3 # Copyright (c) The Exim Maintainers 2023
4 # Copyright (c) 2001-2017 University of Cambridge.
5 # See the file NOTICE for conditions of use and distribution.
6 # SPDX-License-Identifier: GPL-2.0-or-later
8 # Perl script to generate statistics from one or more Exim log files.
10 # Usage: eximstats [<options>] <log file> <log file> ...
12 # 1996-05-21: Ignore lines not starting with valid date/time, just in case
13 # these get into a log file.
14 # 1996-11-19: Add the -h option to control the size of the histogram,
15 # and optionally turn it off.
16 # Use some Perl 5 things; it should be everywhere by now.
17 # Add the Perl -w option and rewrite so no warnings are given.
18 # Add the -t option to control the length of the "top" listing.
19 # Add the -ne, -nt options to turn off errors and transport
21 # Add information about length of time on queue, and -q<list> to
22 # control the intervals and turn it off.
23 # Add count and percentage of delayed messages to the Received
25 # Show total number of errors.
26 # Add count and percentage of messages with errors to Received
28 # Add information about relaying and -nr to suppress it.
29 # 1997-02-03 Merged in some of the things Nigel Metheringham had done:
31 # Added received histogram as well as delivered
32 # Added local senders' league table
33 # Added local recipients' league table
34 # 1997-03-10 Fixed typo "destinationss"
35 # Allow for intermediate address between final and original
36 # when testing for relaying
37 # Give better message when no input
38 # 1997-04-24 Fixed bug in layout of error listing that was depending on
39 # text length (output line got repeated).
40 # 1997-05-06 Bug in option decoding when only one option.
41 # Overflow bug when handling very large volumes.
42 # 1997-10-28 Updated to handle revised log format that might show
43 # HELO name as well as host name before IP number
44 # 1998-01-26 Bugs in the function for calculating the number of seconds
45 # since 1970 from a log date
46 # 1998-02-02 Delivery to :blackhole: doesn't have a T= entry in the log
47 # line; cope with this, thereby avoiding undefined problems
48 # Very short log line gave substring error
49 # 1998-02-03 A routed delivery to a local transport may not have <> in the
50 # log line; terminate the address at white space, not <
51 # 1998-09-07 If first line of input was a => line, $thissize was undefined;
53 # 1998-12-21 Adding of $thissize from => line should have been adding $size.
54 # Oops. Should have looked more closely when fixing the previous
56 # 1999-11-12 Increased the field widths for printed integers; numbers are
57 # bigger than originally envisaged.
58 # 2001-03-21 Converted seconds() routine to use Time::Local, fixing a bug
59 # whereby seconds($timestamp) - id_seconds($id) gave an
61 # Added POD documentation.
62 # Moved usage instructions into help() subroutine.
63 # Added 'use strict' and declared all global variables.
64 # Added '-html' flag and resultant code.
65 # Added '-cache' flag and resultant code.
66 # Added add_volume() routine and converted all volume variables
67 # to use it, fixing the overflow problems for individual hosts
69 # Converted all volume output to GB/MB/KB as appropriate.
70 # Don't store local user stats if -nfl is specified.
71 # Modifications done by: Steve Campbell (<steve@computurn.com>)
72 # 2001-04-02 Added the -t_remote_users flag. Steve Campbell.
73 # 2001-10-15 Added the -domain flag. Steve Campbell.
74 # 2001-10-16 Accept files on STDIN or on the command line. Steve Campbell.
75 # 2001-10-21 Removed -domain flag and added -bydomain, -byhost, and -byemail.
76 # We now generate our main parsing subroutine as an eval statement
77 # which improves performance dramatically when not all the results
78 # are required. We also cache the last timestamp to time conversion.
80 # NOTE: 'Top 50 destinations by (message count|volume)' lines are
81 # now 'Top N (host|email|domain) destinations by (message count|volume)'
82 # where N is the topcount. Steve Campbell.
84 # 2001-10-30 V1.16 Joachim Wieland.
85 # Fixed minor bugs in add_volume() when taking over this version
86 # for use in Exim 4: -w gave uninitialized value warnings in
87 # two situations: for the first addition to a counter, and if
88 # there were never any gigabytes, thereby leaving the $gigs
90 # Initialized $last_timestamp to stop a -w uninitialized warning.
91 # Minor layout tweak for grand totals (nitpicking).
92 # Put the IP addresses for relaying stats in [] and separated by
93 # a space from the domain name.
94 # Removed the IPv4-specific address test when picking out addresses
95 # for relaying. Anything inside [] is OK.
97 # 2002-07-02 Philip Hazel
98 # Fixed "uninitialized variable" message that occurred for relay
99 # messages that arrived from H=[1.2.3.4] hosts (no name shown).
100 # This bug didn't affect the output.
102 # 2002-04-15 V1.17 Joachim Wieland.
103 # Added -charts, -chartdir. -chartrel options which use
104 # GD::Graph modules to create graphical charts of the statistics.
106 # 2002-04-15 V1.18 Steve Campbell.
107 # Added a check for $domain to to stop a -w uninitialized warning.
108 # Added -byemaildomain option.
109 # Only print HTML header links to included tables!
111 # 2002-08-02 V1.19 Steve Campbell.
112 # Changed the debug mode to dump the parser onto STDERR rather
113 # than STDOUT. Documented the -d flag into the help().
114 # Rejoined the divergent 2002-04-15 and 2002-07-02 releases.
116 # 2002-08-21 V1.20 Steve Campbell.
117 # Added the '-merge' option to allow merging of previous reports.
118 # Fixed a missing semicolon when doing -bydomain.
119 # Make volume charts plot the data gigs and bytes rather than just bytes.
120 # Only process log lines with $flag =~ /<=|=>|->|==|\*\*|Co/
121 # Converted Emaildomain to Edomain - the column header was too wide!
122 # This changes the text output slightly. You can revert to the old
123 # column widths by changing $COLUMN_WIDTHS to 7;
125 # 2002-09-04 V1.21 Andreas J Mueller
126 # Local deliveries domain now defaults to 'localdomain'.
127 # Don't match F=<From> when looking for the user.
129 # 2002-09-05 V1.22 Steve Campbell
130 # Fixed a perl 5.005 incompatibility problem ('our' variables).
132 # 2002-09-11 V1.23 Steve Campbell
133 # Stopped -charts option from throwing errors on null data.
134 # Don't print out 'Errors encountered' unless there are any.
136 # 2002-10-21 V1.23a Philip Hazel - patch from Tony Finch put in until
137 # Steve's eximstats catches up.
138 # Handle log files that include the timezone after the timestamp.
139 # Switch to assuming that log timestamps are in local time, with
140 # an option for UTC timestamps, as in Exim itself.
142 # 2003-02-05 V1.24 Steve Campbell
143 # Added in Sergey Sholokh's code to convert '<' and '>' characters
144 # in HTML output. Also added code to convert them back with -merge.
145 # Fixed timestamp offsets to convert to seconds rather than minutes.
146 # Updated -merge to work with output files using timezones.
147 # Added caching to speed up the calculation of timezone offsets.
149 # 2003-02-07 V1.25 Steve Campbell
150 # Optimised the usage of mktime() in the seconds subroutine.
151 # Removed the now redundant '-cache' option.
152 # html2txt() now explicitly matches HTML tags.
153 # Implemented a new sorting algorithm - the top_n_sort() routine.
154 # Added Danny Carroll's '-nvr' flag and code.
156 # 2003-03-13 V1.26 Steve Campbell
157 # Implemented HTML compliance changes recommended by Bernard Massot.
158 # Bug fix to allow top_n_sort() to handle null keys.
159 # Convert all domains and edomains to lowercase.
160 # Remove preceding dots from domains.
162 # 2003-03-13 V1.27 Steve Campbell
163 # Replaced border attributes with 'border=1', as recommended by
166 # 2003-06-03 V1.28 John Newman
167 # Added in the ability to skip over the parsing and evaluation of
168 # specific transports as passed to eximstats via the new "-nt/.../"
169 # command line argument. This new switch allows the viewing of
170 # not more accurate statistics but more applicable statistics when
171 # special transports are in use (ie; SpamAssassin). We need to be
172 # able to ignore transports such as this otherwise the resulting
173 # local deliveries are significantly skewed (doubled)...
175 # 2003-11-06 V1.29 Steve Campbell
176 # Added the '-pattern "Description" "/pattern/"' option.
178 # 2004-02-17 V1.30 Steve Campbell
179 # Added warnings if required GD::Graph modules are not available or
180 # insufficient -chart* options are specified.
182 # 2004-02-20 V1.31 Andrea Balzi
183 # Only show the Local Sender/Destination links if the tables exist.
185 # 2004-07-05 V1.32 Steve Campbell
186 # Fix '-merge -h0' divide by zero error.
188 # 2004-07-15 V1.33 Steve Campbell
189 # Documentation update - I've converted the subroutine
190 # documentation from POD to comments.
192 # 2004-12-10 V1.34 Steve Campbell
193 # Eximstats can now parse syslog lines as well as mainlog lines.
195 # 2004-12-20 V1.35 Wouter Verhelst
196 # Pie charts by volume were actually generated by count. Fixed.
198 # 2005-02-07 V1.36 Gregor Herrmann / Steve Campbell
199 # Added average sizes to HTML Top tables.
201 # 2005-04-26 V1.37 Frank Heydlauf
202 # Added -xls and the ability to specify output files.
204 # 2005-04-29 V1.38 Steve Campbell
205 # Use FileHandles for outputting results.
206 # Allow any combination of xls, txt, and html output.
207 # Fixed display of large numbers with -nvr option
208 # Fixed merging of reports with empty tables.
210 # 2005-05-27 V1.39 Steve Campbell
211 # Added the -include_original_destination flag
212 # Removed tabs and trailing whitespace.
214 # 2005-06-03 V1.40 Steve Campbell
215 # Whilst parsing the mainlog(s), store information about
216 # the messages in a hash of arrays rather than using
217 # individual hashes. This is a bit cleaner and results in
218 # dramatic memory savings, albeit at a slight CPU cost.
220 # 2005-06-15 V1.41 Steve Campbell
221 # Added the -show_rt<list> flag.
222 # Added the -show_dt<list> flag.
224 # 2005-06-24 V1.42 Steve Campbell
225 # Added Histograms for user specified patterns.
227 # 2005-06-30 V1.43 Steve Campbell
228 # Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
230 # 2005-07-26 V1.44 Steve Campbell
231 # Use a glob alias rather than an array ref in the generated
232 # parser. This improves both readability and performance.
234 # 2005-09-30 V1.45 Marco Gaiarin / Steve Campbell
235 # Collect SpamAssassin and rejection statistics.
236 # Don't display local sender or destination tables unless
237 # there is data to show.
238 # Added average volumes into the top table text output.
240 # 2006-02-07 V1.46 Steve Campbell
241 # Collect data on the number of addresses (recipients)
242 # as well as the number of messages.
244 # 2006-05-05 V1.47 Steve Campbell
245 # Added 'Message too big' to the list of mail rejection
246 # reasons (thanks to Marco Gaiarin).
248 # 2006-06-05 V1.48 Steve Campbell
249 # Mainlog lines which have GMT offsets and are too short to
250 # have a flag are now skipped.
252 # 2006-11-10 V1.49 Alain Williams
253 # Added the -emptyok flag.
255 # 2006-11-16 V1.50 Steve Campbell
256 # Fixes for obtaining the IP address from reject messages.
258 # 2006-11-27 V1.51 Steve Campbell
259 # Another update for obtaining the IP address from reject messages.
261 # 2006-11-27 V1.52 Steve Campbell
262 # Tally any reject message containing SpamAssassin.
264 # 2007-01-31 V1.53 Philip Hazel
265 # Allow for [pid] after date in log lines
267 # 2007-02-14 V1.54 Daniel Tiefnig
268 # Improved the '($parent) =' pattern match.
270 # 2007-03-19 V1.55 Steve Campbell
271 # Differentiate between permanent and temporary rejects.
273 # 2007-03-29 V1.56 Jez Hancock
274 # Fixed some broken HTML links and added missing column headers.
276 # 2007-03-30 V1.57 Steve Campbell
277 # Fixed Grand Total Summary Domains, Edomains, and Email columns
278 # for Rejects, Temp Rejects, Ham, and Spam rows.
280 # 2007-04-11 V1.58 Steve Campbell
281 # Fix to get <> and blackhole to show in edomain tables.
283 # 2007-09-20 V1.59 Steve Campbell
284 # Added the -bylocaldomain option
286 # 2007-09-20 V1.60 Heiko Schlittermann
287 # Fix for misinterpreted log lines
289 # 2013-01-14 V1.61 Steve Campbell
290 # Watch out for senders sending "HELO [IpAddr]"
293 # For documentation on the logfile format, see
294 # http://www.exim.org/exim-html-4.50/doc/html/spec_48.html#IX2793
298 eximstats - generates statistics from Exim mainlog or syslog files.
302 eximstats [Output] [Options] mainlog1 mainlog2 ...
303 eximstats -merge [Options] report.1.txt report.2.txt ... > weekly_report.txt
311 Output the results in plain text to STDOUT.
313 =item B<-txt>=I<filename>
315 Output the results in plain text. Filename '-' for STDOUT is accepted.
319 Output the results in HTML to STDOUT.
321 =item B<-html>=I<filename>
323 Output the results in HTML. Filename '-' for STDOUT is accepted.
327 Output the results in Excel compatible Format to STDOUT.
328 Requires the Spreadsheet::WriteExcel CPAN module.
330 =item B<-xls>=I<filename>
332 Output the results in Excel compatible format. Filename '-' for STDOUT is accepted.
343 histogram divisions per hour. The default is 1, and
344 0 suppresses histograms. Valid values are:
346 0, 1, 2, 3, 5, 10, 15, 20, 30 or 60.
350 Don't display error information.
354 Don't display relaying information.
356 =item B<-nr>I</pattern/>
358 Don't display relaying information that matches.
362 Don't display transport information.
364 =item B<-nt>I</pattern/>
366 Don't display transport information that matches
370 List of times for queuing information single 0 item suppresses.
374 Display top <number> sources/destinations
375 default is 50, 0 suppresses top listing.
379 Omit local sources/destinations in top listing.
381 =item B<-t_remote_users>
383 Include remote users in the top source/destination listings.
385 =item B<-include_original_destination>
387 Include the original destination email addresses rather than just
388 using the final ones.
389 Useful for finding out which of your mailing lists are receiving mail.
391 =item B<-show_dt>I<list>
393 Show the delivery times (B<DT>)for all the messages.
395 Exim must have been configured to use the +deliver_time logging option
396 for this option to work.
398 I<list> is an optional list of times. Eg -show_dt1,2,4,8 will show
399 the number of messages with delivery times under 1 second, 2 seconds, 4 seconds,
400 8 seconds, and over 8 seconds.
402 =item B<-show_rt>I<list>
404 Show the receipt times for all the messages. The receipt time is
405 defined as the Completed hh:mm:ss - queue_time_overall - the Receipt hh:mm:ss.
406 These figures will be skewed by pipelined messages so might not be that useful.
408 Exim must have been configured to use the +queue_time_overall logging option
409 for this option to work.
411 I<list> is an optional list of times. Eg -show_rt1,2,4,8 will show
412 the number of messages with receipt times under 1 second, 2 seconds, 4 seconds,
413 8 seconds, and over 8 seconds.
417 Show results by sending host. This may be combined with
418 B<-bydomain> and/or B<-byemail> and/or B<-byedomain>. If none of these options
419 are specified, then B<-byhost> is assumed as a default.
423 Show results by sending domain.
424 May be combined with B<-byhost> and/or B<-byemail> and/or B<-byedomain>.
428 Show results by sender's email address.
429 May be combined with B<-byhost> and/or B<-bydomain> and/or B<-byedomain>.
431 =item B<-byemaildomain> or B<-byedomain>
433 Show results by sender's email domain.
434 May be combined with B<-byhost> and/or B<-bydomain> and/or B<-byemail>.
436 =item B<-pattern> I<Description> I</Pattern/>
438 Look for the specified pattern and count the number of lines in which it appears.
439 This option can be specified multiple times. Eg:
441 -pattern 'Refused connections' '/refused connection/'
446 This option allows eximstats to merge old eximstat reports together. Eg:
448 eximstats mainlog.sun > report.sun.txt
449 eximstats mainlog.mon > report.mon.txt
450 eximstats mainlog.tue > report.tue.txt
451 eximstats mainlog.wed > report.web.txt
452 eximstats mainlog.thu > report.thu.txt
453 eximstats mainlog.fri > report.fri.txt
454 eximstats mainlog.sat > report.sat.txt
455 eximstats -merge report.*.txt > weekly_report.txt
456 eximstats -merge -html report.*.txt > weekly_report.html
462 You can merge text or html reports and output the results as text or html.
466 You can use all the normal eximstat output options, but only data
467 included in the original reports can be shown!
471 When merging reports, some loss of accuracy may occur in the top I<n> lists.
472 This will be towards the ends of the lists.
476 The order of items in the top I<n> lists may vary when the data volumes
477 round to the same value.
483 Create graphical charts to be displayed in HTML output.
484 Only valid in combination with I<-html>.
486 This requires the following modules which can be obtained
487 from http://www.cpan.org/modules/01modules.index.html
499 To install these, download and unpack them, then use the normal perl installation procedure:
506 =item B<-chartdir>I <dir>
508 Create the charts in the directory <dir>
510 =item B<-chartrel>I <dir>
512 Specify the relative directory for the "img src=" tags from where to include
517 Specify that it's OK to not find any valid log lines. Without this
518 we will output an error message if we don't find any.
522 Debug flag. This outputs the eval()'d parser onto STDOUT which makes it
523 easier to trap errors in the eval section. Remember to add 1 to the line numbers to allow for the
530 Eximstats parses exim mainlog and syslog files to output a statistical
531 analysis of the messages processed. By default, a text
532 analysis is generated, but you can request other output formats
533 using flags. See the help (B<-help>) to learn
534 about how to create charts from the tables.
538 There is a website at https://www.exim.org - this contains details of the
539 mailing list exim-users@exim.org.
543 This program does not perfectly handle messages whose received
544 and delivered log lines are in different files, which can happen
545 when you have multiple mail servers and a message cannot be
546 immediately delivered. Fixing this could be tricky...
