "
\n\n"; } - if ($xls_fh) { - $row++; - } - } @@ -2790,25 +3223,50 @@ sub parse_old_eximstat_reports { $end = $3 if ($3 gt $end); } elsif (/Grand total summary/) { - # Fill in $report_totals{Received|Delivered}{Volume|Messages|Hosts|Domains|...|Delayed|DelayedPercent|Failed|FailedPercent} - my(@fields); + # Fill in $report_totals{Received|Delivered}{Volume|Messages|Addresses|Hosts|Domains|...|Delayed|DelayedPercent|Failed|FailedPercent} + my(@fields, @delivered_fields); + my $doing_table = 0; while (<$fh>) { $_ = html2txt($_); #Convert general HTML markup to text. s/At least one addr//g; #Another part of the HTML output we don't want. -# TOTAL Volume Messages Hosts Domains Delayed Failed -# Received 26MB 237 177 23 8 3.4% 28 11.8% -# Delivered 13MB 233 99 88 +# TOTAL Volume Messages Addresses Hosts Domains Delayed Failed +# Received 26MB 237 177 23 8 3.4% 28 11.8% +# Delivered 13MB 233 250 99 88 if (/TOTAL\s+(.*?)\s*$/) { - @fields = split(/\s+/,$1); + $doing_table = 1; + @delivered_fields = split(/\s+/,$1); + #Delayed and Failed have two columns each, so add the extra field names in. - splice(@fields,-1,1,'DelayedPercent','Failed','FailedPercent'); + splice(@delivered_fields,-1,1,'DelayedPercent','Failed','FailedPercent'); + + # Addresses only figure in the Delivered row, so remove them from the + # normal fields. + @fields = grep !/Addresses/, @delivered_fields; } - elsif (/(Received|Delivered)\s+(.*?)\s*$/) { + elsif (/(Received)\s+(.*?)\s*$/) { print STDERR "Parsing $_" if $debug; add_to_totals($report_totals{$1},\@fields,$2); } - last if (/Delivered/); #Last line of this section. + elsif (/(Delivered)\s+(.*?)\s*$/) { + print STDERR "Parsing $_" if $debug; + add_to_totals($report_totals{$1},\@delivered_fields,$2); + my $data = $2; + # If we're merging an old report which doesn't include addresses, + # then use the Messages field instead. + unless (grep(/Addresses/, @delivered_fields)) { + my %tmp; + line_to_hash(\%tmp,\@delivered_fields,$data); + add_to_totals($report_totals{Delivered},['Addresses'],$tmp{Messages}); + } + } + elsif (/(Temp Rejects|Rejects|Ham|Spam)\s+(.*?)\s*$/) { + print STDERR "Parsing $_" if $debug; + add_to_totals($report_totals{$1},['Messages','Hosts'],$2); + } + else { + last if $doing_table; + } } } @@ -2909,16 +3367,21 @@ sub parse_old_eximstat_reports { } - my $reached_table = 0; + my ($blank_lines, $reached_table) = (0,0); while (<$fh>) { $_ = html2txt($_); #Convert general HTML markup to text. - $reached_table = 1 if (/^\s*Under/); + # The table is preceded by one blank line, and has one blank line + # following it. As the table may be empty, the best way to determine + # that we've finished it is to look for the second blank line. + ++$blank_lines if /^\s*$/; + last if ($blank_lines >=2); #Finished the table ? + $reached_table = 1 if (/\d/); next unless $reached_table; my $previous_seconds_on_queue = 0; if (/^\s*(Under|Over|)\s+(\d+[smhdw])\s+(\d+)/) { print STDERR "Parsing $_" if $debug; - my($modifier,$formated_time,$count) = ($1,$2,$3); - my $seconds = unformat_time($formated_time); + my($modifier,$formatted_time,$count) = ($1,$2,$3); + my $seconds = unformat_time($formatted_time); my $time_on_queue = ($seconds + $previous_seconds_on_queue) / 2; $previous_seconds_on_queue = $seconds; $time_on_queue = $seconds * 2 if ($modifier eq 'Over'); @@ -2932,9 +3395,6 @@ sub parse_old_eximstat_reports { $$overflow_sref += $count if ($i > $#$times_aref); } - else { - last; #Finished the table ? - } } } @@ -2992,73 +3452,88 @@ sub parse_old_eximstat_reports { #As this section processes multiple different table categories, #set up pointers to the hashes to be updated. - my($count_href,$data_href,$data_gigs_href); + my($messages_href,$addresses_href,$data_href,$data_gigs_href); if ($category =~ /local sender/) { - $count_href = \%received_count_user; + $messages_href = \%received_count_user; + $addresses_href = undef; $data_href = \%received_data_user; $data_gigs_href = \%received_data_gigs_user; } elsif ($category =~ /sending (\S+?)s?