| 1 | #!PERL_COMMAND -w |
| 2 | # $Cambridge: exim/src/src/eximstats.src,v 1.18 2009/10/19 11:41:27 nm4 Exp $ |
| 3 | |
| 4 | # Copyright (c) 2001 University of Cambridge. |
| 5 | # See the file NOTICE for conditions of use and distribution. |
| 6 | |
| 7 | # Perl script to generate statistics from one or more Exim log files. |
| 8 | |
| 9 | # Usage: eximstats [<options>] <log file> <log file> ... |
| 10 | |
| 11 | # 1996-05-21: Ignore lines not starting with valid date/time, just in case |
| 12 | # these get into a log file. |
| 13 | # 1996-11-19: Add the -h option to control the size of the histogram, |
| 14 | # and optionally turn it off. |
| 15 | # Use some Perl 5 things; it should be everywhere by now. |
| 16 | # Add the Perl -w option and rewrite so no warnings are given. |
| 17 | # Add the -t option to control the length of the "top" listing. |
| 18 | # Add the -ne, -nt options to turn off errors and transport |
| 19 | # information. |
| 20 | # Add information about length of time on queue, and -q<list> to |
| 21 | # control the intervals and turn it off. |
| 22 | # Add count and percentage of delayed messages to the Received |
| 23 | # line. |
| 24 | # Show total number of errors. |
| 25 | # Add count and percentage of messages with errors to Received |
| 26 | # line. |
| 27 | # Add information about relaying and -nr to suppress it. |
| 28 | # 1997-02-03 Merged in some of the things Nigel Metheringham had done: |
| 29 | # Re-worded headings |
| 30 | # Added received histogram as well as delivered |
| 31 | # Added local senders' league table |
| 32 | # Added local recipients' league table |
| 33 | # 1997-03-10 Fixed typo "destinationss" |
| 34 | # Allow for intermediate address between final and original |
| 35 | # when testing for relaying |
| 36 | # Give better message when no input |
| 37 | # 1997-04-24 Fixed bug in layout of error listing that was depending on |
| 38 | # text length (output line got repeated). |
| 39 | # 1997-05-06 Bug in option decoding when only one option. |
| 40 | # Overflow bug when handling very large volumes. |
| 41 | # 1997-10-28 Updated to handle revised log format that might show |
| 42 | # HELO name as well as host name before IP number |
| 43 | # 1998-01-26 Bugs in the function for calculating the number of seconds |
| 44 | # since 1970 from a log date |
| 45 | # 1998-02-02 Delivery to :blackhole: doesn't have a T= entry in the log |
| 46 | # line; cope with this, thereby avoiding undefined problems |
| 47 | # Very short log line gave substring error |
| 48 | # 1998-02-03 A routed delivery to a local transport may not have <> in the |
| 49 | # log line; terminate the address at white space, not < |
| 50 | # 1998-09-07 If first line of input was a => line, $thissize was undefined; |
| 51 | # ensure it is zero. |
| 52 | # 1998-12-21 Adding of $thissize from => line should have been adding $size. |
| 53 | # Oops. Should have looked more closely when fixing the previous |
| 54 | # bug! |
| 55 | # 1999-11-12 Increased the field widths for printed integers; numbers are |
| 56 | # bigger than originally envisaged. |
| 57 | # 2001-03-21 Converted seconds() routine to use Time::Local, fixing a bug |
| 58 | # whereby seconds($timestamp) - id_seconds($id) gave an |
| 59 | # incorrect result. |
| 60 | # Added POD documentation. |
| 61 | # Moved usage instructions into help() subroutine. |
| 62 | # Added 'use strict' and declared all global variables. |
| 63 | # Added '-html' flag and resultant code. |
| 64 | # Added '-cache' flag and resultant code. |
| 65 | # Added add_volume() routine and converted all volume variables |
| 66 | # to use it, fixing the overflow problems for individual hosts |
| 67 | # on large sites. |
| 68 | # Converted all volume output to GB/MB/KB as appropriate. |
| 69 | # Don't store local user stats if -nfl is specified. |
| 70 | # Modifications done by: Steve Campbell (<steve@computurn.com>) |
| 71 | # 2001-04-02 Added the -t_remote_users flag. Steve Campbell. |