548 Merging of xls files is not (yet) possible. Be free to implement :)
554 BEGIN { pop @INC if $INC[-1] eq '.' };
559 # use Time::Local; # PH/FANF
562 if (@ARGV and ($ARGV[0] eq '--version' || $ARGV[0] eq '-v')) {
563 print basename($0) . ": $0\n",
564 "build: EXIM_RELEASE_VERSIONEXIM_VARIANT_VERSION\n",
565 "perl(runtime): $]\n";
569 use vars qw($HAVE_GD_Graph_pie $HAVE_GD_Graph_linespoints $HAVE_Spreadsheet_WriteExcel);
570 eval { require GD::Graph::pie; };
571 $HAVE_GD_Graph_pie = $@ ? 0 : 1;
572 eval { require GD::Graph::linespoints; };
573 $HAVE_GD_Graph_linespoints = $@ ? 0 : 1;
574 eval { require Spreadsheet::WriteExcel; };
575 $HAVE_Spreadsheet_WriteExcel = $@ ? 0 : 1;
578 ##################################################
580 ##################################################
581 # 'use vars' instead of 'our' as perl5.005 is still in use out there!
582 use vars qw(@tab62 @days_per_month $gig);
583 use vars qw($VERSION);
584 use vars qw($COLUMN_WIDTHS);
585 use vars qw($WEEK $DAY $HOUR $MINUTE);
589 (0,1,2,3,4,5,6,7,8,9,0,0,0,0,0,0, # 0-9
590 0,10,11,12,13,14,15,16,17,18,19,20, # A-K
591 21,22,23,24,25,26,27,28,29,30,31,32, # L-W
592 33,34,35, 0, 0, 0, 0, 0, # X-Z
593 0,36,37,38,39,40,41,42,43,44,45,46, # a-k
594 47,48,49,50,51,52,53,54,55,56,57,58, # l-w
597 @days_per_month = (0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334);
598 $gig = 1024 * 1024 * 1024;
601 # How much space do we allow for the Hosts/Domains/Emails/Edomains column headers?
605 $HOUR = 60 * $MINUTE;
609 # Declare global variables.
610 use vars qw($total_received_data $total_received_data_gigs $total_received_count);
611 use vars qw($total_delivered_data $total_delivered_data_gigs $total_delivered_messages $total_delivered_addresses);
612 use vars qw(%timestamp2time); #Hash of timestamp => time.
613 use vars qw($last_timestamp $last_time); #The last time conversion done.
614 use vars qw($last_date $date_seconds); #The last date conversion done.
615 use vars qw($last_offset $offset_seconds); #The last time offset conversion done.
616 use vars qw($localtime_offset);
617 use vars qw($i); #General loop counter.
618 use vars qw($debug); #Debug mode?
619 use vars qw($ntopchart); #How many entries should make it into the chart?
620 use vars qw($gddirectory); #Where to put files from GD::Graph
622 # SpamAssassin variables
623 use vars qw($spam_score $spam_score_gigs);
624 use vars qw($ham_score $ham_score_gigs);
625 use vars qw(%ham_count_by_ip %spam_count_by_ip);
626 use vars qw(%rejected_count_by_ip %rejected_count_by_reason);
627 use vars qw(%temporarily_rejected_count_by_ip %temporarily_rejected_count_by_reason);
629 #For use in Spreadsheet::WriteExcel
630 use vars qw($workbook $ws_global $ws_relayed $ws_errors);
631 use vars qw($row $col $row_hist $col_hist);
632 use vars qw($run_hist);
633 use vars qw($f_default $f_header1 $f_header2 $f_header2_m $f_headertab $f_percent); #Format Header
636 use vars qw($txt_fh $htm_fh $xls_fh);
640 # The following are parameters whose values are
641 # set by command line switches:
642 use vars qw($show_errors $show_relay $show_transport $transport_pattern);
643 use vars qw($topcount $local_league_table $include_remote_users $do_local_domain);
644 use vars qw($hist_opt $hist_interval $hist_number $volume_rounding $emptyOK);
645 use vars qw($relay_pattern @queue_times @user_patterns @user_descriptions);
646 use vars qw(@rcpt_times @delivery_times);
647 use vars qw($include_original_destination);
648 use vars qw($txt_fh $htm_fh $xls_fh);
650 use vars qw(%do_sender); #Do sender by Host, Domain, Email, and/or Edomain tables.
651 use vars qw($charts $chartrel $chartdir $charts_option_specified);
652 use vars qw($merge_reports); #Merge old reports ?
654 # The following are modified in the parse() routine, and
655 # referred to in the print_*() routines.
656 use vars qw($delayed_count $relayed_unshown $begin $end);
657 use vars qw(%messages @message);
658 use vars qw(%received_count %received_data %received_data_gigs);
659 use vars qw(%delivered_messages %delivered_data %delivered_data_gigs %delivered_addresses);
660 use vars qw(%received_count_user %received_data_user %received_data_gigs_user);
661 use vars qw(%delivered_messages_user %delivered_addresses_user %delivered_data_user %delivered_data_gigs_user);
662 use vars qw(%delivered_messages_local_domain %delivered_addresses_local_domain %delivered_data_local_domain %delivered_data_gigs_local_domain);
663 use vars qw(%transported_count %transported_data %transported_data_gigs);
664 use vars qw(%relayed %errors_count $message_errors);
665 use vars qw(@qt_all_bin @qt_remote_bin);
666 use vars qw($qt_all_overflow $qt_remote_overflow);
667 use vars qw(@dt_all_bin @dt_remote_bin %rcpt_times_bin);
668 use vars qw($dt_all_overflow $dt_remote_overflow %rcpt_times_overflow);
669 use vars qw(@received_interval_count @delivered_interval_count);
670 use vars qw(@user_pattern_totals @user_pattern_interval_count);
672 use vars qw(%report_totals);
675 use vars qw($SIZE $FROM_HOST $FROM_ADDRESS $ARRIVAL_TIME $REMOTE_DELIVERED $PROTOCOL);
676 use vars qw($DELAYED $HAD_ERROR);
681 $REMOTE_DELIVERED = 4;
688 ##################################################
690 ##################################################
692 #######################################################################
693 # get_filehandle($file,\%output_files);
694 # Return a filehandle writing to $file.
696 # If %output_files is defined, check that $output_files{$file}
697 # doesn't exist and die if it does, or set it if it doesn't.
698 #######################################################################
700 my($file,$output_files_href) = @_;
702 $file = '-' if ($file eq '');
704 if (defined $output_files_href) {
705 die "You can only output to '$file' once! Use -h for help.\n" if exists $output_files_href->{$file};
706 $output_files_href->{$file} = 1;
714 unlink $file or die "Failed to rm $file: $!";
717 my $fh = new IO::File $file, O_WRONLY|O_CREAT|O_EXCL;
718 die "new IO::File $file failed: $!" unless (defined $fh);
723 #######################################################################
726 # $rounded_volume = volume_rounded($bytes,$gigabytes);
728 # Given a data size in bytes, round it to KB, MB, or GB
731 # Eg 12000 => 12KB, 15000000 => 14GB, etc.
733 # Note: I've experimented with Math::BigInt and it results in a 33%
734 # performance degredation as opposed to storing numbers split into
735 # bytes and gigabytes.
736 #######################################################################
748 if ($volume_rounding) {
752 $rounded = sprintf("%6d", $x);
754 elsif ($x < 10000000) {
755 $rounded = sprintf("%4dKB", ($x + 512)/1024);
758 $rounded = sprintf("%4dMB", ($x + 512*1024)/(1024*1024));
761 # Values between 1GB and 10GB are printed in MB
763 $rounded = sprintf("%4dMB", ($g * 1024) + ($x + 512*1024)/(1024*1024));
766 # Handle values over 10GB
767 $rounded = sprintf("%4dGB", $g + ($x + $gig/2)/$gig);
771 # We don't want any rounding to be done.
772 # and we don't need broken formatted output which on one hand avoids numbers from
773 # being interpreted as string by Spreadsheet Calculators, on the other hand
774 # breaks if more than 4 digits! -> flexible length instead of fixed length
775 # Format the return value at the output routine! -fh
776 #$rounded = sprintf("%d", ($g * $gig) + $x);
778 $rounded = sprintf("%.0f", ($g * $gig) + $x);
785 #######################################################################
788 # un_round($rounded_volume,\$bytes,\$gigabytes);
790 # Given a volume in KB, MB or GB, as generated by volume_rounded(),
791 # do the reverse transformation and convert it back into Bytes and Gigabytes.
792 # These are added to the $bytes and $gigabytes parameters.
794 # Given a data size in bytes, round it to KB, MB, or GB
797 # EG: 500 => (500,0), 14GB => (0,14), etc.
798 #######################################################################
800 my($rounded,$bytes_sref,$gigabytes_sref) = @_;
802 if ($rounded =~ /(\d+)GB/) {
803 $$gigabytes_sref += $1;
805 elsif ($rounded =~ /(\d+)MB/) {
806 $$gigabytes_sref += $1 / 1024;
807 $$bytes_sref += (($1 % 1024 ) * 1024 * 1024);
809 elsif ($rounded =~ /(\d+)KB/) {
810 $$gigabytes_sref += $1 / (1024 * 1024);
811 $$bytes_sref += ($1 % (1024 * 1024) * 1024);
813 elsif ($rounded =~ /(\d+)/) {
814 # We need to turn off integer in case we are merging an -nvr report.
816 $$gigabytes_sref += int($1 / $gig);
817 $$bytes_sref += $1 % $gig;
820 #Now reduce the bytes down to less than 1GB.
821 add_volume($bytes_sref,$gigabytes_sref,0) if ($$bytes_sref > $gig);
825 #######################################################################
828 # add_volume(\$bytes,\$gigs,$size);
830 # Add $size to $bytes/$gigs where this is a number split into
831 # bytes ($bytes) and gigabytes ($gigs). This is significantly
832 # faster than using Math::BigInt.
833 #######################################################################
835 my($bytes_ref,$gigs_ref,$size) = @_;
836 $$bytes_ref = 0 if ! defined $$bytes_ref;
837 $$gigs_ref = 0 if ! defined $$gigs_ref;
838 $$bytes_ref += $size;
839 while ($$bytes_ref > $gig) {
846 #######################################################################
849 # $formatted_time = format_time($seconds);
851 # Given a time in seconds, break it down into
852 # weeks, days, hours, minutes, and seconds.
854 # Eg 12005 => 3h20m5s
855 #######################################################################
867 $p .= "$w"."w" if $w > 0;
868 $p .= "$d"."d" if $d > 0;
869 $p .= "$h"."h" if $h > 0;
870 $p .= "$m"."m" if $m > 0;
871 $p .= "$s"."s" if $s > 0 || $p eq "";
876 #######################################################################
879 # $seconds = unformat_time($formatted_time);
881 # Given a time in weeks, days, hours, minutes, or seconds, convert it to seconds.
883 # Eg 3h20m5s => 12005
884 #######################################################################
886 my($formatted_time) = pop @_;
889 while ($formatted_time =~ s/^(\d+)([wdhms]?)//) {
890 $time += $1 if ($2 eq '' || $2 eq 's');
891 $time += $1 * 60 if ($2 eq 'm');
892 $time += $1 * 60 * 60 if ($2 eq 'h');
893 $time += $1 * 60 * 60 * 24 if ($2 eq 'd');
894 $time += $1 * 60 * 60 * 24 * 7 if ($2 eq 'w');
900 #######################################################################
903 # $time = seconds($timestamp);
905 # Given a time-of-day timestamp, convert it into a time() value using
906 # POSIX::mktime. We expect the timestamp to be of the form
907 # "$year-$mon-$day $hour:$min:$sec", with month going from 1 to 12,
908 # and the year to be absolute (we do the necessary conversions). The
909 # seconds value can be followed by decimals, which we ignore. The
910 # timestamp may be followed with an offset from UTC like "+$hh$mm"; if the
911 # offset is not present, and we have not been told that the log is in UTC
912 # (with the -utc option), then we adjust the time by the current local
913 # time offset so that it can be compared with the time recorded in message
916 # To improve performance, we only use mktime on the date ($year-$mon-$day),
917 # and only calculate it if the date is different to the previous time we
918 # came here. We then add on seconds for the '$hour:$min:$sec'.
920 # We also store the results of the last conversion done, and only
921 # recalculate if the date is different.
923 # We used to have the '-cache' flag which would store the results of the
924 # mktime() call. However, the current way of just using mktime() on the
925 # date obsoletes this.
926 #######################################################################
930 # Is the timestamp the same as the last one?
931 return $last_time if ($last_timestamp eq $timestamp);
933 return 0 unless ($timestamp =~ /^((\d{4})\-(\d\d)-(\d\d))\s(\d\d):(\d\d):(\d\d)(?:\.\d+)?( ([+-])(\d\d)(\d\d))?/o);
935 unless ($last_date eq $1) {
937 my(@timestamp) = (0,0,0,$4,$3,$2);
938 $timestamp[5] -= 1900;
940 $date_seconds = mktime(@timestamp);
942 my $time = $date_seconds + ($5 * 3600) + ($6 * 60) + $7;
944 # SC. Use caching. Also note we want seconds not minutes.
945 #my($this_offset) = ($10 * 60 + $12) * ($9 . "1") if defined $8;
946 if (defined $8 && ($8 ne $last_offset)) {
948 $offset_seconds = ($10 * 60 + $11) * 60;
949 $offset_seconds = -$offset_seconds if ($9 eq '-');
954 #$time -= $this_offset;
955 $time -= $offset_seconds;
956 } elsif (defined $localtime_offset) {
957 $time -= $localtime_offset;
960 # Store the last timestamp received.
961 $last_timestamp = $timestamp;
968 #######################################################################
971 # $time = id_seconds($message_id);
973 # Given a message ID, convert it into a time() value.
974 #######################################################################
976 my($sub_id) = substr((pop @_), 0, 6);
978 my(@c) = split(//, $sub_id);
979 while($#c >= 0) { $s = $s * 62 + $tab62[ord(shift @c) - ord('0')] }
983 #######################################################################
986 # $seconds = wdhms_seconds($string);
988 # Convert a string in a week/day/hour/minute/second format (eg 4h10s)
990 #######################################################################
992 if ($_[0] =~ /^(?:(\d+)w)?(?:(\d+)d)?(?:(\d+)h)?(?:(\d+)m)?(?:(\d+)s)?/) {
993 return((($1||0) * $WEEK) + (($2||0) * $DAY) + (($3||0) * $HOUR) + (($4||0) * $MINUTE) + ($5||0));
998 #######################################################################
1001 # $queued = queue_time($completed_tod, $arrival_time, $id);
1003 # Given the completed time of day and either the arrival time
1004 # (preferred), or the message ID, calculate how long the message has
1005 # been on the queue.
1007 #######################################################################
1009 my($completed_tod, $arrival_time, $id) = @_;
1011 # Note: id_seconds() benchmarks as 42% slower than seconds()
1012 # and computing the time accounts for a significant portion of
1014 if (defined $arrival_time) {
1015 return(seconds($completed_tod) - seconds($arrival_time));
1018 return(seconds($completed_tod) - id_seconds($id));
1023 #######################################################################
1024 # calculate_localtime_offset();
1026 # $localtime_offset = calculate_localtime_offset();
1028 # Calculate the the localtime offset from gmtime in seconds.
1030 # $localtime = time() + $localtime_offset.
1032 # These are the same semantics as ISO 8601 and RFC 2822 timezone offsets.
1033 # (West is negative, East is positive.)
1034 #######################################################################
1036 # $localtime = gmtime() + $localtime_offset. OLD COMMENT
1037 # This subroutine commented out as it's not currently in use.
1039 #sub calculate_localtime_offset {
1040 # # Pick an arbitrary date, convert it to localtime & gmtime, and return the difference.
1041 # my (@sample_date) = (0,0,0,5,5,100);
1042 # my $localtime = timelocal(@sample_date);
1043 # my $gmtime = timegm(@sample_date);
1044 # my $offset = $localtime - $gmtime;
1048 sub calculate_localtime_offset {
1049 # Assume that the offset at the moment is valid across the whole
1050 # period covered by the logs that we're analysing. This may not
1051 # be true around the time the clocks change in spring or autumn.
1053 # mktime works on local time and gmtime works in UTC
1054 my $local = mktime(gmtime($utc));
1055 return $local - $utc;
1060 #######################################################################
1061 # print_duration_table();
1063 # print_duration_table($title, $message_type, \@times, \@values, $overflow);
1065 # Print a table showing how long a particular step took for
1066 # the messages. The parameters are:
1067 # $title Eg "Time spent on the queue"
1068 # $message_type Eg "Remote"
1069 # \@times The maximum time a message took for it to increment
1070 # the corresponding @values counter.
1071 # \@values An array of message counters.
1072 # $overflow The number of messages which exceeded the maximum
1074 #######################################################################
1075 sub print_duration_table {
1077 my($title, $message_type, $times_aref, $values_aref, $overflow) = @_;
1078 my(@chartdatanames);
1081 my $printed_one = 0;
1082 my $cumulative_percent = 0;
1084 my $queue_total = $overflow;
1085 map {$queue_total += $_} @$values_aref;
1087 my $temp = "$title: $message_type";
1090 my $txt_format = "%5s %4s %6d %5.1f%% %5.1f%%\n";
1091 my $htm_format = "<tr><td align=\"right\">%s %s</td><td align=\"right\">%d</td><td align=\"right\">%5.1f%%</td><td align=\"right\">%5.1f%%</td>\n";
1094 printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh;
1096 print $htm_fh "<hr><a name=\"$title $message_type\"></a><h2>$temp</h2>\n";
1097 print $htm_fh "<table border=0 width=\"100%\"><tr><td><table border=1>\n";
1098 print $htm_fh "<tr><th>Time</th><th>Messages</th><th>Percentage</th><th>Cumulative Percentage</th>\n";
1101 $ws_global->write($row++, $col, "$title: ".$message_type, $f_header2);
1102 my @content=("Time", "Messages", "Percentage", "Cumulative Percentage");
1103 &set_worksheet_line($ws_global, $row++, 1, \@content, $f_headertab);
1107 for ($i = 0; $i <= $#$times_aref; ++$i) {
1108 if ($$values_aref[$i] > 0)
1110 my $percent = ($values_aref->[$i] * 100)/$queue_total;
1111 $cumulative_percent += $percent;
1113 my @content=($printed_one? " " : "Under",
1114 format_time($times_aref->[$i]),
1115 $values_aref->[$i], $percent, $cumulative_percent);
1118 printf $htm_fh ($htm_format, @content);
1119 if (!defined($values_aref->[$i])) {
1120 print $htm_fh "Not defined";
1124 printf $txt_fh ($txt_format, @content);
1125 if (!defined($times_aref->[$i])) {
1126 print $txt_fh "Not defined";
1132 &set_worksheet_line($ws_global, $row, 0, [@content[0,1,2]], $f_default);
1133 &set_worksheet_line($ws_global, $row++, 3, [$content[3]/100,$content[4]/100], $f_percent);
1135 if (!defined($times_aref->[$i])) {
1137 $ws_global->write($row++, $col, "Not defined" );
1141 push(@chartdatanames,
1142 ($printed_one? "" : "Under") . format_time($times_aref->[$i]));
1143 push(@chartdatavals, $$values_aref[$i]);
1148 if ($overflow && $overflow > 0) {
1149 my $percent = ($overflow * 100)/$queue_total;
1150 $cumulative_percent += $percent;
1152 my @content = ("Over ", format_time($times_aref->[-1]),
1153 $overflow, $percent, $cumulative_percent);
1155 printf $txt_fh ($txt_format, @content) if $txt_fh;
1156 printf $htm_fh ($htm_format, @content) if $htm_fh;
1159 &set_worksheet_line($ws_global, $row, 0, [@content[0,1,2]], $f_default);
1160 &set_worksheet_line($ws_global, $row++, 3, [$content[3]/100,$content[4]/100], $f_percent);
1165 push(@chartdatanames, "Over " . format_time($times_aref->[-1]));
1166 push(@chartdatavals, $overflow);
1168 #printf("Unknown %6d\n", $queue_unknown) if $queue_unknown > 0;
1170 print $htm_fh "</table></td><td>";
1172 if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) {
1177 my $graph = GD::Graph::pie->new(200, 200);
1178 my $pngname = "$title-$message_type.png";
1179 $pngname =~ s/[^\w\-\.]/_/;
1181 my $graph_title = "$title ($message_type)";
1182 $graph->set(title => $graph_title) if (length($graph_title) < 21);
1184 my $gd = $graph->plot(\@data) or warn($graph->error);
1186 open(IMG, ">$chartdir/$pngname") or die "Could not write $chartdir/$pngname: $!\n";
1190 print $htm_fh "<img src=\"$chartrel/$pngname\">";
1193 print $htm_fh "</td></tr></table>\n";
1200 print $txt_fh "\n" if $txt_fh;
1201 print $htm_fh "\n" if $htm_fh;
1206 #######################################################################
1207 # print_histogram();
1209 # print_histogram('Deliveries|Messages received|$pattern', $unit, @interval_count);
1211 # Print a histogram of the messages delivered/received per time slot
1212 # (hour by default).