\b/) { #Top 50 sending (host|domain|email|edomain)s #Top sending (host|domain|email|edomain) - $count_href = \%{$received_count{"\u$1"}}; + $messages_href = \%{$received_count{"\u$1"}}; $data_href = \%{$received_data{"\u$1"}}; $data_gigs_href = \%{$received_data_gigs{"\u$1"}}; } elsif ($category =~ /local destination/) { - $count_href = \%delivered_count_user; + $messages_href = \%delivered_messages_user; + $addresses_href = \%delivered_addresses_user; $data_href = \%delivered_data_user; $data_gigs_href = \%delivered_data_gigs_user; } + elsif ($category =~ /local domain destination/) { + $messages_href = \%delivered_messages_local_domain; + $addresses_href = \%delivered_addresses_local_domain; + $data_href = \%delivered_data_local_domain; + $data_gigs_href = \%delivered_data_gigs_local_domain; + } elsif ($category =~ /(\S+) destination/) { #Top 50 (host|domain|email|edomain) destinations #Top (host|domain|email|edomain) destination - $count_href = \%{$delivered_count{"\u$1"}}; + $messages_href = \%{$delivered_messages{"\u$1"}}; + $addresses_href = \%{$delivered_addresses{"\u$1"}}; $data_href = \%{$delivered_data{"\u$1"}}; $data_gigs_href = \%{$delivered_data_gigs{"\u$1"}}; } + elsif ($category =~ /temporarily rejected ips/) { + $messages_href = \%temporarily_rejected_count_by_ip; + } + elsif ($category =~ /rejected ips/) { + $messages_href = \%rejected_count_by_ip; + } + elsif ($category =~ /non-rejected spamming ips/) { + $messages_href = \%spam_count_by_ip; + } + elsif ($category =~ /mail temporary rejection reasons/) { + $messages_href = \%temporarily_rejected_count_by_reason; + } + elsif ($category =~ /mail rejection reasons/) { + $messages_href = \%rejected_count_by_reason; + } my $reached_table = 0; + my $row_re; while (<$fh>) { # Watch out for empty tables. - goto PARSE_OLD_REPORT_LINE if (/

/ or /^[a-zA-Z]/); + goto PARSE_OLD_REPORT_LINE if (/

/ or (/^\s*[a-zA-Z]/ && !/^\s*Messages/)); $_ = html2txt($_); #Convert general HTML markup to text. - - $reached_table = 1 if (/^\s*\d/); + # Messages Addresses Bytes Average + if (/^\s*Messages/) { + my $pattern = '^\s*(\d+)'; + $pattern .= (/Addresses/) ? '\s+(\d+)' : '()'; + $pattern .= (/Bytes/) ? '\s+([\dKMGB]+)' : '()'; + $pattern .= (/Average/) ? '\s+[\dKMGB]+' : ''; + $pattern .= '\s+(.*?)\s*$'; + $row_re = qr/$pattern/; + $reached_table = 1; + next; + } next unless $reached_table; - # Remove optional 'average value' column. - s/^\s*(\d+)\s+(\S+)\s+(\d+(KB|MB|GB|\b)\s+)/$1 $2 /; - - if (/^\s*(\d+)\s+(\S+)\s*(.*?)\s*$/) { - my($count,$rounded_volume,$entry) = ($1,$2,$3); - #Note: $entry fields can be both null and can contain spaces. - - #Add the entry into the %table_order hash if it has a rounded volume (KB/MB/GB). - push(@{$table_order{$rounded_volume}{$by_count_or_volume}},$entry) if ($rounded_volume =~ /\D/); - - unless ($league_table_value_entered{$entry}) { - $league_table_value_entered{$entry} = 1; - unless ($$count_href{$entry}) { - $$count_href{$entry} = 0; - $$data_href{$entry} = 0; - $$data_gigs_href{$entry} = 0; - $league_table_value_was_zero{$entry} = 1; - } + my($messages, $addresses, $rounded_volume, $entry); - $$count_href{$entry} += $count; - #Add the rounded value to the data and data_gigs hashes. - un_round($rounded_volume,\$$data_href{$entry},\$$data_gigs_href{$entry}); - print STDERR "$category by $by_count_or_volume: added $count,$rounded_volume to $entry\n" if $debug; - } + if (/$row_re/) { + ($messages, $addresses, $rounded_volume, $entry) = ($1, $2, $3, $4); } - else { #Finished the table ? + else { + #Else we have finished the table and we may need to do some + #kludging to retain the order of the entries. + if ($by_count_or_volume =~ /volume/) { #Add a few bytes to appropriate entries to preserve the order. - - my($rounded_volume); foreach $rounded_volume (keys %table_order) { #For each rounded volume, we want to create a list which has things #ordered from the volume table at the front, and additional things @@ -3082,9 +3557,37 @@ sub parse_old_eximstat_reports { } } } - last; } + + # Store a new table entry. + + # Add the entry into the %table_order hash if it has a rounded + # volume (KB/MB/GB). + push(@{$table_order{$rounded_volume}{$by_count_or_volume}},$entry) if ($rounded_volume =~ /\D/); + + unless ($league_table_value_entered{$entry}) { + $league_table_value_entered{$entry} = 1; + unless ($$messages_href{$entry}) { + $$messages_href{$entry} = 0; + $$addresses_href{$entry} = 0; + $$data_href{$entry} = 0; + $$data_gigs_href{$entry} = 0; + $league_table_value_was_zero{$entry} = 1; + } + + $$messages_href{$entry} += $messages; + + # When adding the addresses, be aware that we could be merging + # an old report which does not include addresses. In this case, + # we add the messages instead. + $$addresses_href{$entry} += ($addresses) ? $addresses : $messages; + + #Add the rounded value to the data and data_gigs hashes. + un_round($rounded_volume,\$$data_href{$entry},\$$data_gigs_href{$entry}) if $rounded_volume; + print STDERR "$category by $by_count_or_volume: added $messages,$rounded_volume to $entry\n" if $debug; + } + } } elsif (/List of errors/) { @@ -3193,7 +3696,7 @@ sub update_relayed { # # add_to_totals(\%totals,\@keys,$values); # -# Given a line of space seperated values, add them into the provided hash using @keys +# Given a line of space separated values, add them into the provided hash using @keys # as the hash keys. # # If the value contains a '%', then the value is set rather than added. Otherwise, we @@ -3202,22 +3705,39 @@ sub update_relayed { sub add_to_totals { my($totals_href,$keys_aref,$values) = @_; my(@values) = split(/\s+/,$values); - my(@keys) = @$keys_aref; #Make a copy as we destroy the one we use. - my($value); - foreach $value (@values) { - my $key = shift(@keys) or next; - if ($value =~ /%/) { - $$totals_href{$key} = $value; + + for(my $i = 0; $i < @values && $i < @$keys_aref; ++$i) { + my $key = $keys_aref->[$i]; + if ($values[$i] =~ /%/) { + $$totals_href{$key} = $values[$i]; } else { $$totals_href{$key} = 0 unless ($$totals_href{$key}); $$totals_href{"$key-gigs"} = 0 unless ($$totals_href{"$key-gigs"}); - un_round($value, \$$totals_href{$key}, \$$totals_href{"$key-gigs"}); - print STDERR "Added $value to $key - $$totals_href{$key} , " . $$totals_href{"$key-gigs"} . "GB.\n" if $debug; + un_round($values[$i], \$$totals_href{$key}, \$$totals_href{"$key-gigs"}); + print STDERR "Added $values[$i] to $key - $$totals_href{$key} , " . $$totals_href{"$key-gigs"} . "GB.