| 72 | # 2001-10-15 Added the -domain flag. Steve Campbell. |
| 73 | # 2001-10-16 Accept files on STDIN or on the command line. Steve Campbell. |
| 74 | # 2001-10-21 Removed -domain flag and added -bydomain, -byhost, and -byemail. |
| 75 | # We now generate our main parsing subroutine as an eval statement |
| 76 | # which improves performance dramatically when not all the results |
| 77 | # are required. We also cache the last timestamp to time convertion. |
| 78 | # |
| 79 | # NOTE: 'Top 50 destinations by (message count|volume)' lines are |
| 80 | # now 'Top N (host|email|domain) destinations by (message count|volume)' |
| 81 | # where N is the topcount. Steve Campbell. |
| 82 | # |
| 83 | # 2001-10-30 V1.16 Joachim Wieland. |
| 84 | # Fixed minor bugs in add_volume() when taking over this version |
| 85 | # for use in Exim 4: -w gave uninitialized value warnings in |
| 86 | # two situations: for the first addition to a counter, and if |
| 87 | # there were never any gigabytes, thereby leaving the $gigs |
| 88 | # value unset. |
| 89 | # Initialized $last_timestamp to stop a -w uninitialized warning. |
| 90 | # Minor layout tweak for grand totals (nitpicking). |
| 91 | # Put the IP addresses for relaying stats in [] and separated by |
| 92 | # a space from the domain name. |
| 93 | # Removed the IPv4-specific address test when picking out addresses |
| 94 | # for relaying. Anything inside [] is OK. |
| 95 | # |
| 96 | # 2002-07-02 Philip Hazel |
| 97 | # Fixed "uninitialized variable" message that occurred for relay |
| 98 | # messages that arrived from H=[1.2.3.4] hosts (no name shown). |
| 99 | # This bug didn't affect the output. |
| 100 | # |
| 101 | # 2002-04-15 V1.17 Joachim Wieland. |
| 102 | # Added -charts, -chartdir. -chartrel options which use |
| 103 | # GD::Graph modules to create graphical charts of the statistics. |
| 104 | # |
| 105 | # 2002-04-15 V1.18 Steve Campbell. |
| 106 | # Added a check for $domain to to stop a -w uninitialized warning. |
| 107 | # Added -byemaildomain option. |
| 108 | # Only print HTML header links to included tables! |
| 109 | # |
| 110 | # 2002-08-02 V1.19 Steve Campbell. |
| 111 | # Changed the debug mode to dump the parser onto STDERR rather |
| 112 | # than STDOUT. Documented the -d flag into the help(). |
| 113 | # Rejoined the divergent 2002-04-15 and 2002-07-02 releases. |
| 114 | # |
| 115 | # 2002-08-21 V1.20 Steve Campbell. |
| 116 | # Added the '-merge' option to allow merging of previous reports. |
| 117 | # Fixed a missing semicolon when doing -bydomain. |
| 118 | # Make volume charts plot the data gigs and bytes rather than just bytes. |
| 119 | # Only process log lines with $flag =~ /<=|=>|->|==|\*\*|Co/ |
| 120 | # Converted Emaildomain to Edomain - the column header was too wide! |
| 121 | # This changes the text output slightly. You can revert to the old |
| 122 | # column widths by changing $COLUMN_WIDTHS to 7; |
| 123 | # |
| 124 | # 2002-09-04 V1.21 Andreas J Mueller |
| 125 | # Local deliveries domain now defaults to 'localdomain'. |
| 126 | # Don't match F=<From> when looking for the user. |
| 127 | # |
| 128 | # 2002-09-05 V1.22 Steve Campbell |
| 129 | # Fixed a perl 5.005 incompatibility problem ('our' variables). |
| 130 | # |
| 131 | # 2002-09-11 V1.23 Steve Campbell |
| 132 | # Stopped -charts option from throwing errors on null data. |
| 133 | # Don't print out 'Errors encountered' unless there are any. |
| 134 | |
| 135 | # 2002-10-21 V1.23a Philip Hazel - patch from Tony Finch put in until |
| 136 | # Steve's eximstats catches up. |
| 137 | # Handle log files that include the timezone after the timestamp. |
| 138 | # Switch to assuming that log timestamps are in local time, with |
| 139 | # an option for UTC timestamps, as in Exim itself. |
| 140 | # |
| 141 | # 2003-02-05 V1.24 Steve Campbell |
| 142 | # Added in Sergey Sholokh's code to convert '<' and '>' characters |
| 143 | # in HTML output. Also added code to convert them back with -merge. |