1213 #######################################################################
1214 sub print_histogram {
1215 my($text, $unit, @interval_count) = @_;
1216 my(@chartdatanames);
1220 # save first row of print_histogram for xls output
1228 for ($i = 0; $i < $hist_number; $i++)
1229 { $maxd = $interval_count[$i] if $interval_count[$i] > $maxd; }
1231 my $scale = int(($maxd + 25)/50);
1232 $scale = 1 if $scale == 0;
1235 if ($unit !~ s/y$/ies/) {
1240 # make and output title
1241 my $title = sprintf("$text per %s",
1242 ($hist_interval == 60)? "hour" :
1243 ($hist_interval == 1)? "minute" : "$hist_interval minutes");
1245 my $txt_htm_title = $title . " (each dot is $scale $unit)";
1247 printf $txt_fh ("%s\n%s\n\n", $txt_htm_title, "-" x length($txt_htm_title)) if $txt_fh;
1250 print $htm_fh "<hr><a name=\"$text\"></a><h2>$txt_htm_title</h2>\n";
1251 print $htm_fh "<table border=0 width=\"100%\">\n";
1252 print $htm_fh "<tr><td><pre>\n";
1256 $title =~ s/Messages/Msg/ ;
1258 $ws_global->write($row++, $col_hist+1, $title, $f_headertab);
1264 for ($i = 0; $i < $hist_number; $i++) {
1265 my $c = $interval_count[$i];
1267 # If the interval is an hour (the maximum) print the starting and
1268 # ending hours as a label. Otherwise print the starting hour and
1269 # minutes, which take up the same space.
1272 if ($hist_opt == 1) {
1273 $temp = sprintf("%02d-%02d", $hour, $hour + 1);
1275 print $txt_fh $temp if $txt_fh;
1276 print $htm_fh $temp if $htm_fh;
1281 $ws_global->write($row, 0, [$temp], $f_default);
1285 push(@chartdatanames, $temp);
1290 { $temp = sprintf("%02d:%02d", $hour, $minutes) }
1292 { $temp = sprintf(" :%02d", $minutes) }
1294 print $txt_fh $temp if $txt_fh;
1295 print $htm_fh $temp if $htm_fh;
1296 if (($xls_fh) and ($run_hist==0)) {
1298 $temp = sprintf("%02d:%02d", $hour, $minutes);
1299 $ws_global->write($row, 0, [$temp], $f_default);
1302 push(@chartdatanames, $temp);
1303 $minutes += $hist_interval;
1304 if ($minutes >= 60) {
1309 push(@chartdatavals, $c);
1311 printf $txt_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $txt_fh;
1312 printf $htm_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $htm_fh;
1313 $ws_global->write($row++, $col_hist+1, [$c], $f_default) if $xls_fh;
1317 printf $txt_fh "\n" if $txt_fh;
1318 printf $htm_fh "\n" if $htm_fh;
1322 print $htm_fh "</pre>\n";
1323 print $htm_fh "</td><td>\n";
1324 if ($HAVE_GD_Graph_linespoints && $charts && ($#chartdatavals > 0)) {
1325 # calculate the graph
1330 my $graph = GD::Graph::linespoints->new(300, 300);
1333 y_label => 'Amount',
1335 x_labels_vertical => 1
1337 my $pngname = "histogram_$text.png";
1338 $pngname =~ s/[^\w\._]/_/g;
1340 my $gd = $graph->plot(\@data) or warn($graph->error);
1342 open(IMG, ">$chartdir/$pngname") or die "Could not write $chartdir/$pngname: $!\n";
1346 print $htm_fh "<img src=\"$chartrel/$pngname\">";
1349 print $htm_fh "</td></tr></table>\n";
1352 $col_hist++; # where to continue next times
1354 $row+=2; # leave some space after history block
1355 $run_hist=1; # we have done this once or more
1360 #######################################################################
1361 # print_league_table();
1363 # print_league_table($league_table_type,\%message_count,\%address_count,\%message_data,\%message_data_gigs, $spreadsheet, $row_sref);
1365 # Given hashes of message count, address count, and message data,
1366 # which are keyed by the table type (eg by the sending host), print a
1367 # league table showing the top $topcount (defaults to 50).
1368 #######################################################################
1369 sub print_league_table {
1370 my($text,$m_count,$a_count,$m_data,$m_data_gigs,$spreadsheet, $row_sref) = @_;
1371 my($name) = ($topcount == 1)? "$text" : "$topcount ${text}s";
1372 my($title) = "Top $name by message count";
1373 my(@chartdatanames) = ();
1374 my(@chartdatavals) = ();
1375 my $chartotherval = 0;
1376 $text = ucfirst($text);
1378 # Align non-local addresses to the right (so all the .com's line up).
1379 # Local addresses are aligned on the left as they are userids.
1380 my $align = ($text !~ /local/i) ? 'right' : 'left';
1383 ################################################
1384 # Generate the printf formats and table headers.
1385 ################################################
1386 my(@headers) = ('Messages');
1387 #push(@headers,'Addresses') if defined $a_count;
1388 push(@headers,'Addresses') if defined $a_count && %$a_count;
1389 push(@headers,'Bytes','Average') if defined $m_data;
1391 my $txt_format = "%10s " x @headers . " %s\n";
1392 my $txt_col_headers = sprintf $txt_format, @headers, $text;
1393 my $htm_format = "<tr>" . '<td align="right">%s</td>'x@headers . "<td align=\"$align\" nowrap>%s</td></tr>\n";
1394 my $htm_col_headers = sprintf $htm_format, @headers, $text;
1395 $htm_col_headers =~ s/(<\/?)td/$1th/g; #Convert <td>'s to <th>'s for the header.
1398 ################################################
1399 # Write the table headers
1400 ################################################
1401 printf $txt_fh ("%s\n%s\n%s", $title, "-" x length($title),$txt_col_headers) if $txt_fh;
1404 print $htm_fh <<EoText;
1405 <hr><a name="$text count"></a><h2>$title</h2>
1406 <table border=0 width="100%">
1410 print $htm_fh $htm_col_headers
1414 $spreadsheet->write(${$row_sref}++, 0, $title, $f_header2);
1415 $spreadsheet->write(${$row_sref}++, 0, [@headers, $text], $f_headertab);
1420 foreach my $key (top_n_sort($topcount,$m_count,$m_data_gigs,$m_data)) {
1422 # When displaying the average figures, we calculate the average of
1423 # the rounded data, as the user would calculate it. This reduces
1424 # the accuracy slightly, but we have to do it this way otherwise
1425 # when using -merge to convert results from text to HTML and
1426 # vice-versa discrepencies would occur.
1427 my $messages = $$m_count{$key};
1428 my @content = ($messages);
1429 push(@content, $$a_count{$key}) if defined $a_count;
1430 if (defined $m_data) {
1431 my $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key});
1432 my($data,$gigs) = (0,0);
1433 un_round($rounded_volume,\$data,\$gigs);
1434 my $rounded_average = volume_rounded($data/$messages,$gigs/$messages);
1435 push(@content, $rounded_volume, $rounded_average);
1439 printf $txt_fh ($txt_format, @content, $key) if $txt_fh;
1443 $htmlkey =~ s/>/\>\;/g;
1444 $htmlkey =~ s/</\<\;/g;
1445 printf $htm_fh ($htm_format, @content, $htmlkey);
1447 $spreadsheet->write(${$row_sref}++, 0, [@content, $key], $f_default) if $xls_fh;
1449 if (scalar @chartdatanames < $ntopchart) {
1450 push(@chartdatanames, $key);
1451 push(@chartdatavals, $$m_count{$key});
1454 $chartotherval += $$m_count{$key};
1458 push(@chartdatanames, "Other");
1459 push(@chartdatavals, $chartotherval);
1461 print $txt_fh "\n" if $txt_fh;
1463 print $htm_fh "</table>\n";
1464 print $htm_fh "</td><td>\n";
1465 if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0))
1467 # calculate the graph
1472 my $graph = GD::Graph::pie->new(300, 300);
1475 y_label => 'Amount',
1476 title => 'By count',
1478 my $gd = $graph->plot(\@data) or warn($graph->error);
1482 open(IMG, ">$chartdir/${temp}_count.png") or die "Could not write $chartdir/${temp}_count.png: $!\n";
1486 print $htm_fh "<img src=\"$chartrel/${temp}_count.png\">";
1489 print $htm_fh "</td><td>\n";
1490 print $htm_fh "</td></tr></table>\n\n";
1492 ++${$row_sref} if $xls_fh;
1495 if (defined $m_data) {
1498 $title = "Top $name by volume";
1500 printf $txt_fh ("%s\n%s\n%s", $title, "-" x length($title),$txt_col_headers) if $txt_fh;
1503 print $htm_fh <<EoText;
1504 <hr><a name="$text volume"></a><h2>$title</h2>
1505 <table border=0 width="100%">
1509 print $htm_fh $htm_col_headers;
1512 $spreadsheet->write(${$row_sref}++, 0, $title, $f_header2);
1513 $spreadsheet->write(${$row_sref}++, 0, [@headers, $text], $f_headertab);
1516 @chartdatanames = ();
1517 @chartdatavals = ();
1520 foreach my $key (top_n_sort($topcount,$m_data_gigs,$m_data,$m_count)) {
1521 # The largest volume will be the first (top of the list).
1522 # If it has at least 1 gig, then just use gigabytes to avoid
1523 # risking an integer overflow when generating the pie charts.
1524 if ($$m_data_gigs{$key}) {
1528 my $messages = $$m_count{$key};
1529 my @content = ($messages);
1530 push(@content, $$a_count{$key}) if defined $a_count;
1531 my $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key});
1532 my($data ,$gigs) = (0,0);
1533 un_round($rounded_volume,\$data,\$gigs);
1534 my $rounded_average = volume_rounded($data/$messages,$gigs/$messages);
1535 push(@content, $rounded_volume, $rounded_average );
1538 printf $txt_fh ($txt_format, @content, $key) if $txt_fh;
1541 $htmlkey =~ s/>/\>\;/g;
1542 $htmlkey =~ s/</\<\;/g;
1543 printf $htm_fh ($htm_format, @content, $htmlkey);
1545 $spreadsheet->write(${$row_sref}++, 0, [@content, $key], $f_default) if $xls_fh;
1548 if (scalar @chartdatanames < $ntopchart) {
1550 if ($$m_data_gigs{$key}) {
1551 push(@chartdatanames, $key);
1552 push(@chartdatavals, $$m_data_gigs{$key});
1556 push(@chartdatanames, $key);
1557 push(@chartdatavals, $$m_data{$key});
1561 $chartotherval += ($use_gig) ? $$m_data_gigs{$key} : $$m_data{$key};
1564 push(@chartdatanames, "Other");
1565 push(@chartdatavals, $chartotherval);
1567 print $txt_fh "\n" if $txt_fh;
1569 print $htm_fh "</table>\n";
1570 print $htm_fh "</td><td>\n";
1571 if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) {
1572 # calculate the graph
1577 my $graph = GD::Graph::pie->new(300, 300);
1580 y_label => 'Volume' ,
1581 title => 'By Volume',
1583 my $gd = $graph->plot(\@data) or warn($graph->error);
1587 open(IMG, ">$chartdir/${temp}_volume.png") or die "Could not write $chartdir/${temp}_volume.png: $!\n";
1591 print $htm_fh "<img src=\"$chartrel/${temp}_volume.png\">";
1594 print $htm_fh "</td><td>\n";
1595 print $htm_fh "</td></tr></table>\n\n";
1598 ++${$row_sref} if $xls_fh;
1603 #######################################################################
1606 # @sorted_keys = top_n_sort($n,$href1,$href2,$href3);
1608 # Given a hash which has numerical values, return the sorted $n keys which
1609 # point to the top values. The second and third hashes are used as
1610 # tiebreakers. They all must have the same keys.
1612 # The idea behind this routine is that when you only want to see the
1613 # top n members of a set, rather than sorting the entire set and then
1614 # plucking off the top n, sort through the stack as you go, discarding
1615 # any member which is lower than your current n'th highest member.
1617 # This proves to be an order of magnitude faster for large hashes.
1618 # On 200,000 lines of mainlog it benchmarked 9 times faster.
1619 # On 700,000 lines of mainlog it benchmarked 13.8 times faster.
1621 # We assume the values are > 0.
1622 #######################################################################
1624 my($n,$href1,$href2,$href3) = @_;
1626 # PH's original sort was:
1628 # foreach $key (sort
1630 # $$m_count{$b} <=> $$m_count{$a} ||
1631 # $$m_data_gigs{$b} <=> $$m_data_gigs{$a} ||
1632 # $$m_data{$b} <=> $$m_data{$a} ||
1638 #We use a key of '_' to represent non-existant values, as null keys are valid.
1639 #'_' is not a valid domain, edomain, host, or email.
1640 my(@top_n_keys) = ('_') x $n;
1641 my($minimum_value1,$minimum_value2,$minimum_value3) = (0,0,0);
1643 my $n_minus_1 = $n - 1;
1644 my $n_minus_2 = $n - 2;
1646 # Create a dummy hash incase the user has not provided us with
1647 # tiebreaker hashes.
1649 $href2 = \%dummy_hash unless defined $href2;
1650 $href3 = \%dummy_hash unless defined $href3;
1652 # Pick out the top $n keys.
1653 my($key,$value1,$value2,$value3,$i,$comparison,$insert_position);
1654 while (($key,$value1) = each %$href1) {
1656 #print STDERR "key $key ($value1,",$href2->{$key},",",$href3->{$key},") <=> ($minimum_value1,$minimum_value2,$minimum_value3)\n";
1658 # Check to see that the new value is bigger than the lowest of the
1659 # top n keys that we're keeping. We test the main key first, because
1660 # for the majority of cases we can skip creating dummy hash values
1661 # should the user have not provided real tie-breaking hashes.
1662 next unless $value1 >= $minimum_value1;
1664 # Create a dummy hash entry for the key if required.
1665 # Note that setting the dummy_hash value sets it for both href2 &
1666 # href3. Also note that currently we are guaranteed to have a real
1667 # value for href3 if a real value for href2 exists so don't need to
1668 # test for it as well.
1669 $dummy_hash{$key} = 0 unless exists $href2->{$key};
1671 $comparison = $value1 <=> $minimum_value1 ||
1672 $href2->{$key} <=> $minimum_value2 ||
1673 $href3->{$key} <=> $minimum_value3 ||
1674 $top_n_key cmp $key;
1675 next unless ($comparison == 1);
1677 # As we will be using these values a few times, extract them into scalars.
1678 $value2 = $href2->{$key};
1679 $value3 = $href3->{$key};
1681 # This key is bigger than the bottom n key, so the lowest position we
1682 # will insert it into is $n minus 1 (the bottom of the list).
1683 $insert_position = $n_minus_1;
1685 # Now go through the list, stopping when we find a key that we're
1686 # bigger than, or we come to the penultimate position - we've
1687 # already tested bigger than the last.
1689 # Note: we go top down as the list starts off empty.
1690 # Note: stepping through the list in this way benchmarks nearly
1691 # three times faster than doing a sort() on the reduced list.
1692 # I assume this is because the list is already in order, and
1693 # we get a performance boost from not having to do hash lookups
1695 for ($i = 0; $i < $n_minus_1; $i++) {
1696 $top_n_key = $top_n_keys[$i];
1697 if ( ($top_n_key eq '_') ||
1698 ( ($value1 <=> $href1->{$top_n_key} ||
1699 $value2 <=> $href2->{$top_n_key} ||
1700 $value3 <=> $href3->{$top_n_key} ||
1701 $top_n_key cmp $key) == 1
1704 $insert_position = $i;
1709 # Remove the last element, then insert the new one.
1710 $#top_n_keys = $n_minus_2;
1711 splice(@top_n_keys,$insert_position,0,$key);
1713 # Extract our new minimum values.
1714 $top_n_key = $top_n_keys[$n_minus_1];
1715 if ($top_n_key ne '_') {
1716 $minimum_value1 = $href1->{$top_n_key};
1717 $minimum_value2 = $href2->{$top_n_key};
1718 $minimum_value3 = $href3->{$top_n_key};
1722 # Return the top n list, grepping out non-existant values, just in case
1723 # we didn't have that many values.
1724 return(grep(!/^_$/,@top_n_keys));
1729 #######################################################################
1732 # $header = html_header($title);
1734 # Print our HTML header and start the <body> block.
1735 #######################################################################
1738 my $text = << "EoText";
1739 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
1742 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-15">
1743 <title>$title</title>
1745 <body bgcolor="white">
1753 #######################################################################
1758 # Display usage instructions and exit.
1759 #######################################################################
1763 eximstats Version $VERSION
1766 eximstats [Output] [Options] mainlog1 mainlog2 ...
1767 eximstats -merge -html [Options] report.1.html ... > weekly_rep.html
1770 eximstats -html=eximstats.html mainlog1 mainlog2 ...