\n" if $debug; } } } + +####################################################################### +# line_to_hash(); +# +# line_to_hash(\%hash,\@keys,$line); +# +# Given a line of space separated values, set them into the provided hash +# using @keys as the hash keys. +####################################################################### +sub line_to_hash { + my($href,$keys_aref,$values) = @_; + my(@values) = split(/\s+/,$values); + for(my $i = 0; $i < @values && $i < @$keys_aref; ++$i) { + $$href{$keys_aref->[$i]} = $values[$i]; + } +} + + ####################################################################### # get_report_total(); # @@ -3272,7 +3792,7 @@ sub html2txt { # until we've got all of the argument. # # This isn't perfect as all white space gets reduced to one space, -# but it's as good as we can get! If it's esential that spacing +# but it's as good as we can get! If it's essential that spacing # be preserved precisely, then you get that by not using shell # variables. ####################################################################### @@ -3314,7 +3834,7 @@ sub set_worksheet_line { ####################################################################### # @rcpt_times = parse_time_list($string); # -# Parse a comma seperated list of time values in seconds given by +# Parse a comma separated list of time values in seconds given by # the user and fill an array. # # Return a default list if $string is undefined. @@ -3377,7 +3897,6 @@ $last_offset = ''; $offset_seconds = 0; $row=1; -$row_league_table=1; $col=0; $col_hist=0; $run_hist=0; @@ -3424,6 +3943,8 @@ while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') { elsif ($ARGV[0] =~ /^-byemail$/) { $do_sender{Email} = 1 } elsif ($ARGV[0] =~ /^-byemaildomain$/) { $do_sender{Edomain} = 1 } elsif ($ARGV[0] =~ /^-byedomain$/) { $do_sender{Edomain} = 1 } + elsif ($ARGV[0] =~ /^-bylocaldomain$/) { $do_local_domain = 1 } + elsif ($ARGV[0] =~ /^-emptyok$/) { $emptyOK = 1 } elsif ($ARGV[0] =~ /^-nvr$/) { $volume_rounding = 0 } elsif ($ARGV[0] =~ /^-show_rt([,\d\+\-\*\/]+)?$/) { @rcpt_times = parse_time_list($1) } elsif ($ARGV[0] =~ /^-show_dt([,\d\+\-\*\/]+)?$/) { @delivery_times = parse_time_list($1) } @@ -3460,8 +3981,7 @@ while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') { $do_sender{Host} = 1 unless ($do_sender{Domain} || $do_sender{Email} || $do_sender{Edomain}); # prepare xls Excel Workbook - if (defined $xls_fh) - { + if (defined $xls_fh) { # Create a new Excel workbook $workbook = Spreadsheet::WriteExcel->new($xls_fh); @@ -3476,9 +3996,6 @@ while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') { $ws_relayed = $workbook->addworksheet('Relayed Messages'); $ws_relayed->set_column(1, 2, 80); } - if ($topcount) { - $ws_top50 = $workbook->addworksheet('Deliveries'); - } if ($show_errors) { $ws_errors = $workbook->addworksheet('Errors'); } @@ -3506,6 +4023,13 @@ while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') { $f_header2->set_valign(); # $ws_global->write($row++, 2, "Testing Headers 2", $f_header2); + # Create another header2 for use in merged cells. + $f_header2_m = $workbook->add_format(); + $f_header2_m->set_bold(); + $f_header2_m->set_size('8'); + $f_header2_m->set_valign(); + $f_header2_m->set_align('center'); + $f_percent = $workbook->add_format(); $f_percent->set_num_format('0.0%'); @@ -3557,7 +4081,8 @@ $total_received_count = 0; $total_delivered_data = 0; $total_delivered_data_gigs = 0; -$total_delivered_count = 0; +$total_delivered_messages = 0; +$total_delivered_addresses = 0; $qt_all_overflow = 0; $qt_remote_overflow = 0; @@ -3569,7 +4094,7 @@ $message_errors = 0; $begin = "9999-99-99 99:99:99"; $end = "0000-00-00 00:00:00"; my($section,$type); -foreach $section ('Received','Delivered') { +foreach $section ('Received','Delivered','Temp Rejects', 'Rejects','Ham','Spam') { foreach $type ('Volume','Messages','Delayed','Failed','Hosts','Domains','Emails','Edomains') { $report_totals{$section}{$type} = 0; } @@ -3612,7 +4137,7 @@ else { } -if ($begin eq "9999-99-99 99:99:99") { +if ($begin eq "9999-99-99 99:99:99" && ! $emptyOK) { print STDERR "**** No valid log lines read\n"; exit 1; } @@ -3624,6 +4149,9 @@ print_grandtotals(); # Print