| 144 | # Fixed timestamp offsets to convert to seconds rather than minutes. |
| 145 | # Updated -merge to work with output files using timezones. |
| 146 | # Added cacheing to speed up the calculation of timezone offsets. |
| 147 | # |
| 148 | # 2003-02-07 V1.25 Steve Campbell |
| 149 | # Optimised the usage of mktime() in the seconds subroutine. |
| 150 | # Removed the now redundant '-cache' option. |
| 151 | # html2txt() now explicitly matches HTML tags. |
| 152 | # Implemented a new sorting algorithm - the top_n_sort() routine. |
| 153 | # Added Danny Carroll's '-nvr' flag and code. |
| 154 | # |
| 155 | # 2003-03-13 V1.26 Steve Campbell |
| 156 | # Implemented HTML compliance changes recommended by Bernard Massot. |
| 157 | # Bug fix to allow top_n_sort() to handle null keys. |
| 158 | # Convert all domains and edomains to lowercase. |
| 159 | # Remove preceding dots from domains. |
| 160 | # |
| 161 | # 2003-03-13 V1.27 Steve Campbell |
| 162 | # Replaced border attributes with 'border=1', as recommended by |
| 163 | # Bernard Massot. |
| 164 | # |
| 165 | # 2003-06-03 V1.28 John Newman |
| 166 | # Added in the ability to skip over the parsing and evaulation of |
| 167 | # specific transports as passed to eximstats via the new "-nt/.../" |
| 168 | # command line argument. This new switch allows the viewing of |
| 169 | # not more accurate statistics but more applicable statistics when |
| 170 | # special transports are in use (ie; SpamAssassin). We need to be |
| 171 | # able to ignore transports such as this otherwise the resulting |
| 172 | # local deliveries are significantly skewed (doubled)... |
| 173 | # |
| 174 | # 2003-11-06 V1.29 Steve Campbell |
| 175 | # Added the '-pattern "Description" "/pattern/"' option. |
| 176 | # |
| 177 | # 2004-02-17 V1.30 Steve Campbell |
| 178 | # Added warnings if required GD::Graph modules are not available or |
| 179 | # insufficient -chart* options are specified. |
| 180 | # |
| 181 | # 2004-02-20 V1.31 Andrea Balzi |
| 182 | # Only show the Local Sender/Destination links if the tables exist. |
| 183 | # |
| 184 | # 2004-07-05 V1.32 Steve Campbell |
| 185 | # Fix '-merge -h0' divide by zero error. |
| 186 | # |
| 187 | # 2004-07-15 V1.33 Steve Campbell |
| 188 | # Documentation update - I've converted the subroutine |
| 189 | # documentation from POD to comments. |
| 190 | # |
| 191 | # 2004-12-10 V1.34 Steve Campbell |
| 192 | # Eximstats can now parse syslog lines as well as mainlog lines. |
| 193 | # |
| 194 | # 2004-12-20 V1.35 Wouter Verhelst |
| 195 | # Pie charts by volume were actually generated by count. Fixed. |
| 196 | # |
| 197 | # 2005-02-07 V1.36 Gregor Herrmann / Steve Campbell |
| 198 | # Added average sizes to HTML Top tables. |
| 199 | # |
| 200 | # 2005-04-26 V1.37 Frank Heydlauf |
| 201 | # Added -xls and the ability to specify output files. |
| 202 | # |
| 203 | # 2005-04-29 V1.38 Steve Campbell |
| 204 | # Use FileHandles for outputing results. |
| 205 | # Allow any combination of xls, txt, and html output. |
| 206 | # Fixed display of large numbers with -nvr option |
| 207 | # Fixed merging of reports with empty tables. |
| 208 | # |
| 209 | # 2005-05-27 V1.39 Steve Campbell |
| 210 | # Added the -include_original_destination flag |
| 211 | # Removed tabs and trailing whitespace. |
| 212 | # |
| 213 | # 2005-06-03 V1.40 Steve Campbell |
| 214 | # Whilst parsing the mainlog(s), store information about |
| 215 | # the messages in a hash of arrays rather than using |
| 216 | # individual hashes. This is a bit cleaner and results in |
| 217 | # dramatic memory savings, albeit at a slight CPU cost. |
| 218 | # |
| 219 | # 2005-06-15 V1.41 Steve Campbell |
| 220 | # Added the -show_rt<list> flag. |
| 221 | # Added the -show_dt<list> flag. |
| 222 | # |
| 223 | # 2005-06-24 V1.42 Steve Campbell |
| 224 | # Added Histograms for user specified patterns. |
| 225 | # |
| 226 | # 2005-06-30 V1.43 Steve Campbell |