1771 eximstats mainlog1 mainlog2 ... > report.txt
1773 Parses exim mainlog or syslog files and generates a statistical analysis
1774 of the messages processed.
1776 Valid output types are:
1777 -txt[=<file>] plain text (default unless no other type is specified)
1780 With no type and file given, defaults to -txt and STDOUT.
1783 -h<number> histogram divisions per hour. The default is 1, and
1784 0 suppresses histograms. Other valid values are:
1785 2, 3, 5, 10, 15, 20, 30 or 60.
1786 -ne don't display error information
1787 -nr don't display relaying information
1788 -nr/pattern/ don't display relaying information that matches
1789 -nt don't display transport information
1790 -nt/pattern/ don't display transport information that matches
1791 -nvr don't do volume rounding. Display in bytes, not KB/MB/GB.
1792 -t<number> display top <number> sources/destinations
1793 default is 50, 0 suppresses top listing
1794 -tnl omit local sources/destinations in top listing
1795 -t_remote_users show top user sources/destinations from non-local domains
1796 -q<list> list of times for queuing information. -q0 suppresses.
1797 -show_rt<list> Show the receipt times for all the messages.
1798 -show_dt<list> Show the delivery times for all the messages.
1799 <list> is an optional list of times in seconds.
1802 -include_original_destination show both the final and original
1803 destinations in the results rather than just the final ones.
1805 -byhost show results by sending host (default unless bydomain or
1806 byemail is specified)
1807 -bydomain show results by sending domain.
1808 -byemail show results by sender's email address
1809 -byedomain show results by sender's email domain
1810 -bylocaldomain show results by local domain
1812 -pattern "Description" /pattern/
1813 Count lines matching specified patterns and show them in
1814 the results. It can be specified multiple times. Eg:
1815 -pattern 'Refused connections' '/refused connection/'
1817 -merge merge previously generated reports into a new report
1819 -charts Create charts (this requires the GD::Graph modules).
1820 Only valid with -html.
1821 -chartdir <dir> Create the charts' png files in the directory <dir>
1822 -chartrel <dir> Specify the relative directory for the "img src=" tags
1823 from where to include the charts in the html file
1824 -chartdir and -chartrel default to '.'
1826 -emptyok It is OK if there is no valid input, don't print an error.
1828 -d Debug mode - dump the eval'ed parser onto STDERR.
1837 #######################################################################
1838 # generate_parser();
1840 # $parser = generate_parser();
1842 # This subroutine generates the parsing routine which will be
1843 # used to parse the mainlog. We take the base operation, and remove bits not in use.
1844 # This improves performance depending on what bits you take out or add.
1846 # I've tested using study(), but this does not improve performance.
1848 # We store our parsing routing in a variable, and process it looking for #IFDEF (Expression)
1849 # or #IFNDEF (Expression) statements and corresponding #ENDIF (Expression) statements. If
1850 # the expression evaluates to true, then it is included/excluded accordingly.
1851 #######################################################################
1852 sub generate_parser {
1854 my($ip,$host,$email,$edomain,$domain,$thissize,$size,$old,$new);
1855 my($tod,$m_hour,$m_min,$id,$flag,$extra,$length);
1856 my($seconds,$queued,$rcpt_time,$local_domain);
1860 # Convert syslog lines to mainlog format.
1862 next unless s/^.*? exim\\b.*?: //;
1865 $length = length($_);
1866 next if ($length < 38);
1868 (\\d{4}\\-\\d\\d-\\d\\d\\s # 1: YYYYMMDD HHMMSS
1874 (\\.\\d+)? # 4: subseconds
1875 (\s[-+]\\d\\d\\d\\d)? # 5: tz-offset
1876 (\s\\[\\d+\\])? # 6: pid
1879 $tod = defined($5) ? $1 . $5 : $1;
1880 ($m_hour,$m_min) = ($2,$3);
1882 # PH - watch for GMT offsets in the timestamp.
1885 next if ($length < 44);
1891 # watch for subsecond precision
1893 $extra += length($4);
1894 next if ($length < 38 + $extra);
1897 # PH - watch for PID added after the timestamp.
1899 $extra += length($6);
1900 next if ($length < 38 + $extra);
1903 # $id = substr($_, 20 + $extra, 16); # old ID was 16 chars
1904 $id = substr($_, 20 + $extra, 23); # new IS is 23 chars
1905 $id =~ s/(\S+).*/$1/;
1906 $extra += length($id) - 16;
1908 $flag = substr($_, 37 + $extra, 2);
1910 if ($flag !~ /^([<>=*-]+|SA)$/ && /rejected|refused|dropped/) {
1915 # Rejects can have no MSGID...
1916 if ($flag eq "Re" && $id !~ /^[-0-9a-zA-Z]+$/) {
1917 $id = "reject:" . ++$rej_id;
1922 # Watch for user specified patterns.
1923 my $user_pattern_index = 0;
1924 foreach (@user_patterns) {
1925 $user_pattern_totals[$user_pattern_index] = 0;
1926 $parser .= " if ($_) {\n";
1927 $parser .= " \$user_pattern_totals[$user_pattern_index]++;\n";
1928 $parser .= " \$user_pattern_interval_count[$user_pattern_index][(\$m_hour*60 + \$m_min)/$hist_interval]++;\n" if ($hist_opt > 0);
1930 $user_pattern_index++;
1934 next unless ($flag =~ /<=|=>|->|==|\\*\\*|Co|SA|Re/);
1936 #Strip away the timestamp, ID and flag to speed up later pattern matches.
1937 #The flags include Co (Completed), Re (Rejected), and SA (SpamAssassin).
1938 $_ = substr($_, 40 + $extra); # PH
1940 # Alias @message to the array of information about the message.
1941 # This minimises the number of calls to hash functions.
1942 $messages{$id} = [] unless exists $messages{$id};
1943 *message = $messages{$id};
1946 # JN - Skip over certain transports as specified via the "-nt/.../" command
1947 # line switch (where ... is a perl style regular expression). This is
1948 # required so that transports that skew stats such as SpamAssassin can be
1950 #IFDEF ($transport_pattern)
1951 if (/\\sT=(\\S+)/) {
1952 next if ($1 =~ /$transport_pattern/o) ;
1954 #ENDIF ($transport_pattern)
1958 # Do some pattern matches to get the host and IP address.
1959 # We expect lines to be of the form "H=[IpAddr]" or "H=Host [IpAddr]" or
1960 # "H=Host (UnverifiedHost) [IpAddr]" or "H=(UnverifiedHost) [IpAddr]".
1961 # We do 2 separate matches to keep the matches simple and fast.
1962 # Host is local unless otherwise specified.
1963 # Watch out for "H=([IpAddr])" in case they send "[IpAddr]" as their HELO!
1964 $ip = (/\\bH=(?:|.*? )(\\[[^]]+\\])/) ? $1
1965 # 2008-03-31 06:25:22 Connection from [213.246.33.217]:39456 refused: too many connections from that IP address // .hs
1966 : (/Connection from (\[\S+\])/) ? $1
1967 # 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
1968 : (/SMTP call from .*?(\[\S+\])/) ? $1
1970 $host = (/\\bH=(\\S+)/) ? $1 : "local";
1972 $domain = "localdomain"; #Domain is localdomain unless otherwise specified.
1974 #IFDEF ($do_sender{Domain})
1975 if ($host =~ /^\\[/ || $host =~ /^[\\d\\.]+$/) {
1976 # Host is just an IP address.
1979 elsif ($host =~ /^(\\(?)[^\\.]+\\.([^\\.]+\\..*)/) {
1980 # Remove the host portion from the DNS name. We ensure that we end up
1981 # with at least xxx.yyy. $host can be "(x.y.z)" or "x.y.z".
1982 $domain = lc("$1.$2");
1983 $domain =~ s/^\\.//; #Remove preceding dot.
1985 #ENDIF ($do_sender{Domain})
1987 #IFDEF ($do_sender{Email})
1988 #IFDEF ($include_original_destination)
1989 # Catch both "a@b.com <c@d.com>" and "e@f.com"
1990 #$email = (/^(\S+) (<(\S*?)>)?/) ? $3 || $1 : "";
1991 $email = (/^(\S+ (<[^@>]+@?[^>]*>)?)/) ? $1 : "";
1993 #ENDIF ($include_original_destination)
1995 #IFNDEF ($include_original_destination)
1996 $email = (/^(\S+)/) ? $1 : "";
1997 #ENDIF ($include_original_destination)
1998 #ENDIF ($do_sender{Email})
2000 #IFDEF ($do_sender{Edomain})
2001 if (/^(<>|blackhole)/) {
2004 #IFDEF ($include_original_destination)
2005 elsif (/^(\S+ (<\S*?\\@(\S+?)>)?)/) {
2008 $edomain =~ s/@(\S+?)>/"@" . lc($1) . ">"/e;
2010 #ENDIF ($include_original_destination)
2011 #IFNDEF ($include_original_destination)
2012 elsif (/^\S*?\\@(\S+)/) {
2015 #ENDIF ($include_original_destination)
2020 #ENDIF ($do_sender{Edomain})
2022 if ($tod lt $begin) {
2025 elsif ($tod gt $end) {
2030 if ($flag eq "<=") {
2031 $thissize = (/\\sS=(\\d+)( |$)/) ? $1 : 0;
2032 $message[$SIZE] = $thissize;
2033 $message[$PROTOCOL] = (/ P=(\S+)/) ? $1 : undef;
2035 #IFDEF ($show_relay)
2036 if ($host ne "local") {
2037 # Save incoming information in case it becomes interesting
2038 # later, when delivery lines are read.
2039 my($from) = /^(\\S+)/;
2040 $message[$FROM_HOST] = "$host$ip";
2041 $message[$FROM_ADDRESS] = $from;
2043 #ENDIF ($show_relay)
2045 #IFDEF ($local_league_table || $include_remote_users)
2049 #IFDEF ($local_league_table && $include_remote_users)
2050 { #Store both local and remote users.
2051 #ENDIF ($local_league_table && $include_remote_users)
2053 #IFDEF ($local_league_table && ! $include_remote_users)
2054 if ($host eq "local") { #Store local users only.
2055 #ENDIF ($local_league_table && ! $include_remote_users)
2057 #IFDEF ($include_remote_users && ! $local_league_table)
2058 if ($host ne "local") { #Store remote users only.
2059 #ENDIF ($include_remote_users && ! $local_league_table)
2061 ++$received_count_user{$user};
2062 add_volume(\\$received_data_user{$user},\\$received_data_gigs_user{$user},$thissize);
2065 #ENDIF ($local_league_table || $include_remote_users)
2067 #IFDEF ($do_sender{Host})
2068 ++$received_count{Host}{$host};
2069 add_volume(\\$received_data{Host}{$host},\\$received_data_gigs{Host}{$host},$thissize);
2070 #ENDIF ($do_sender{Host})
2072 #IFDEF ($do_sender{Domain})
2074 ++$received_count{Domain}{$domain};
2075 add_volume(\\$received_data{Domain}{$domain},\\$received_data_gigs{Domain}{$domain},$thissize);
2077 #ENDIF ($do_sender{Domain})
2079 #IFDEF ($do_sender{Email})
2080 ++$received_count{Email}{$email};
2081 add_volume(\\$received_data{Email}{$email},\\$received_data_gigs{Email}{$email},$thissize);
2082 #ENDIF ($do_sender{Email})
2084 #IFDEF ($do_sender{Edomain})
2085 ++$received_count{Edomain}{$edomain};
2086 add_volume(\\$received_data{Edomain}{$edomain},\\$received_data_gigs{Edomain}{$edomain},$thissize);
2087 #ENDIF ($do_sender{Edomain})
2089 ++$total_received_count;
2090 add_volume(\\$total_received_data,\\$total_received_data_gigs,$thissize);
2092 #IFDEF ($#queue_times >= 0 || $#rcpt_times >= 0)
2093 $message[$ARRIVAL_TIME] = $tod;
2094 #ENDIF ($#queue_times >= 0 || $#rcpt_times >= 0)
2096 #IFDEF ($hist_opt > 0)
2097 $received_interval_count[($m_hour*60 + $m_min)/$hist_interval]++;
2098 #ENDIF ($hist_opt > 0)
2101 elsif ($flag eq "=>") {
2102 $size = $message[$SIZE] || 0;
2103 if ($host ne "local") {
2104 $message[$REMOTE_DELIVERED] = 1;
2107 #IFDEF ($show_relay)
2108 # Determine relaying address if either only one address listed,
2109 # or two the same. If they are different, it implies a forwarding
2110 # or aliasing, which is not relaying. Note that for multi-aliased
2111 # addresses, there may be a further address between the first
2114 if (defined $message[$FROM_HOST]) {
2115 if (/^(\\S+)(?:\\s+\\([^)]\\))?\\s+<([^>]+)>/) {
2116 ($old,$new) = ($1,$2);
2122 if ("\\L$new" eq "\\L$old") {
2123 ($old) = /^(\\S+)/ if $old eq "";
2124 my $key = "H=\\L$message[$FROM_HOST]\\E A=\\L$message[$FROM_ADDRESS]\\E => " .
2125 "H=\\L$host\\E$ip A=\\L$old\\E";
2126 if (!defined $relay_pattern || $key !~ /$relay_pattern/o) {
2127 $relayed{$key} = 0 if !defined $relayed{$key};
2135 #ENDIF ($show_relay)
2139 #IFDEF ($local_league_table || $include_remote_users)
2140 #IFDEF ($local_league_table && $include_remote_users)
2141 { #Store both local and remote users.
2142 #ENDIF ($local_league_table && $include_remote_users)
2144 #IFDEF ($local_league_table && ! $include_remote_users)
2145 if ($host eq "local") { #Store local users only.
2146 #ENDIF ($local_league_table && ! $include_remote_users)
2148 #IFDEF ($include_remote_users && ! $local_league_table)
2149 if ($host ne "local") { #Store remote users only.
2150 #ENDIF ($include_remote_users && ! $local_league_table)
2152 if (my($user) = split((/\\s</)? " <" : " ", $_)) {
2153 #IFDEF ($include_original_destination)
2155 #ENDIF ($include_original_destination)
2156 #IFNDEF ($include_original_destination)
2157 if ($user =~ /^[\\/|]/) {
2158 #ENDIF ($include_original_destination)
2159 #my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/;
2160 my($parent) = $_ =~ / (<.+?>) /; #DT 1.54
2161 if (defined $parent) {
2162 $user = "$user $parent";
2163 #IFDEF ($do_local_domain)
2164 if ($parent =~ /\\@(.+)>/) {
2165 $local_domain = lc($1);
2166 ++$delivered_messages_local_domain{$local_domain};
2167 ++$delivered_addresses_local_domain{$local_domain};
2168 add_volume(\\$delivered_data_local_domain{$local_domain},\\$delivered_data_gigs_local_domain{$local_domain},$size);
2170 #ENDIF ($do_local_domain)
2173 ++$delivered_messages_user{$user};
2174 ++$delivered_addresses_user{$user};
2175 add_volume(\\$delivered_data_user{$user},\\$delivered_data_gigs_user{$user},$size);
2178 #ENDIF ($local_league_table || $include_remote_users)
2180 #IFDEF ($do_sender{Host})
2181 $delivered_messages{Host}{$host}++;
2182 $delivered_addresses{Host}{$host}++;
2183 add_volume(\\$delivered_data{Host}{$host},\\$delivered_data_gigs{Host}{$host},$size);
2184 #ENDIF ($do_sender{Host})
2185 #IFDEF ($do_sender{Domain})
2187 ++$delivered_messages{Domain}{$domain};
2188 ++$delivered_addresses{Domain}{$domain};
2189 add_volume(\\$delivered_data{Domain}{$domain},\\$delivered_data_gigs{Domain}{$domain},$size);
2191 #ENDIF ($do_sender{Domain})
2192 #IFDEF ($do_sender{Email})
2193 ++$delivered_messages{Email}{$email};
2194 ++$delivered_addresses{Email}{$email};
2195 add_volume(\\$delivered_data{Email}{$email},\\$delivered_data_gigs{Email}{$email},$size);
2196 #ENDIF ($do_sender{Email})
2197 #IFDEF ($do_sender{Edomain})
2198 ++$delivered_messages{Edomain}{$edomain};
2199 ++$delivered_addresses{Edomain}{$edomain};
2200 add_volume(\\$delivered_data{Edomain}{$edomain},\\$delivered_data_gigs{Edomain}{$edomain},$size);
2201 #ENDIF ($do_sender{Edomain})
2203 ++$total_delivered_messages;
2204 ++$total_delivered_addresses;
2205 add_volume(\\$total_delivered_data,\\$total_delivered_data_gigs,$size);
2207 #IFDEF ($show_transport)
2208 my $transport = (/\\sT=(\\S+)/) ? $1 : ":blackhole:";
2209 ++$transported_count{$transport};
2210 add_volume(\\$transported_data{$transport},\\$transported_data_gigs{$transport},$size);
2211 #ENDIF ($show_transport)
2213 #IFDEF ($hist_opt > 0)
2214 $delivered_interval_count[($m_hour*60 + $m_min)/$hist_interval]++;
2215 #ENDIF ($hist_opt > 0)
2217 #IFDEF ($#delivery_times > 0)
2219 $seconds = wdhms_seconds($1);
2220 for ($i = 0; $i <= $#delivery_times; $i++) {
2221 if ($seconds < $delivery_times[$i]) {
2223 ++$dt_remote_bin[$i] if $message[$REMOTE_DELIVERED];
2227 if ($i > $#delivery_times) {
2229 ++$dt_remote_overflow if $message[$REMOTE_DELIVERED];
2232 #ENDIF ($#delivery_times > 0)
2236 elsif ($flag eq "->") {
2238 #IFDEF ($local_league_table || $include_remote_users)
2239 #IFDEF ($local_league_table && $include_remote_users)
2240 { #Store both local and remote users.
2241 #ENDIF ($local_league_table && $include_remote_users)
2243 #IFDEF ($local_league_table && ! $include_remote_users)
2244 if ($host eq "local") { #Store local users only.
2245 #ENDIF ($local_league_table && ! $include_remote_users)
2247 #IFDEF ($include_remote_users && ! $local_league_table)
2248 if ($host ne "local") { #Store remote users only.