counts of user specified patterns if required. print_user_patterns() if @user_patterns; +# Print rejection reasons. +# print_rejects(); + # Print totals by transport if required. print_transport() if $show_transport; @@ -3658,19 +4186,35 @@ print_relay() if $show_relay; # Print the league tables, if topcount isn't zero. if ($topcount > 0) { + my($ws_rej, $ws_top50, $ws_rej_row, $ws_top50_row, $ws_temp_rej, $ws_temp_rej_row); + $ws_rej_row = $ws_temp_rej_row = $ws_top50_row = 0; + if ($xls_fh) { + $ws_top50 = $workbook->addworksheet('Deliveries'); + $ws_rej = $workbook->addworksheet('Rejections') if (%rejected_count_by_reason || %rejected_count_by_ip || %spam_count_by_ip); + $ws_temp_rej = $workbook->addworksheet('Temporary Rejections') if (%temporarily_rejected_count_by_reason || %temporarily_rejected_count_by_ip); + } + + print_league_table("mail rejection reason", \%rejected_count_by_reason, undef, undef, undef, $ws_rej, \$ws_rej_row) if %rejected_count_by_reason; + print_league_table("mail temporary rejection reason", \%temporarily_rejected_count_by_reason, undef, undef, undef, $ws_temp_rej, \$ws_temp_rej_row) if %temporarily_rejected_count_by_reason; + foreach ('Host','Domain','Email','Edomain') { next unless $do_sender{$_}; - print_league_table("sending \l$_", $received_count{$_}, $received_data{$_},$received_data_gigs{$_}); + print_league_table("sending \l$_", $received_count{$_}, undef, $received_data{$_},$received_data_gigs{$_}, $ws_top50, \$ws_top50_row); } - print_league_table("local sender", \%received_count_user, - \%received_data_user,\%received_data_gigs_user) if ($local_league_table || $include_remote_users); + print_league_table("local sender", \%received_count_user, undef, + \%received_data_user,\%received_data_gigs_user, $ws_top50, \$ws_top50_row) if (($local_league_table || $include_remote_users) && %received_count_user); foreach ('Host','Domain','Email','Edomain') { next unless $do_sender{$_}; - print_league_table("\l$_ destination", $delivered_count{$_}, $delivered_data{$_},$delivered_data_gigs{$_}); + print_league_table("\l$_ destination", $delivered_messages{$_}, $delivered_addresses{$_}, $delivered_data{$_},$delivered_data_gigs{$_}, $ws_top50, \$ws_top50_row); } - print_league_table("local destination", \%delivered_count_user, - \%delivered_data_user,\%delivered_data_gigs_user) if ($local_league_table || $include_remote_users); + print_league_table("local destination", \%delivered_messages_user, \%delivered_addresses_user, \%delivered_data_user,\%delivered_data_gigs_user, $ws_top50, \$ws_top50_row) if (($local_league_table || $include_remote_users) && %delivered_messages_user); + print_league_table("local domain destination", \%delivered_messages_local_domain, \%delivered_addresses_local_domain, \%delivered_data_local_domain,\%delivered_data_gigs_local_domain, $ws_top50, \$ws_top50_row) if (($local_league_table || $include_remote_users) && %delivered_messages_local_domain); + + print_league_table("rejected ip", \%rejected_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %rejected_count_by_ip; + print_league_table("temporarily rejected ip", \%temporarily_rejected_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %temporarily_rejected_count_by_ip; + print_league_table("non-rejected spamming ip", \%spam_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %spam_count_by_ip; + } # Print the error statistics if required. @@ -3679,7 +4223,7 @@ print_errors() if $show_errors; print $htm_fh "\n\n" if $htm_fh; -$txt_fh->close if $txt_fh; +$txt_fh->close if $txt_fh && ref $txt_fh; $htm_fh->close if $htm_fh; if ($xls_fh) { @@ -3692,6 +4236,3 @@ if ($xls_fh) { # End of eximstats - - -# FIXME: Doku