2249 #ENDIF ($include_remote_users && ! $local_league_table)
2251 if (my($user) = split((/\\s</)? " <" : " ", $_)) {
2252 #IFDEF ($include_original_destination)
2254 #ENDIF ($include_original_destination)
2255 #IFNDEF ($include_original_destination)
2256 if ($user =~ /^[\\/|]/) {
2257 #ENDIF ($include_original_destination)
2258 #my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/;
2259 my($parent) = $_ =~ / (<.+?>) /; #DT 1.54
2260 $user = "$user $parent" if defined $parent;
2262 ++$delivered_addresses_user{$user};
2265 #ENDIF ($local_league_table || $include_remote_users)
2267 #IFDEF ($do_sender{Host})
2268 $delivered_addresses{Host}{$host}++;
2269 #ENDIF ($do_sender{Host})
2270 #IFDEF ($do_sender{Domain})
2272 ++$delivered_addresses{Domain}{$domain};
2274 #ENDIF ($do_sender{Domain})
2275 #IFDEF ($do_sender{Email})
2276 ++$delivered_addresses{Email}{$email};
2277 #ENDIF ($do_sender{Email})
2278 #IFDEF ($do_sender{Edomain})
2279 ++$delivered_addresses{Edomain}{$edomain};
2280 #ENDIF ($do_sender{Edomain})
2282 ++$total_delivered_addresses;
2285 elsif ($flag eq "==" && defined($message[$SIZE]) && !defined($message[$DELAYED])) {
2287 $message[$DELAYED] = 1;
2290 elsif ($flag eq "**") {
2291 if (defined ($message[$SIZE])) {
2292 unless (defined $message[$HAD_ERROR]) {
2294 $message[$HAD_ERROR] = 1;
2298 #IFDEF ($show_errors)
2299 ++$errors_count{$_};
2300 #ENDIF ($show_errors)
2304 elsif ($flag eq "Co") {
2306 #IFDEF ($#queue_times >= 0)
2307 $queued = queue_time($tod, $message[$ARRIVAL_TIME], $id);
2309 for ($i = 0; $i <= $#queue_times; $i++) {
2310 if ($queued < $queue_times[$i]) {
2312 ++$qt_remote_bin[$i] if $message[$REMOTE_DELIVERED];
2316 if ($i > $#queue_times) {
2318 ++$qt_remote_overflow if $message[$REMOTE_DELIVERED];
2320 #ENDIF ($#queue_times >= 0)
2322 #IFDEF ($#rcpt_times >= 0)
2324 $seconds = wdhms_seconds($1);
2325 #Calculate $queued if not previously calculated above.
2326 #IFNDEF ($#queue_times >= 0)
2327 $queued = queue_time($tod, $message[$ARRIVAL_TIME], $id);
2328 #ENDIF ($#queue_times >= 0)
2329 $rcpt_time = $seconds - $queued;
2332 if (defined $message[$PROTOCOL]) {
2333 $protocol = $message[$PROTOCOL];
2335 # Create the bin if its not already defined.
2336 unless (exists $rcpt_times_bin{$protocol}) {
2337 initialise_rcpt_times($protocol);
2342 for ($i = 0; $i <= $#rcpt_times; ++$i) {
2343 if ($rcpt_time < $rcpt_times[$i]) {
2344 ++$rcpt_times_bin{all}[$i];
2345 ++$rcpt_times_bin{$protocol}[$i] if defined $protocol;
2350 if ($i > $#rcpt_times) {
2351 ++$rcpt_times_overflow{all};
2352 ++$rcpt_times_overflow{$protocol} if defined $protocol;
2355 #ENDIF ($#rcpt_times >= 0)
2357 delete($messages{$id});
2359 elsif ($flag eq "SA") {
2360 $ip = (/From.*?(\\[[^]]+\\])/ || /\\((local)\\)/) ? $1 : "";
2361 #SpamAssassin message
2362 if (/Action: ((permanently|temporarily) rejected message|flagged as Spam but accepted): score=(\d+\.\d)/) {
2363 #add_volume(\\$spam_score,\\$spam_score_gigs,$3);
2364 ++$spam_count_by_ip{$ip};
2365 } elsif (/Action: scanned but message isn\'t spam: score=(-?\d+\.\d)/) {
2366 #add_volume(\\$ham_score,\\$ham_score_gigs,$1);
2367 ++$ham_count_by_ip{$ip};
2368 } elsif (/(Not running SA because SAEximRunCond expanded to false|check skipped due to message size)/) {
2369 ++$ham_count_by_ip{$ip};
2373 # Look for Reject messages or blackholed messages (deliveries
2374 # without a transport)
2375 if ($flag eq "Re" || ($flag eq "=>" && ! /\\sT=\\S+/)) {
2376 # Correct the IP address for rejects:
2377 # rejected EHLO from my.test.net [10.0.0.5]: syntactically invalid argument(s):
2378 # rejected EHLO from [10.0.0.6]: syntactically invalid argument(s):
2379 $ip = $1 if ($ip eq "local" && /^rejected [HE][HE]LO from .*?(\[.+?\]):/);
2380 if (/SpamAssassin/) {
2381 ++$rejected_count_by_reason{"Rejected by SpamAssassin"};
2382 ++$rejected_count_by_ip{$ip};
2385 /(temporarily rejected [A-Z]*) .*?(: .*?)(:|\s*$)/
2387 ++$temporarily_rejected_count_by_reason{"\u$1$2"};
2388 ++$temporarily_rejected_count_by_ip{$ip};
2391 /(temporarily refused connection)/
2393 ++$temporarily_rejected_count_by_reason{"\u$1"};
2394 ++$temporarily_rejected_count_by_ip{$ip};
2397 /(listed at [^ ]+)/ ||
2398 /(Forged IP detected in HELO)/ ||
2399 /(Invalid domain or IP given in HELO\/EHLO)/ ||
2400 /(unqualified recipient rejected)/ ||
2401 /(closed connection (after|in response) .*?)\s*$/ ||
2402 /(sender rejected)/ ||
2403 # 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.
2404 # 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.
2405 / rejected after DATA: (.*)/ ||
2406 / (rejected DATA: .*)/ ||
2407 /.DATA ACL discarded recipients.: (.*)/ ||
2408 /rejected after DATA: (unqualified address not permitted)/ ||
2409 /(VRFY rejected)/ ||
2410 # /(sender verify (defer|fail))/i ||
2411 /(too many recipients)/ ||
2412 /(refused relay.*?) to/ ||
2413 /(rejected by non-SMTP ACL: .*)/ ||
2414 /(rejected by local_scan.*)/ ||
2415 # SMTP call from %s dropped: too many syntax or protocol errors (last command was "%s"
2416 # SMTP call from %s dropped: too many nonmail commands
2417 /(dropped: too many ((nonmail|unrecognized) commands|syntax or protocol errors))/ ||
2419 # local_scan() function crashed with signal %d - message temporarily rejected
2420 # local_scan() function timed out - message temporarily rejected
2421 /(local_scan.. function .* - message temporarily rejected)/ ||
2422 # SMTP protocol synchronization error (input sent without waiting for greeting): rejected connection from %s
2423 /(SMTP protocol .*?(error|violation))/ ||
2426 ++$rejected_count_by_reason{"\u$1"};
2427 ++$rejected_count_by_ip{$ip};
2429 elsif (/rejected [HE][HE]LO from [^:]*: syntactically invalid argument/) {
2430 ++$rejected_count_by_reason{"Rejected HELO/EHLO: syntactically invalid argument"};
2431 ++$rejected_count_by_ip{$ip};
2433 elsif (/response to "RCPT TO.*? was: (.*)/) {
2434 ++$rejected_count_by_reason{"Response to RCPT TO was: $1"};
2435 ++$rejected_count_by_ip{$ip};
2438 /(lookup of host )\S+ (failed)/ ||
2440 # rejected from <%s>%s%s%s%s: message too big:
2441 /(rejected [A-Z]*) .*?(: .*?)(:|\s*$)/ ||
2442 # refused connection from %s (host_reject_connection)
2443 # refused connection from %s (tcp wrappers)
2444 /(refused connection )from.*? (\(.*)/ ||
2446 # error from remote mailer after RCPT TO:<a@b.c>: host a.b.c [10.0.0.1]: 450 <a@b.c>: Recipient address rejected: Greylisted for 60 seconds
2447 # 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.
2448 /(error from remote mailer after .*?:).*(: .*?)(:|\s*$)/ ||
2450 # a.b.c F=<a@b.c> rejected after DATA: "@" or "." expected after "Undisclosed-Recipient": failing address in "To" header is: <Undisclosed-Recipient:;>
2451 /rejected after DATA: ("." or "." expected).*?(: failing address in .*? header)/ ||
2453 # connection from %s refused load average = %.2f
2454 /(Connection )from.*? (refused: load average)/ ||
2455 # connection from %s refused (IP options)
2456 # Connection from %s refused: too many connections
2457 # connection from %s refused
2458 /([Cc]onnection )from.*? (refused.*)/ ||
2459 # [10.0.0.1]: connection refused
2460 /: (Connection refused)()/
2462 ++$rejected_count_by_reason{"\u$1$2"};
2463 ++$rejected_count_by_ip{$ip};
2466 # 2008-03-31 06:25:22 H=mail.densitron.com [216.70.140.224]:45386 temporarily rejected connection in "connect" ACL: too fast reconnects // .hs
2467 # 2008-03-31 06:25:22 H=mail.densitron.com [216.70.140.224]:45386 temporarily rejected connection in "connect" ACL // .hs
2468 /(temporarily rejected connection in .*?ACL:?.*)/
2470 ++$temporarily_rejected_count_by_ip{$ip};
2471 ++$temporarily_rejected_count_by_reason{"\u$1"};
2474 ++$rejected_count_by_reason{Unknown};
2475 ++$rejected_count_by_ip{$ip};
2476 print STDERR "Unknown rejection: $_" if $debug;
2481 # We now do a 'C preprocessor style operation on our parser
2482 # to remove bits not in use.
2483 my(%defines_in_operation,$removing_lines,$processed_parser);
2484 foreach (split (/\n/,$parser)) {
2485 if ((/^\s*#\s*IFDEF\s*\((.*?)\)/i && ! eval $1) ||
2486 (/^\s*#\s*IFNDEF\s*\((.*?)\)/i && eval $1) ) {
2487 $defines_in_operation{$1} = 1;
2488 $removing_lines = 1;
2491 # Convert constants.
2492 while (/(\$[A-Z][A-Z_]*)\b/) {
2493 my $constant = eval $1;
2494 s/(\$[A-Z][A-Z_]*)\b/$constant/;
2497 $processed_parser .= $_."\n" unless $removing_lines;
2499 if (/^\s*#\s*ENDIF\s*\((.*?)\)/i) {
2500 delete $defines_in_operation{$1};
2501 unless (keys %defines_in_operation) {
2502 $removing_lines = 0;
2506 print STDERR "# START OF PARSER:$processed_parser\n# END OF PARSER\n\n" if $debug;
2508 return $processed_parser;
2513 #######################################################################
2516 # parse($parser,\*FILEHANDLE);
2518 # This subroutine accepts a parser and a filehandle from main and parses each
2519 # line. We store the results into global variables.
2520 #######################################################################
2522 my($parser,$fh) = @_;
2524 if ($merge_reports) {
2525 parse_old_eximstat_reports($fh);
2536 #######################################################################
2541 # Print our headers and contents.
2542 #######################################################################
2546 my $title = "Exim statistics from $begin to $end";
2548 print $txt_fh "\n$title\n" if $txt_fh;
2550 print $htm_fh html_header($title);
2551 print $htm_fh "<ul>\n";
2552 print $htm_fh "<li><a href=\"#Grandtotal\">Grand total summary</a>\n";
2553 print $htm_fh "<li><a href=\"#Patterns\">User Specified Patterns</a>\n" if @user_patterns;
2554 print $htm_fh "<li><a href=\"#Transport\">Deliveries by Transport</a>\n" if $show_transport;
2556 print $htm_fh "<li><a href=\"#Messages received\">Messages received per hour</a>\n";
2557 print $htm_fh "<li><a href=\"#Deliveries\">Deliveries per hour</a>\n";
2560 if ($#queue_times >= 0) {
2561 print $htm_fh "<li><a href=\"#Time spent on the queue all messages\">Time spent on the queue: all messages</a>\n";
2562 print $htm_fh "<li><a href=\"#Time spent on the queue messages with at least one remote delivery\">Time spent on the queue: messages with at least one remote delivery</a>\n";
2565 if ($#delivery_times >= 0) {
2566 print $htm_fh "<li><a href=\"#Delivery times all messages\">Delivery times: all messages</a>\n";
2567 print $htm_fh "<li><a href=\"#Delivery times messages with at least one remote delivery\">Delivery times: messages with at least one remote delivery</a>\n";
2570 if ($#rcpt_times >= 0) {
2571 print $htm_fh "<li><a href=\"#Receipt times all messages\">Receipt times</a>\n";
2574 print $htm_fh "<li><a href=\"#Relayed messages\">Relayed messages</a>\n" if $show_relay;
2576 print $htm_fh "<li><a href=\"#Mail rejection reason count\">Top $topcount mail rejection reasons by message count</a>\n" if %rejected_count_by_reason;
2577 foreach ('Host','Domain','Email','Edomain') {
2578 next unless $do_sender{$_};
2579 print $htm_fh "<li><a href=\"#Sending \l$_ count\">Top $topcount sending \l${_}s by message count</a>\n";
2580 print $htm_fh "<li><a href=\"#Sending \l$_ volume\">Top $topcount sending \l${_}s by volume</a>\n";
2582 if (($local_league_table || $include_remote_users) && %received_count_user) {
2583 print $htm_fh "<li><a href=\"#Local sender count\">Top $topcount local senders by message count</a>\n";
2584 print $htm_fh "<li><a href=\"#Local sender volume\">Top $topcount local senders by volume</a>\n";
2586 foreach ('Host','Domain','Email','Edomain') {
2587 next unless $do_sender{$_};
2588 print $htm_fh "<li><a href=\"#$_ destination count\">Top $topcount \l$_ destinations by message count</a>\n";
2589 print $htm_fh "<li><a href=\"#$_ destination volume\">Top $topcount \l$_ destinations by volume</a>\n";
2591 if (($local_league_table || $include_remote_users) && %delivered_messages_user) {
2592 print $htm_fh "<li><a href=\"#Local destination count\">Top $topcount local destinations by message count</a>\n";
2593 print $htm_fh "<li><a href=\"#Local destination volume\">Top $topcount local destinations by volume</a>\n";
2595 if (($local_league_table || $include_remote_users) && %delivered_messages_local_domain) {
2596 print $htm_fh "<li><a href=\"#Local domain destination count\">Top $topcount local domain destinations by message count</a>\n";
2597 print $htm_fh "<li><a href=\"#Local domain destination volume\">Top $topcount local domain destinations by volume</a>\n";
2600 print $htm_fh "<li><a href=\"#Rejected ip count\">Top $topcount rejected ips by message count</a>\n" if %rejected_count_by_ip;
2601 print $htm_fh "<li><a href=\"#Temporarily rejected ip count\">Top $topcount temporarily rejected ips by message count</a>\n" if %temporarily_rejected_count_by_ip;
2602 print $htm_fh "<li><a href=\"#Non-rejected spamming ip count\">Top $topcount non-rejected spamming ips by message count</a>\n" if %spam_count_by_ip;
2605 print $htm_fh "<li><a href=\"#errors\">List of errors</a>\n" if %errors_count;
2606 print $htm_fh "</ul>\n<hr>\n";
2610 $ws_global->write($row++, $col+0, "Exim Statistics", $f_header1);
2611 &set_worksheet_line($ws_global, $row, $col, ["from:", $begin, "to:", $end], $f_default);
2617 #######################################################################
2618 # print_grandtotals();
2620 # print_grandtotals();
2622 # Print the grand totals.
2623 #######################################################################
2624 sub print_grandtotals {
2626 # Get the sender by headings and results. This is complicated as we can have
2627 # different numbers of columns.
2628 my($sender_txt_header,$sender_txt_format,$sender_html_format);
2629 my(@received_totals,@delivered_totals);
2630 my($row_tablehead, $row_max);
2631 my(@col_headers) = ('TOTAL', 'Volume', 'Messages', 'Addresses');
2633 foreach ('Host','Domain','Email','Edomain') {
2634 next unless $do_sender{$_};
2635 if ($merge_reports) {
2636 push(@received_totals, get_report_total($report_totals{Received},"${_}s"));
2637 push(@delivered_totals,get_report_total($report_totals{Delivered},"${_}s"));
2640 push(@received_totals,scalar(keys %{$received_data{$_}}));
2641 push(@delivered_totals,scalar(keys %{$delivered_data{$_}}));
2643 $sender_txt_header .= " " x ($COLUMN_WIDTHS - length($_)) . $_ . 's';
2644 $sender_html_format .= "<td align=\"right\">%s</td>";
2645 $sender_txt_format .= " " x ($COLUMN_WIDTHS - 5) . "%6s";
2646 push(@col_headers,"${_}s");
2649 my $txt_format1 = " %-16s %9s %6d %6s $sender_txt_format";
2650 my $txt_format2 = " %6d %4.1f%% %6d %4.1f%%",
2651 my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%s</td><td align=\"right\">%s</td><td align=\"right\">%s</td>$sender_html_format";
2652 my $htm_format2 = "<td align=\"right\">%d</td><td align=\"right\">%4.1f%%</td><td align=\"right\">%d</td><td align=\"right\">%4.1f%%</td>";
2655 my $sender_spaces = " " x length($sender_txt_header);
2657 print $txt_fh "Grand total summary\n";
2658 print $txt_fh "-------------------\n";
2659 print $txt_fh " $sender_spaces At least one address\n";
2660 print $txt_fh " TOTAL Volume Messages Addresses $sender_txt_header Delayed Failed\n";
2663 print $htm_fh "<a name=\"Grandtotal\"></a>\n";
2664 print $htm_fh "<h2>Grand total summary</h2>\n";
2665 print $htm_fh "<table border=1>\n";
2666 print $htm_fh "<tr><th>" . join('</th><th>',@col_headers) . "</th><th colspan=2>At least one addr<br>Delayed</th><th colspan=2>At least one addr<br>Failed</th>\n";
2669 $ws_global->write($row++, 0, "Grand total summary", $f_header2);
2670 $ws_global->write($row, 0, \@col_headers, $f_header2);
2671 $ws_global->merge_range($row, scalar(@col_headers), $row, scalar(@col_headers)+1, "At least one addr Delayed", $f_header2_m);
2672 $ws_global->merge_range($row, scalar(@col_headers)+2, $row, scalar(@col_headers)+3, "At least one addr Failed", $f_header2_m);
2673 #$ws_global->write(++$row, scalar(@col_headers), ['Total','Percent','Total','Percent'], $f_header2);
2677 my($volume,$failed_count);
2678 if ($merge_reports) {
2679 $volume = volume_rounded($report_totals{Received}{Volume}, $report_totals{Received}{'Volume-gigs'});
2680 $total_received_count = get_report_total($report_totals{Received},'Messages');
2681 $failed_count = get_report_total($report_totals{Received},'Failed');
2682 $delayed_count = get_report_total($report_totals{Received},'Delayed');
2685 $volume = volume_rounded($total_received_data, $total_received_data_gigs);
2686 $failed_count = $message_errors;
2693 $volume,$total_received_count,'',
2696 ($total_received_count) ? ($delayed_count*100/$total_received_count) : 0,
2698 ($total_received_count) ? ($failed_count*100/$total_received_count) : 0
2701 printf $txt_fh ("$txt_format1$txt_format2\n", 'Received', @content) if $txt_fh;
2702 printf $htm_fh ("$htm_format1$htm_format2\n", 'Received', @content) if $htm_fh;
2704 $ws_global->write(++$row, 0, 'Received', $f_default);
2705 for (my $i=0; $i < scalar(@content); $i++) {
2706 if ($i == 4 || $i == 6) {
2707 $ws_global->write($row, $i+1, $content[$i]/100, $f_percent);
2710 $ws_global->write($row, $i+1, $content[$i], $f_default);
2716 if ($merge_reports) {
2717 $volume = volume_rounded($report_totals{Delivered}{Volume}, $report_totals{Delivered}{'Volume-gigs'});
2718 $total_delivered_messages = get_report_total($report_totals{Delivered},'Messages');
2719 $total_delivered_addresses = get_report_total($report_totals{Delivered},'Addresses');
2722 $volume = volume_rounded($total_delivered_data, $total_delivered_data_gigs);
2725 my @content=($volume, $total_delivered_messages, $total_delivered_addresses, @delivered_totals);
2726 printf $txt_fh ("$txt_format1\n", 'Delivered', @content) if $txt_fh;
2727 printf $htm_fh ("$htm_format1\n", 'Delivered', @content) if $htm_fh;
2730 $ws_global->write(++$row, 0, 'Delivered', $f_default);
2731 for (my $i=0; $i < scalar(@content); $i++) {
2732 $ws_global->write($row, $i+1, $content[$i], $f_default);
2736 if ($merge_reports) {
2737 foreach ('Rejects', 'Temp Rejects', 'Ham', 'Spam') {
2738 my $messages = get_report_total($report_totals{$_},'Messages');
2739 my $addresses = get_report_total($report_totals{$_},'Addresses');
2741 @content = ($_, '', $messages, '');
2742 push(@content,get_report_total($report_totals{$_},'Hosts')) if $do_sender{Host};
2743 #These rows do not have entries for the following columns (if specified)
2744 foreach ('Domain','Email','Edomain') {
2745 push(@content,'') if $do_sender{$_};
2748 printf $txt_fh ("$txt_format1\n", @content) if $txt_fh;
2749 printf $htm_fh ("$htm_format1\n", @content) if $htm_fh;
2750 $ws_global->write(++$row, 0, \@content) if $xls_fh;
2755 foreach my $total_aref (['Rejects',\%rejected_count_by_ip],
2756 ['Temp Rejects',\%temporarily_rejected_count_by_ip],
2757 ['Ham',\%ham_count_by_ip],
2758 ['Spam',\%spam_count_by_ip]) {
2759 #Count the number of messages of this type.
2761 map {$messages += $_} values %{$total_aref->[1]};
2763 if ($messages > 0) {
2764 @content = ($total_aref->[0], '', $messages, '');
2766 #Count the number of distinct IPs for the Hosts column.
2767 push(@content,scalar(keys %{$total_aref->[1]})) if $do_sender{Host};
2769 #These rows do not have entries for the following columns (if specified)
2770 foreach ('Domain','Email','Edomain') {
2771 push(@content,'') if $do_sender{$_};
2774 printf $txt_fh ("$txt_format1\n", @content) if $txt_fh;
2775 printf $htm_fh ("$htm_format1\n", @content) if $htm_fh;
2776 $ws_global->write(++$row, 0, \@content) if $xls_fh;
2781 printf $txt_fh "\n" if $txt_fh;
2782 printf $htm_fh "</table>\n" if $htm_fh;
2787 #######################################################################
2788 # print_user_patterns()
2790 # print_user_patterns();
2792 # Print the counts of user specified patterns.
2793 #######################################################################
2794 sub print_user_patterns {
2795 my $txt_format1 = " %-18s %6d";
2796 my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%d</td>";
2799 print $txt_fh "User Specified Patterns\n";
2800 print $txt_fh "-----------------------";
2801 print $txt_fh "\n Total\n";
2804 print $htm_fh "<hr><a name=\"Patterns\"></a><h2>User Specified Patterns</h2>\n";
2805 print $htm_fh "<table border=0 width=\"100%\">\n";
2806 print $htm_fh "<tr><td>\n";
2807 print $htm_fh "<table border=1>\n";
2808 print $htm_fh "<tr><th> </th><th>Total</th>\n";
2811 $ws_global->write($row++, $col, "User Specified Patterns", $f_header2);
2812 &set_worksheet_line($ws_global, $row++, 1, ["Total"], $f_headertab);
2817 if ($merge_reports) {
2818 # We are getting our data from previous reports.
2819 foreach $key (@user_descriptions) {
2820 my $count = get_report_total($report_totals{patterns}{$key},'Total');
2821 printf $txt_fh ("$txt_format1\n",$key,$count) if $txt_fh;
2822 printf $htm_fh ("$htm_format1\n",$key,$count) if $htm_fh;
2825 &set_worksheet_line($ws_global, $row++, 0, [$key,$count], $f_default);
2830 # We are getting our data from mainlog files.
2831 my $user_pattern_index = 0;
2832 foreach $key (@user_descriptions) {
2833 printf $txt_fh ("$txt_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $txt_fh;
2834 printf $htm_fh ("$htm_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $htm_fh;
2835 $ws_global->write($row++, 0, [$key,$user_pattern_totals[$user_pattern_index]]) if $xls_fh;
2836 $user_pattern_index++;
2839 print $txt_fh "\n" if $txt_fh;
2840 print $htm_fh "</table>\n\n" if $htm_fh;
2846 if ($hist_opt > 0) {
2847 my $user_pattern_index = 0;
2848 foreach $key (@user_descriptions) {
2849 print_histogram($key, 'occurence', @{$user_pattern_interval_count[$user_pattern_index]});
2850 $user_pattern_index++;
2855 #######################################################################
2860 # Print statistics about rejected mail.
2861 #######################################################################
2863 my($format1,$reason);
2865 my $txt_format1 = " %-40s %6d";
2866 my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%d</td>";
2869 print $txt_fh "Rejected mail by reason\n";
2870 print $txt_fh "-----------------------";
2871 print $txt_fh "\n Total\n";
2874 print $htm_fh "<hr><a name=\"patterns\"></a><h2>Rejected mail by reason</h2>\n";
2875 print $htm_fh "<table border=0 width=\"100%\"><tr><td><table border=1>\n";
2876 print $htm_fh "<tr><th> </th><th>Total</th>\n";
2879 $ws_global->write($row++, $col, "Rejected mail by reason", $f_header2);
2880 &set_worksheet_line($ws_global, $row++, 1, ["Total"], $f_headertab);
2884 my $href = ($merge_reports) ? $report_totals{rejected_mail_by_reason} : \%rejected_count_by_reason;
2885 my(@chartdatanames, @chartdatavals_count);
2887 foreach $reason (top_n_sort($topcount, $href, undef, undef)) {
2888 printf $txt_fh ("$txt_format1\n",$reason,$href->{$reason}) if $txt_fh;
2889 printf $htm_fh ("$htm_format1\n",$reason,$href->{$reason}) if $htm_fh;
2890 set_worksheet_line($ws_global, $row++, 0, [$reason,$href->{$reason}], $f_default) if $xls_fh;
2891 push(@chartdatanames, $reason);
2892 push(@chartdatavals_count, $href->{$reason});
2896 print $txt_fh "\n" if $txt_fh;
2899 print $htm_fh "</tr></table></td><td>";
2900 if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_count > 0)) {
2901 # calculate the graph
2904 \@chartdatavals_count
2906 my $graph = GD::Graph::pie->new(200, 200);
2908 x_label => 'Rejection Reasons',
2909 y_label => 'Messages',
2910 title => 'By count',
2912 my $gd = $graph->plot(\@data) or warn($graph->error);
2914 open(IMG, ">$chartdir/rejections_count.png") or die "Could not write $chartdir/rejections_count.png: $!\n";
2918 print $htm_fh "<img src=\"$chartrel/rejections_count.png\">";
2921 print $htm_fh "</td></tr></table>\n\n";
2929 #######################################################################
2930 # print_transport();
2932 # print_transport();
2934 # Print totals by transport.
2935 #######################################################################
2936 sub print_transport {
2937 my(@chartdatanames);
2938 my(@chartdatavals_count);
2939 my(@chartdatavals_vol);
2940 no integer; #Lose this for charting the data.
2942 my $txt_format1 = " %-18s %6s %6d";
2943 my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%s</td><td align=\"right\">%d</td>";
2946 print $txt_fh "Deliveries by transport\n";
2947 print $txt_fh "-----------------------";
2948 print $txt_fh "\n Volume Messages\n";
2951 print $htm_fh "<hr><a name=\"Transport\"></a><h2>Deliveries by Transport</h2>\n";
2952 print $htm_fh "<table border=0 width=\"100%\"><tr><td><table border=1>\n";
2953 print $htm_fh "<tr><th> </th><th>Volume</th><th>Messages</th>\n";
2956 $ws_global->write(++$row, $col, "Deliveries by transport", $f_header2);
2957 $ws_global->write(++$row, 1, ["Volume", "Messages"], $f_headertab);
2961 if ($merge_reports) {
2962 # We are getting our data from previous reports.
2963 foreach $key (sort keys %{$report_totals{transport}}) {
2964 my $count = get_report_total($report_totals{transport}{$key},'Messages');
2965 my @content=($key, volume_rounded($report_totals{transport}{$key}{Volume},
2966 $report_totals{transport}{$key}{'Volume-gigs'}), $count);
2967 push(@chartdatanames, $key);
2968 push(@chartdatavals_count, $count);
2969 push(@chartdatavals_vol, $report_totals{transport}{$key}{'Volume-gigs'}*$gig + $report_totals{transport}{$key}{Volume} );
2970 printf $txt_fh ("$txt_format1\n", @content) if $txt_fh;
2971 printf $htm_fh ("$htm_format1\n", @content) if $htm_fh;
2972 $ws_global->write(++$row, 0, \@content) if $xls_fh;
2976 # We are getting our data from mainlog files.
2977 foreach $key (sort keys %transported_data) {
2978 my @content=($key, volume_rounded($transported_data{$key},$transported_data_gigs{$key}),
2979 $transported_count{$key});
2980 push(@chartdatanames, $key);
2981 push(@chartdatavals_count, $transported_count{$key});
2982 push(@chartdatavals_vol, $transported_data_gigs{$key}*$gig + $transported_data{$key});
2983 printf $txt_fh ("$txt_format1\n", @content) if $txt_fh;
2984 printf $htm_fh ("$htm_format1\n", @content) if $htm_fh;
2985 $ws_global->write(++$row, 0, \@content) if $xls_fh;
2988 print $txt_fh "\n" if $txt_fh;
2990 print $htm_fh "</tr></table></td><td>";
2992 if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_count > 0))
2994 # calculate the graph
2997 \@chartdatavals_count
2999 my $graph = GD::Graph::pie->new(200, 200);
3001 x_label => 'Transport',
3002 y_label => 'Messages',
3003 title => 'By count',
3005 my $gd = $graph->plot(\@data) or warn($graph->error);
3007 open(IMG, ">$chartdir/transports_count.png") or die "Could not write $chartdir/transports_count.png: $!\n";
3011 print $htm_fh "<img src=\"$chartrel/transports_count.png\">";
3014 print $htm_fh "</td><td>";
3016 if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_vol > 0)) {
3021 my $graph = GD::Graph::pie->new(200, 200);
3023 title => 'By volume',
3025 my $gd = $graph->plot(\@data) or warn($graph->error);
3027 open(IMG, ">$chartdir/transports_vol.png") or die "Could not write $chartdir/transports_vol.png: $!\n";
3031 print $htm_fh "<img src=\"$chartrel/transports_vol.png\">";
3035 print $htm_fh "</td></tr></table>\n\n";
3041 #######################################################################
3046 # Print our totals by relay.
3047 #######################################################################
3049 my $row_print_relay=1;
3050 my $temp = "Relayed messages";
3051 print $htm_fh "<hr><a name=\"$temp\"></a><h2>$temp</h2>\n" if $htm_fh;
3052 if (scalar(keys %relayed) > 0 || $relayed_unshown > 0) {
3055 my $txt_format = "%7d %s\n => %s\n";
3056 my $htm_format = "<tr><td align=\"right\">%d</td><td>%s</td><td>%s</td>\n";
3058 printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh;
3060 print $htm_fh "<table border=1>\n";
3061 print $htm_fh "<tr><th>Count</th><th>From</th><th>To</th>\n";
3064 $ws_relayed->write($row_print_relay++, $col, $temp, $f_header2);
3065 &set_worksheet_line($ws_relayed, $row_print_relay++, 0, ["Count", "From", "To"], $f_headertab);
3070 foreach $key (sort keys %relayed) {
3071 my $count = $relayed{$key};
3074 my($one,$two) = split(/=> /, $key);
3075 my @content=($count, $one, $two);
3076 printf $txt_fh ($txt_format, @content) if $txt_fh;
3077 printf $htm_fh ($htm_format, @content) if $htm_fh;
3080 &set_worksheet_line($ws_relayed, $row_print_relay++, 0, \@content);
3085 print $htm_fh "</table>\n<p>\n" if $htm_fh;
3086 print $txt_fh "${spacing}Total: $shown (plus $relayed_unshown unshown)\n\n" if $txt_fh;
3087 print $htm_fh "${spacing}Total: $shown (plus $relayed_unshown unshown)\n\n" if $htm_fh;
3090 &set_worksheet_line($ws_relayed, $row_print_relay++, 0, [$shown, "Sum of shown" ]);
3091 &set_worksheet_line($ws_relayed, $row_print_relay++, 0, [$relayed_unshown, "unshown"]);
3096 print $txt_fh "No relayed messages\n-------------------\n\n" if $txt_fh;
3097 print $htm_fh "No relayed messages\n\n" if $htm_fh;
3107 #######################################################################
3112 # Print our errors. In HTML, we display them as a list rather than a table -
3113 # Netscape doesn't like large tables!
3114 #######################################################################
3116 my $total_errors = 0;
3119 if (scalar(keys %errors_count) != 0) {
3120 my $temp = "List of errors";
3121 my $htm_format = "<li>%d - %s\n";
3123 printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh;
3125 print $htm_fh "<hr><a name=\"errors\"></a><h2>$temp</h2>\n";
3126 print $htm_fh "<ul><li><b>Count - Error</b>\n";
3130 $ws_errors->write($row++, 0, $temp, $f_header2);
3131 &set_worksheet_line($ws_errors, $row++, 0, ["Count", "Error"], $f_headertab);
3136 foreach $key (sort keys %errors_count) {
3139 $text =~ s/\s\s+/ /g; #Convert multiple spaces to a single space.
3140 $total_errors += $errors_count{$key};
3143 printf $txt_fh ("%5d ", $errors_count{$key});
3144 my $text_remaining = $text;
3145 while (length($text_remaining) > 65) {
3146 my($first,$rest) = $text_remaining =~ /(.{50}\S*)\s+(.+)/;
3148 printf $txt_fh ("%s\n\t ", $first);
3149 $text_remaining = $rest;
3151 printf $txt_fh ("%s\n\n", $text_remaining);
3156 #Translate HTML tag characters. Sergey Sholokh.
3157 $text =~ s/\</\<\;/g;
3158 $text =~ s/\>/\>\;/g;
3160 printf $htm_fh ($htm_format,$errors_count{$key},$text);
3164 &set_worksheet_line($ws_errors, $row++, 0, [$errors_count{$key},$text]);
3168 $temp = "Errors encountered: $total_errors";
3171 print $txt_fh $temp, "\n";
3172 print $txt_fh "-" x length($temp),"\n";
3175 print $htm_fh "</ul>\n<p>\n";
3176 print $htm_fh $temp, "\n";
3180 &set_worksheet_line($ws_errors, $row++, 0, [$total_errors, "Sum of Errors encountered"]);
3187 #######################################################################
3188 # parse_old_eximstat_reports();
3190 # parse_old_eximstat_reports($fh);
3192 # Parse old eximstat output so we can merge daily stats to weekly stats and weekly to monthly etc.
3194 # To test that the merging still works after changes, do something like the following.
3195 # All the diffs should produce no output.
3197 # options='-bydomain -byemail -byhost -byedomain'
3198 # options="$options -show_rt1,2,4 -show_dt 1,2,4"
3199 # options="$options -pattern 'Completed Messages' /Completed/"
3200 # options="$options -pattern 'Received Messages' /<=/"
3202 # ./eximstats $options mainlog > mainlog.txt
3203 # ./eximstats $options -merge mainlog.txt > mainlog.2.txt
3204 # diff mainlog.txt mainlog.2.txt
3206 # ./eximstats $options -html mainlog > mainlog.html
3207 # ./eximstats $options -merge -html mainlog.txt > mainlog.2.html
3208 # diff mainlog.html mainlog.2.html
3210 # ./eximstats $options -merge mainlog.html > mainlog.3.txt
3211 # diff mainlog.txt mainlog.3.txt
3213 # ./eximstats $options -merge -html mainlog.html > mainlog.3.html
3214 # diff mainlog.html mainlog.3.html
3216 # ./eximstats $options -nvr mainlog > mainlog.nvr.txt
3217 # ./eximstats $options -merge mainlog.nvr.txt > mainlog.4.txt
3218 # diff mainlog.txt mainlog.4.txt
3220 # # double_mainlog.txt should have twice the values that mainlog.txt has.
3221 # ./eximstats $options mainlog mainlog > double_mainlog.txt
3222 #######################################################################
3223 sub parse_old_eximstat_reports {
3226 my(%league_table_value_entered, %league_table_value_was_zero, %table_order);
3228 my(%user_pattern_index);
3229 my $user_pattern_index = 0;
3230 map {$user_pattern_index{$_} = $user_pattern_index++} @user_descriptions;
3231 my $user_pattern_keys = join('|', @user_descriptions);
3234 PARSE_OLD_REPORT_LINE:
3235 if (/Exim statistics from ([\d\-]+ [\d:]+(\s+[\+\-]\d+)?) to ([\d\-]+ [\d:]+(\s+[\+\-]\d+)?)/) {
3236 $begin = $1 if ($1 lt $begin);
3237 $end = $3 if ($3 gt $end);
3239 elsif (/Grand total summary/) {
3240 # Fill in $report_totals{Received|Delivered}{Volume|Messages|Addresses|Hosts|Domains|...|Delayed|DelayedPercent|Failed|FailedPercent}
3241 my(@fields, @delivered_fields);
3242 my $doing_table = 0;
3244 $_ = html2txt($_); #Convert general HTML markup to text.
3245 s/At least one addr//g; #Another part of the HTML output we don't want.
3247 # TOTAL Volume Messages Addresses Hosts Domains Delayed Failed
3248 # Received 26MB 237 177 23 8 3.4% 28 11.8%
3249 # Delivered 13MB 233 250 99 88
3250 if (/TOTAL\s+(.*?)\s*$/) {
3252 @delivered_fields = split(/\s+/,$1);
3254 #Delayed and Failed have two columns each, so add the extra field names in.
3255 splice(@delivered_fields,-1,1,'DelayedPercent','Failed','FailedPercent');
3257 # Addresses only figure in the Delivered row, so remove them from the
3259 @fields = grep !/Addresses/, @delivered_fields;
3261 elsif (/(Received)\s+(.*?)\s*$/) {
3262 print STDERR "Parsing $_" if $debug;
3263 add_to_totals($report_totals{$1},\@fields,$2);
3265 elsif (/(Delivered)\s+(.*?)\s*$/) {
3266 print STDERR "Parsing $_" if $debug;
3267 add_to_totals($report_totals{$1},\@delivered_fields,$2);
3269 # If we're merging an old report which doesn't include addresses,
3270 # then use the Messages field instead.
3271 unless (grep(/Addresses/, @delivered_fields)) {
3273 line_to_hash(\%tmp,\@delivered_fields,$data);
3274 add_to_totals($report_totals{Delivered},['Addresses'],$tmp{Messages});
3277 elsif (/(Temp Rejects|Rejects|Ham|Spam)\s+(.*?)\s*$/) {
3278 print STDERR "Parsing $_" if $debug;
3279 add_to_totals($report_totals{$1},['Messages','Hosts'],$2);
3282 last if $doing_table;
3287 elsif (/User Specified Patterns/i) {
3288 #User Specified Patterns
3289 #-----------------------
3293 while (<$fh>) { last if (/Total/); } #Wait until we get the table headers.
3295 print STDERR "Parsing $_" if $debug;
3296 $_ = html2txt($_); #Convert general HTML markup to text.
3297 if (/^\s*(.*?)\s+(\d+)\s*$/) {
3298 $report_totals{patterns}{$1} = {} unless (defined $report_totals{patterns}{$1});
3299 add_to_totals($report_totals{patterns}{$1},['Total'],$2);
3301 last if (/^\s*$/); #Finished if we have a blank line.
3305 elsif (/(^|<h2>)($user_pattern_keys) per /o) {
3306 # Parse User defined pattern histograms if they exist.
3307 parse_histogram($fh, $user_pattern_interval_count[$user_pattern_index{$2}] );
3311 elsif (/Deliveries by transport/i) {
3312 #Deliveries by transport
3313 #-----------------------
3315 # :blackhole: 70KB 51
3316 # address_pipe 655KB 1
3319 while (<$fh>) { last if (/Volume/); } #Wait until we get the table headers.
3321 print STDERR "Parsing $_" if $debug;
3322 $_ = html2txt($_); #Convert general HTML markup to text.
3323 if (/(\S+)\s+(\d+\S*\s+\d+)/) {
3324 $report_totals{transport}{$1} = {} unless (defined $report_totals{transport}{$1});
3325 add_to_totals($report_totals{transport}{$1},['Volume','Messages'],$2);
3327 last if (/^\s*$/); #Finished if we have a blank line.
3330 elsif (/Messages received per/) {
3331 parse_histogram($fh, \@received_interval_count);
3333 elsif (/Deliveries per/) {
3334 parse_histogram($fh, \@delivered_interval_count);
3337 #elsif (/Time spent on the queue: (all messages|messages with at least one remote delivery)/) {
3338 elsif (/(Time spent on the queue|Delivery times|Receipt times): ((\S+) messages|messages with at least one remote delivery)((<[^>]*>)*\s*)$/) {
3339 #Time spent on the queue: all messages
3340 #-------------------------------------
3342 #Under 1m 217 91.9% 91.9%
3348 # Set a pointer to the queue bin so we can use the same code
3349 # block for both all messages and remote deliveries.
3350 #my $bin_aref = ($1 eq 'all messages') ? \@qt_all_bin : \@qt_remote_bin;
3351 my($bin_aref, $times_aref, $overflow_sref);
3352 if ($1 eq 'Time spent on the queue') {
3353 $times_aref = \@queue_times;
3354 if ($2 eq 'all messages') {
3355 $bin_aref = \@qt_all_bin;
3356 $overflow_sref = \$qt_all_overflow;
3359 $bin_aref = \@qt_remote_bin;
3360 $overflow_sref = \$qt_remote_overflow;
3363 elsif ($1 eq 'Delivery times') {
3364 $times_aref = \@delivery_times;
3365 if ($2 eq 'all messages') {
3366 $bin_aref = \@dt_all_bin;
3367 $overflow_sref = \$dt_all_overflow;
3370 $bin_aref = \@dt_remote_bin;
3371 $overflow_sref = \$dt_remote_overflow;
3375 unless (exists $rcpt_times_bin{$3}) {
3376 initialise_rcpt_times($3);
3378 $bin_aref = $rcpt_times_bin{$3};
3379 $times_aref = \@rcpt_times;
3380 $overflow_sref = \$rcpt_times_overflow{$3};
3384 my ($blank_lines, $reached_table) = (0,0);
3386 $_ = html2txt($_); #Convert general HTML markup to text.
3387 # The table is preceded by one blank line, and has one blank line
3388 # following it. As the table may be empty, the best way to determine
3389 # that we've finished it is to look for the second blank line.
3390 ++$blank_lines if /^\s*$/;
3391 last if ($blank_lines >=2); #Finished the table ?
3392 $reached_table = 1 if (/\d/);
3393 next unless $reached_table;
3394 my $previous_seconds_on_queue = 0;
3395 if (/^\s*(Under|Over|)\s+(\d+[smhdw])\s+(\d+)/) {
3396 print STDERR "Parsing $_" if $debug;
3397 my($modifier,$formatted_time,$count) = ($1,$2,$3);
3398 my $seconds = unformat_time($formatted_time);
3399 my $time_on_queue = ($seconds + $previous_seconds_on_queue) / 2;
3400 $previous_seconds_on_queue = $seconds;
3401 $time_on_queue = $seconds * 2 if ($modifier eq 'Over');
3403 for ($i = 0; $i <= $#$times_aref; $i++) {
3404 if ($time_on_queue < $times_aref->[$i]) {
3405 $$bin_aref[$i] += $count;
3409 $$overflow_sref += $count if ($i > $#$times_aref);
3415 elsif (/Relayed messages/) {
3419 # 1 addr.domain.com [1.2.3.4] a.user@domain.com
3420 # => addr2.domain2.com [5.6.7.8] a2.user2@domain2.com
3422 #<tr><td align="right">1</td><td>addr.domain.com [1.2.3.4] a.user@domain.com </td><td>addr2.domain2.com [5.6.7.8] a2.user2@domain2.com</td>
3424 my $reached_table = 0;
3427 unless ($reached_table) {
3428 last if (/No relayed messages/);
3429 $reached_table = 1 if (/^\s*\d/ || />\d+</);
3430 next unless $reached_table;
3432 if (/>(\d+)<.td><td>(.*?) ?<.td><td>(.*?)</) {
3433 update_relayed($1,$2,$3);
3435 elsif (/^\s*(\d+)\s+(.*?)\s*$/) {
3436 ($count,$sender) = ($1,$2);
3438 elsif (/=>\s+(.*?)\s*$/) {
3439 update_relayed($count,$sender,$1);
3442 last; #Finished the table ?
3447 elsif (/Top (.*?) by (message count|volume)/) {
3448 #Top 50 sending hosts by message count
3449 #-------------------------------------
3452 # Could also have average values for HTML output.
3453 # 48 1468KB 30KB local
3455 my($category,$by_count_or_volume) = ($1,$2);
3457 #As we show 2 views of each table (by count and by volume),
3458 #most (but not all) entries will appear in both tables.
3459 #Set up a hash to record which entries we have already seen
3460 #and one to record which ones we are seeing for the first time.
3461 if ($by_count_or_volume =~ /count/) {
3462 undef %league_table_value_entered;
3463 undef %league_table_value_was_zero;
3467 #As this section processes multiple different table categories,
3468 #set up pointers to the hashes to be updated.
3469 my($messages_href,$addresses_href,$data_href,$data_gigs_href);
3470 if ($category =~ /local sender/) {
3471 $messages_href = \%received_count_user;
3472 $addresses_href = undef;
3473 $data_href = \%received_data_user;
3474 $data_gigs_href = \%received_data_gigs_user;
3476 elsif ($category =~ /sending (\S+?)s?\b/) {
3477 #Top 50 sending (host|domain|email|edomain)s
3478 #Top sending (host|domain|email|edomain)
3479 $messages_href = \%{$received_count{"\u$1"}};
3480 $data_href = \%{$received_data{"\u$1"}};
3481 $data_gigs_href = \%{$received_data_gigs{"\u$1"}};
3483 elsif ($category =~ /local destination/) {
3484 $messages_href = \%delivered_messages_user;
3485 $addresses_href = \%delivered_addresses_user;
3486 $data_href = \%delivered_data_user;
3487 $data_gigs_href = \%delivered_data_gigs_user;
3489 elsif ($category =~ /local domain destination/) {
3490 $messages_href = \%delivered_messages_local_domain;
3491 $addresses_href = \%delivered_addresses_local_domain;
3492 $data_href = \%delivered_data_local_domain;
3493 $data_gigs_href = \%delivered_data_gigs_local_domain;
3495 elsif ($category =~ /(\S+) destination/) {
3496 #Top 50 (host|domain|email|edomain) destinations
3497 #Top (host|domain|email|edomain) destination
3498 $messages_href = \%{$delivered_messages{"\u$1"}};
3499 $addresses_href = \%{$delivered_addresses{"\u$1"}};
3500 $data_href = \%{$delivered_data{"\u$1"}};
3501 $data_gigs_href = \%{$delivered_data_gigs{"\u$1"}};
3503 elsif ($category =~ /temporarily rejected ips/) {
3504 $messages_href = \%temporarily_rejected_count_by_ip;
3506 elsif ($category =~ /rejected ips/) {
3507 $messages_href = \%rejected_count_by_ip;
3509 elsif ($category =~ /non-rejected spamming ips/) {
3510 $messages_href = \%spam_count_by_ip;
3512 elsif ($category =~ /mail temporary rejection reasons/) {
3513 $messages_href = \%temporarily_rejected_count_by_reason;
3515 elsif ($category =~ /mail rejection reasons/) {
3516 $messages_href = \%rejected_count_by_reason;
3519 my $reached_table = 0;
3522 # Watch out for empty tables.
3523 goto PARSE_OLD_REPORT_LINE if (/<h2>/ or (/^\s*[a-zA-Z]/ && !/^\s*Messages/));
3525 $_ = html2txt($_); #Convert general HTML markup to text.
3527 # Messages Addresses Bytes Average
3528 if (/^\s*Messages/) {
3529 my $pattern = '^\s*(\d+)';
3530 $pattern .= (/Addresses/) ? '\s+(\d+)' : '()';
3531 $pattern .= (/Bytes/) ? '\s+([\dKMGB]+)' : '()';
3532 $pattern .= (/Average/) ? '\s+[\dKMGB]+' : '';
3533 $pattern .= '\s+(.*?)\s*$';
3534 $row_re = qr/$pattern/;
3538 next unless $reached_table;
3540 my($messages, $addresses, $rounded_volume, $entry);
3543 ($messages, $addresses, $rounded_volume, $entry) = ($1, $2, $3, $4);
3546 #Else we have finished the table and we may need to do some
3547 #kludging to retain the order of the entries.
3549 if ($by_count_or_volume =~ /volume/) {
3550 #Add a few bytes to appropriate entries to preserve the order.
3551 foreach $rounded_volume (keys %table_order) {
3552 #For each rounded volume, we want to create a list which has things
3553 #ordered from the volume table at the front, and additional things
3554 #from the count table ordered at the back.
3555 @{$table_order{$rounded_volume}{volume}} = () unless defined $table_order{$rounded_volume}{volume};
3556 @{$table_order{$rounded_volume}{'message count'}} = () unless defined $table_order{$rounded_volume}{'message count'};
3558 map {$mark{$_} = 1} @{$table_order{$rounded_volume}{volume}};
3559 @order = @{$table_order{$rounded_volume}{volume}};
3560 map {push(@order,$_)} grep(!$mark{$_},@{$table_order{$rounded_volume}{'message count'}});
3562 my $bonus_bytes = $#order;
3563 $bonus_bytes = 511 if ($bonus_bytes > 511); #Don't go over the half-K boundary!
3564 while (@order and ($bonus_bytes > 0)) {
3565 my $entry = shift(@order);
3566 if ($league_table_value_was_zero{$entry}) {
3567 $$data_href{$entry} += $bonus_bytes;
3568 print STDERR "$category by $by_count_or_volume: added $bonus_bytes bonus bytes to $entry\n" if $debug;
3577 # Store a new table entry.
3579 # Add the entry into the %table_order hash if it has a rounded
3580 # volume (KB/MB/GB).
3581 push(@{$table_order{$rounded_volume}{$by_count_or_volume}},$entry) if ($rounded_volume =~ /\D/);
3583 unless ($league_table_value_entered{$entry}) {
3584 $league_table_value_entered{$entry} = 1;
3585 unless ($$messages_href{$entry}) {
3586 $$messages_href{$entry} = 0;
3587 $$addresses_href{$entry} = 0;
3588 $$data_href{$entry} = 0;
3589 $$data_gigs_href{$entry} = 0;
3590 $league_table_value_was_zero{$entry} = 1;
3593 $$messages_href{$entry} += $messages;
3595 # When adding the addresses, be aware that we could be merging
3596 # an old report which does not include addresses. In this case,
3597 # we add the messages instead.
3598 $$addresses_href{$entry} += ($addresses) ? $addresses : $messages;
3600 #Add the rounded value to the data and data_gigs hashes.
3601 un_round($rounded_volume,\$$data_href{$entry},\$$data_gigs_href{$entry}) if $rounded_volume;
3602 print STDERR "$category by $by_count_or_volume: added $messages,$rounded_volume to $entry\n" if $debug;
3607 elsif (/List of errors/) {
3611 # 1 07904931641@one2one.net R=external T=smtp: SMTP error
3612 # from remote mailer after RCPT TO:<07904931641@one2one.net>:
3613 # host mail.one2one.net [193.133.192.24]: 550 User unknown
3615 #<li>1 - ally.dufc@dunbar.org.uk R=external T=smtp: SMTP error from remote mailer after RCPT TO:<ally.dufc@dunbar.org.uk>: host mail.dunbar.org.uk [216.167.89.88]: 550 Unknown local part ally.dufc in <ally.dufc@dunbar.org.uk>
3618 my $reached_table = 0;
3619 my($count,$error,$blanks);
3621 $reached_table = 1 if (/^( *|<li>)(\d+)/);
3622 next unless $reached_table;
3624 s/^<li>(\d+) -/$1/; #Convert an HTML line to a text line.
3625 $_ = html2txt($_); #Convert general HTML markup to text.
3628 $error .= ' ' . $1; #Join a multiline error.
3630 elsif (/^\s*(\d+)\s+(.*)/) {
3632 #Finished with a previous multiline error so save it.
3633 $errors_count{$error} = 0 unless $errors_count{$error};
3634 $errors_count{$error} += $count;
3636 ($count,$error) = ($1,$2);
3638 elsif (/Errors encountered/) {
3640 #Finished the section, so save our stored last error.
3641 $errors_count{$error} = 0 unless $errors_count{$error};
3642 $errors_count{$error} += $count;
3652 #######################################################################
3653 # parse_histogram($fh, \@delivered_interval_count);
3654 # Parse a histogram into the provided array of counters.
3655 #######################################################################
3656 sub parse_histogram {
3657 my($fh, $counters_aref) = @_;
3659 # Messages received per hour (each dot is 2 messages)
3660 #---------------------------------------------------
3662 #00-01 106 .....................................................
3663 #01-02 103 ...................................................
3665 my $reached_table = 0;
3667 $reached_table = 1 if (/^00/);
3668 next unless $reached_table;
3669 print STDERR "Parsing $_" if $debug;
3670 if (/^(\d+):(\d+)\s+(\d+)/) { #hh:mm start time format ?
3671 $$counters_aref[($1*60 + $2)/$hist_interval] += $3 if $hist_opt;
3673 elsif (/^(\d+)-(\d+)\s+(\d+)/) { #hh-hh start-end time format ?
3674 $$counters_aref[($1*60)/$hist_interval] += $3 if $hist_opt;
3676 else { #Finished the table ?
3683 #######################################################################
3686 # update_relayed($count,$sender,$recipient);
3688 # Adds an entry into the %relayed hash. Currently only used when
3690 #######################################################################
3691 sub update_relayed {
3692 my($count,$sender,$recipient) = @_;
3694 #When generating the key, put in the 'H=' and 'A=' which can be used
3696 my $key = "H=$sender => H=$recipient";
3697 $key =~ s/ ([^=\s]+\@\S+|<>)/ A=$1/g;
3698 if (!defined $relay_pattern || $key !~ /$relay_pattern/o) {
3699 $relayed{$key} = 0 if !defined $relayed{$key};
3700 $relayed{$key} += $count;
3703 $relayed_unshown += $count;
3708 #######################################################################
3711 # add_to_totals(\%totals,\@keys,$values);
3713 # Given a line of space separated values, add them into the provided hash using @keys
3716 # If the value contains a '%', then the value is set rather than added. Otherwise, we
3717 # convert the value to bytes and gigs. The gigs get added to I<Key>-gigs.
3718 #######################################################################
3720 my($totals_href,$keys_aref,$values) = @_;
3721 my(@values) = split(/\s+/,$values);
3723 for(my $i = 0; $i < @values && $i < @$keys_aref; ++$i) {
3724 my $key = $keys_aref->[$i];
3725 if ($values[$i] =~ /%/) {
3726 $$totals_href{$key} = $values[$i];
3729 $$totals_href{$key} = 0 unless ($$totals_href{$key});
3730 $$totals_href{"$key-gigs"} = 0 unless ($$totals_href{"$key-gigs"});
3731 un_round($values[$i], \$$totals_href{$key}, \$$totals_href{"$key-gigs"});
3732 print STDERR "Added $values[$i] to $key - $$totals_href{$key} , " . $$totals_href{"$key-gigs"} . "GB.\n" if $debug;
3738 #######################################################################
3741 # line_to_hash(\%hash,\@keys,$line);
3743 # Given a line of space separated values, set them into the provided hash
3744 # using @keys as the hash keys.
3745 #######################################################################
3747 my($href,$keys_aref,$values) = @_;
3748 my(@values) = split(/\s+/,$values);
3749 for(my $i = 0; $i < @values && $i < @$keys_aref; ++$i) {
3750 $$href{$keys_aref->[$i]} = $values[$i];
3755 #######################################################################
3756 # get_report_total();
3758 # $total = get_report_total(\%hash,$key);
3760 # If %hash contains values split into Units and Gigs, we calculate and return
3762 # $hash{$key} + 1024*1024*1024 * $hash{"${key}-gigs"}
3763 #######################################################################
3764 sub get_report_total {
3766 my($hash_ref,$key) = @_;
3767 if ($$hash_ref{"${key}-gigs"}) {
3768 return $$hash_ref{$key} + $gig * $$hash_ref{"${key}-gigs"};
3770 return $$hash_ref{$key} || 0;
3773 #######################################################################
3776 # $text_line = html2txt($html_line);
3778 # Convert a line from html to text. Currently we just convert HTML tags to spaces
3779 # and convert >, <, and tags back.
3780 #######################################################################
3784 # Convert HTML tags to spacing. Note that the reports may contain <Userid> and
3785 # <Userid@Domain> words, so explicitly specify the HTML tags we will remove
3786 # (the ones used by this program). If someone is careless enough to have their
3787 # Userid the same as an HTML tag, there's not much we can do about it.
3788 s/<\/?(html|head|title|body|h\d|ul|li|a\s+|table|tr|td|th|pre|hr|p|br)\b.*?>/ /g;
3790 s/\<\;/\</og; #Convert '<' to '<'.
3791 s/\>\;/\>/og; #Convert '>' to '>'.
3792 s/\ \;/ /og; #Convert ' ' to ' '.
3796 #######################################################################
3799 # $arg = get_next_arg();
3801 # Because eximstats arguments are often passed as variables,
3802 # we can't rely on shell parsing to deal with quotes. This
3803 # subroutine returns $ARGV[1] and does a shift. If $ARGV[1]
3804 # starts with a quote (' or "), and doesn't end in one, then
3805 # we append the next argument to it and shift again. We repeat
3806 # until we've got all of the argument.
3808 # This isn't perfect as all white space gets reduced to one space,
3809 # but it's as good as we can get! If it's essential that spacing
3810 # be preserved precisely, then you get that by not using shell
3812 #######################################################################
3815 my $matched_pattern = 0;
3817 $arg .= ' ' if $arg;
3818 $arg .= $ARGV[1]; shift(@ARGV);
3819 if ($arg !~ /^['"]/) {
3820 $matched_pattern = 1;
3823 if ($arg =~ s/^(['"])(.*)\1$/$2/) {
3824 $matched_pattern = 1;
3828 die "Mismatched argument quotes - <$arg>.\n" unless $matched_pattern;
3832 #######################################################################
3833 # set_worksheet_line($ws_global, $startrow, $startcol, \@content, $format);
3835 # set values to a sequence of cells in a row.
3837 #######################################################################
3838 sub set_worksheet_line {
3839 my ($worksheet, $row, $col, $content, $format) = @_;
3841 foreach my $token (@$content)
3843 $worksheet->write($row, $col++, $token, $format );
3848 #######################################################################
3849 # @rcpt_times = parse_time_list($string);
3851 # Parse a comma separated list of time values in seconds given by
3852 # the user and fill an array.
3854 # Return a default list if $string is undefined.
3855 # Return () if $string eq '0'.
3856 #######################################################################
3857 sub parse_time_list {
3859 if (! defined $string) {
3860 return(60, 5*60, 15*60, 30*60, 60*60, 3*60*60, 6*60*60, 12*60*60, 24*60*60);
3862 my(@times) = split(/,/, $string);
3863 foreach my $q (@times) { $q = eval($q) + 0 }
3864 @times = sort { $a <=> $b } @times;
3865 @times = () if ($#times == 0 && $times[0] == 0);
3870 #######################################################################
3871 # initialise_rcpt_times($protocol);
3872 # Initialise an array of rcpt_times to 0 for the specified protocol.
3873 #######################################################################
3874 sub initialise_rcpt_times {
3876 for (my $i = 0; $i <= $#rcpt_times; ++$i) {
3877 $rcpt_times_bin{$protocol}[$i] = 0;
3879 $rcpt_times_overflow{$protocol} = 0;
3883 ##################################################
3885 ##################################################
3888 $last_timestamp = '';
3892 $show_transport = 1;
3894 $local_league_table = 1;
3895 $include_remote_users = 0;
3896 $include_original_destination = 0;
3898 $volume_rounding = 1;
3899 $localtime_offset = calculate_localtime_offset(); # PH/FANF
3902 $charts_option_specified = 0;
3906 @queue_times = parse_time_list();
3908 @delivery_times = ();
3911 $offset_seconds = 0;
3917 my(%output_files); # What output files have been specified?
3921 while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') {
3922 if ($ARGV[0] =~ /^\-h(\d+)$/) { $hist_opt = $1 }
3923 elsif ($ARGV[0] =~ /^\-ne$/) { $show_errors = 0 }
3924 elsif ($ARGV[0] =~ /^\-nr(.?)(.*)\1$/) {
3925 if ($1 eq "") { $show_relay = 0 } else { $relay_pattern = $2 }
3927 elsif ($ARGV[0] =~ /^\-q([,\d\+\-\*\/]+)$/) { @queue_times = parse_time_list($1) }
3928 elsif ($ARGV[0] =~ /^-nt$/) { $show_transport = 0 }
3929 elsif ($ARGV[0] =~ /^\-nt(.?)(.*)\1$/)
3931 if ($1 eq "") { $show_transport = 0 } else { $transport_pattern = $2 }
3933 elsif ($ARGV[0] =~ /^-t(\d+)$/) { $topcount = $1 }
3934 elsif ($ARGV[0] =~ /^-tnl$/) { $local_league_table = 0 }
3935 elsif ($ARGV[0] =~ /^-txt=?(\S*)$/) { $txt_fh = get_filehandle($1,\%output_files) }
3936 elsif ($ARGV[0] =~ /^-html=?(\S*)$/) { $htm_fh = get_filehandle($1,\%output_files) }
3937 elsif ($ARGV[0] =~ /^-xls=?(\S*)$/) {
3938 if ($HAVE_Spreadsheet_WriteExcel) {
3939 $xls_fh = get_filehandle($1,\%output_files);
3942 warn "WARNING: CPAN Module Spreadsheet::WriteExcel not installed. Obtain from www.cpan.org\n";
3945 elsif ($ARGV[0] =~ /^-merge$/) { $merge_reports = 1 }
3946 elsif ($ARGV[0] =~ /^-charts$/) {
3948 warn "WARNING: CPAN Module GD::Graph::pie not installed. Obtain from www.cpan.org\n" unless $HAVE_GD_Graph_pie;
3949 warn "WARNING: CPAN Module GD::Graph::linespoints not installed. Obtain from www.cpan.org\n" unless $HAVE_GD_Graph_linespoints;
3951 elsif ($ARGV[0] =~ /^-chartdir$/) { $chartdir = $ARGV[1]; shift; $charts_option_specified = 1; }
3952 elsif ($ARGV[0] =~ /^-chartrel$/) { $chartrel = $ARGV[1]; shift; $charts_option_specified = 1; }
3953 elsif ($ARGV[0] =~ /^-include_original_destination$/) { $include_original_destination = 1 }
3954 elsif ($ARGV[0] =~ /^-cache$/) { } #Not currently used.
3955 elsif ($ARGV[0] =~ /^-byhost$/) { $do_sender{Host} = 1 }
3956 elsif ($ARGV[0] =~ /^-bydomain$/) { $do_sender{Domain} = 1 }
3957 elsif ($ARGV[0] =~ /^-byemail$/) { $do_sender{Email} = 1 }
3958 elsif ($ARGV[0] =~ /^-byemaildomain$/) { $do_sender{Edomain} = 1 }
3959 elsif ($ARGV[0] =~ /^-byedomain$/) { $do_sender{Edomain} = 1 }
3960 elsif ($ARGV[0] =~ /^-bylocaldomain$/) { $do_local_domain = 1 }
3961 elsif ($ARGV[0] =~ /^-emptyok$/) { $emptyOK = 1 }
3962 elsif ($ARGV[0] =~ /^-nvr$/) { $volume_rounding = 0 }
3963 elsif ($ARGV[0] =~ /^-show_rt([,\d\+\-\*\/]+)?$/) { @rcpt_times = parse_time_list($1) }
3964 elsif ($ARGV[0] =~ /^-show_dt([,\d\+\-\*\/]+)?$/) { @delivery_times = parse_time_list($1) }
3965 elsif ($ARGV[0] =~ /^-d$/) { $debug = 1 }
3966 elsif ($ARGV[0] =~ /^--?h(elp)?$/){ help() }
3967 elsif ($ARGV[0] =~ /^-t_remote_users$/) { $include_remote_users = 1 }
3968 elsif ($ARGV[0] =~ /^-pattern$/)
3970 push(@user_descriptions,get_next_arg());
3971 push(@user_patterns,get_next_arg());
3973 elsif ($ARGV[0] =~ /^-utc$/)
3975 # We don't need this value if the log is in UTC.
3976 $localtime_offset = undef;
3980 print STDERR "Eximstats: Unknown or malformed option $ARGV[0]\n";
3986 # keep old default behaviour
3987 if (! ($xls_fh or $htm_fh or $txt_fh)) {
3991 # Check that all the charts options are specified.
3992 warn "-charts option not specified. Use -help for help.\n" if ($charts_option_specified && ! $charts);
3994 # Default to display tables by sending Host.
3995 $do_sender{Host} = 1 unless ($do_sender{Domain} || $do_sender{Email} || $do_sender{Edomain});
3997 # prepare xls Excel Workbook
3998 if (defined $xls_fh) {
4000 # Create a new Excel workbook
4001 $workbook = Spreadsheet::WriteExcel->new($xls_fh);
4004 $ws_global = $workbook->addworksheet('Exim Statistik');
4005 # show $ws_global as initial sheet
4006 $ws_global->set_first_sheet();
4007 $ws_global->activate();
4010 $ws_relayed = $workbook->addworksheet('Relayed Messages');
4011 $ws_relayed->set_column(1, 2, 80);
4014 $ws_errors = $workbook->addworksheet('Errors');
4019 $ws_global->set_column(0, 2, 20); # Columns B-D width set to 30
4020 $ws_global->set_column(3, 3, 15); # Columns B-D width set to 30
4021 $ws_global->set_column(4, 4, 25); # Columns B-D width set to 30
4024 $f_default = $workbook->add_format();
4026 $f_header1 = $workbook->add_format();
4027 $f_header1->set_bold();
4028 #$f_header1->set_color('red');
4029 $f_header1->set_size('15');
4030 $f_header1->set_valign();
4031 # $f_header1->set_align('center');
4032 # $ws_global->write($row++, 2, "Testing Headers 1", $f_header1);
4034 $f_header2 = $workbook->add_format();
4035 $f_header2->set_bold();
4036 $f_header2->set_size('12');
4037 $f_header2->set_valign();
4038 # $ws_global->write($row++, 2, "Testing Headers 2", $f_header2);
4040 # Create another header2 for use in merged cells.
4041 $f_header2_m = $workbook->add_format();
4042 $f_header2_m->set_bold();
4043 $f_header2_m->set_size('8');
4044 $f_header2_m->set_valign();
4045 $f_header2_m->set_align('center');
4047 $f_percent = $workbook->add_format();
4048 $f_percent->set_num_format('0.0%');
4050 $f_headertab = $workbook->add_format();
4051 $f_headertab->set_bold();
4052 $f_headertab->set_valign();
4053 # $ws_global->write($row++, 2, "Testing Headers tab", $f_headertab);
4058 # Initialise the queue/delivery/rcpt time counters.
4059 for (my $i = 0; $i <= $#queue_times; $i++) {
4060 $qt_all_bin[$i] = 0;
4061 $qt_remote_bin[$i] = 0;
4063 for (my $i = 0; $i <= $#delivery_times; $i++) {
4064 $dt_all_bin[$i] = 0;
4065 $dt_remote_bin[$i] = 0;
4067 initialise_rcpt_times('all');
4070 # Compute the number of slots for the histogram
4073 if ($hist_opt > 60 || 60 % $hist_opt != 0)
4075 print STDERR "Eximstats: -h must specify a factor of 60\n";
4078 $hist_interval = 60/$hist_opt; #Interval in minutes.
4079 $hist_number = (24*60)/$hist_interval; #Number of intervals per day.
4080 @received_interval_count = (0) x $hist_number;
4081 @delivered_interval_count = (0) x $hist_number;
4082 my $user_pattern_index = 0;
4083 for (my $user_pattern_index = 0; $user_pattern_index <= $#user_patterns; ++$user_pattern_index) {
4084 @{$user_pattern_interval_count[$user_pattern_index]} = (0) x $hist_number;
4086 @dt_all_bin = (0) x $hist_number;
4087 @dt_remote_bin = (0) x $hist_number;
4090 #$queue_unknown = 0;
4092 $total_received_data = 0;
4093 $total_received_data_gigs = 0;
4094 $total_received_count = 0;
4096 $total_delivered_data = 0;
4097 $total_delivered_data_gigs = 0;
4098 $total_delivered_messages = 0;
4099 $total_delivered_addresses = 0;
4101 $qt_all_overflow = 0;
4102 $qt_remote_overflow = 0;
4103 $dt_all_overflow = 0;
4104 $dt_remote_overflow = 0;
4106 $relayed_unshown = 0;
4107 $message_errors = 0;
4108 $begin = "9999-99-99 99:99:99";
4109 $end = "0000-00-00 00:00:00";
4111 foreach $section ('Received','Delivered','Temp Rejects', 'Rejects','Ham','Spam') {
4112 foreach $type ('Volume','Messages','Delayed','Failed','Hosts','Domains','Emails','Edomains') {
4113 $report_totals{$section}{$type} = 0;
4117 # Generate our parser.
4118 my $parser = generate_parser();
4123 # Scan the input files and collect the data
4124 foreach my $file (@ARGV) {
4125 if ($file =~ /\.gz/) {
4126 unless (open(FILE,"gunzip -c $file |")) {
4127 print STDERR "Failed to gunzip -c $file: $!";
4131 elsif ($file =~ /\.Z/) {
4132 unless (open(FILE,"uncompress -c $file |")) {
4133 print STDERR "Failed to uncompress -c $file: $!";
4138 unless (open(FILE,$file)) {
4139 print STDERR "Failed to read $file: $!";
4143 #Now parse the filehandle, updating the global variables.
4144 parse($parser,\*FILE);
4149 #No files provided. Parse STDIN, updating the global variables.
4150 parse($parser,\*STDIN);
4154 if ($begin eq "9999-99-99 99:99:99" && ! $emptyOK) {
4155 print STDERR "**** No valid log lines read\n";
4159 # Output our results.
4161 print_grandtotals();
4163 # Print counts of user specified patterns if required.
4164 print_user_patterns() if @user_patterns;
4166 # Print rejection reasons.
4169 # Print totals by transport if required.
4170 print_transport() if $show_transport;
4172 # Print the deliveries per interval as a histogram, unless configured not to.
4173 # First find the maximum in one interval and scale accordingly.
4174 if ($hist_opt > 0) {
4175 print_histogram("Messages received", 'message', @received_interval_count);
4176 print_histogram("Deliveries", 'delivery', @delivered_interval_count);
4179 # Print times on queue if required.
4180 if ($#queue_times >= 0) {
4181 print_duration_table("Time spent on the queue", "all messages", \@queue_times, \@qt_all_bin,$qt_all_overflow);
4182 print_duration_table("Time spent on the queue", "messages with at least one remote delivery", \@queue_times, \@qt_remote_bin,$qt_remote_overflow);
4185 # Print delivery times if required.
4186 if ($#delivery_times >= 0) {
4187 print_duration_table("Delivery times", "all messages", \@delivery_times, \@dt_all_bin,$dt_all_overflow);
4188 print_duration_table("Delivery times", "messages with at least one remote delivery", \@delivery_times, \@dt_remote_bin,$dt_remote_overflow);
4191 # Print rcpt times if required.
4192 if ($#rcpt_times >= 0) {
4193 foreach my $protocol ('all', grep(!/^all$/, sort keys %rcpt_times_bin)) {
4194 print_duration_table("Receipt times", "$protocol messages", \@rcpt_times, $rcpt_times_bin{$protocol}, $rcpt_times_overflow{$protocol});
4198 # Print relay information if required.
4199 print_relay() if $show_relay;
4201 # Print the league tables, if topcount isn't zero.
4202 if ($topcount > 0) {
4203 my($ws_rej, $ws_top50, $ws_rej_row, $ws_top50_row, $ws_temp_rej, $ws_temp_rej_row);
4204 $ws_rej_row = $ws_temp_rej_row = $ws_top50_row = 0;
4206 $ws_top50 = $workbook->addworksheet('Deliveries');
4207 $ws_rej = $workbook->addworksheet('Rejections') if (%rejected_count_by_reason || %rejected_count_by_ip || %spam_count_by_ip);
4208 $ws_temp_rej = $workbook->addworksheet('Temporary Rejections') if (%temporarily_rejected_count_by_reason || %temporarily_rejected_count_by_ip);
4211 print_league_table("mail rejection reason", \%rejected_count_by_reason, undef, undef, undef, $ws_rej, \$ws_rej_row) if %rejected_count_by_reason;
4212 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;
4214 foreach ('Host','Domain','Email','Edomain') {
4215 next unless $do_sender{$_};
4216 print_league_table("sending \l$_", $received_count{$_}, undef, $received_data{$_},$received_data_gigs{$_}, $ws_top50, \$ws_top50_row);
4219 print_league_table("local sender", \%received_count_user, undef,
4220 \%received_data_user,\%received_data_gigs_user, $ws_top50, \$ws_top50_row) if (($local_league_table || $include_remote_users) && %received_count_user);
4221 foreach ('Host','Domain','Email','Edomain') {
4222 next unless $do_sender{$_};
4223 print_league_table("\l$_ destination", $delivered_messages{$_}, $delivered_addresses{$_}, $delivered_data{$_},$delivered_data_gigs{$_}, $ws_top50, \$ws_top50_row);
4225 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);
4226 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);
4228 print_league_table("rejected ip", \%rejected_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %rejected_count_by_ip;
4229 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;
4230 print_league_table("non-rejected spamming ip", \%spam_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %spam_count_by_ip;
4234 # Print the error statistics if required.
4235 print_errors() if $show_errors;
4237 print $htm_fh "</body>\n</html>\n" if $htm_fh;
4240 $txt_fh->close if $txt_fh && ref $txt_fh;
4241 $htm_fh->close if $htm_fh;
4244 # close Excel Workbook
4245 $ws_global->set_first_sheet();
4246 # FIXME: whyever - activate does not work :-/
4247 $ws_global->activate();