| 227 | # Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear. |
| 228 | # |
| 229 | # 2005-07-26 V1.44 Steve Campbell |
| 230 | # Use a glob alias rather than an array ref in the generated |
| 231 | # parser. This improves both readability and performance. |
| 232 | # |
| 233 | # 2005-09-30 V1.45 Marco Gaiarin / Steve Campbell |
| 234 | # Collect SpamAssassin and rejection statistics. |
| 235 | # Don't display local sender or destination tables unless |
| 236 | # there is data to show. |
| 237 | # Added average volumes into the top table text output. |
| 238 | # |
| 239 | # 2006-02-07 V1.46 Steve Campbell |
| 240 | # Collect data on the number of addresses (recipients) |
| 241 | # as well as the number of messages. |
| 242 | # |
| 243 | # 2006-05-05 V1.47 Steve Campbell |
| 244 | # Added 'Message too big' to the list of mail rejection |
| 245 | # reasons (thanks to Marco Gaiarin). |
| 246 | # |
| 247 | # 2006-06-05 V1.48 Steve Campbell |
| 248 | # Mainlog lines which have GMT offsets and are too short to |
| 249 | # have a flag are now skipped. |
| 250 | # |
| 251 | # 2006-11-10 V1.49 Alain Williams |
| 252 | # Added the -emptyok flag. |
| 253 | # |
| 254 | # 2006-11-16 V1.50 Steve Campbell |
| 255 | # Fixes for obtaining the IP address from reject messages. |
| 256 | # |
| 257 | # 2006-11-27 V1.51 Steve Campbell |
| 258 | # Another update for obtaining the IP address from reject messages. |
| 259 | # |
| 260 | # 2006-11-27 V1.52 Steve Campbell |
| 261 | # Tally any reject message containing SpamAssassin. |
| 262 | # |
| 263 | # 2007-01-31 V1.53 Philip Hazel |
| 264 | # Allow for [pid] after date in log lines |
| 265 | # |
| 266 | # 2007-02-14 V1.54 Daniel Tiefnig |
| 267 | # Improved the '($parent) =' pattern match. |
| 268 | # |
| 269 | # 2007-03-19 V1.55 Steve Campbell |
| 270 | # Differentiate between permanent and temporary rejects. |
| 271 | # |
| 272 | # 2007-03-29 V1.56 Jez Hancock |
| 273 | # Fixed some broken HTML links and added missing column headers. |
| 274 | # |
| 275 | # 2007-03-30 V1.57 Steve Campbell |
| 276 | # Fixed Grand Total Summary Domains, Edomains, and Email columns |
| 277 | # for Rejects, Temp Rejects, Ham, and Spam rows. |
| 278 | # |
| 279 | # 2007-04-11 V1.58 Steve Campbell |
| 280 | # Fix to get <> and blackhole to show in edomain tables. |
| 281 | # |
| 282 | # 2007-09-20 V1.59 Steve Campbell |
| 283 | # Added the -bylocaldomain option |
| 284 | # |
| 285 | # 2007-09-20 V1.60 Heiko Schlittermann |
| 286 | # Fix for misinterpreted log lines |
| 287 | # |
| 288 | # |
| 289 | # |
| 290 | # For documentation on the logfile format, see |
| 291 | # http://www.exim.org/exim-html-4.50/doc/html/spec_48.html#IX2793 |
| 292 | |
| 293 | =head1 NAME |
| 294 | |
| 295 | eximstats - generates statistics from Exim mainlog or syslog files. |
| 296 | |
| 297 | =head1 SYNOPSIS |
| 298 | |
| 299 | eximstats [Output] [Options] mainlog1 mainlog2 ... |
| 300 | eximstats -merge [Options] report.1.txt report.2.txt ... > weekly_report.txt |
| 301 | |
| 302 | =head2 Output: |
| 303 | |
| 304 | =over 4 |
| 305 | |
| 306 | =item B<-txt> |
| 307 | |
| 308 | Output the results in plain text to STDOUT. |
| 309 | |
| 310 | =item B<-txt>=I<filename> |
| 311 | |
| 312 | Output the results in plain text. Filename '-' for STDOUT is accepted. |
| 313 | |
| 314 | =item B<-html> |
| 315 | |
| 316 | Output the results in HTML to STDOUT. |
| 317 | |
| 318 | =item B<-html>=I<filename> |
| 319 | |
| 320 | Output the results in HTML. Filename '-' for STDOUT is accepted. |
| 321 | |
| 322 | =item B<-xls> |
| 323 | |
| 324 | Output the results in Excel compatible Format to STDOUT. |
| 325 | Requires the Spreadsheet::WriteExcel CPAN module. |
| 326 | |
| 327 | =item B<-xls>=I<filename> |
| 328 | |
| 329 | Output the results in Excel compatible format. Filename '-' for STDOUT is accepted. |
| 330 | |
| 331 | |
| 332 | =back |
| 333 | |
| 334 | =head2 Options: |
| 335 | |
| 336 | =over 4 |
| 337 | |
| 338 | =item B<-h>I<number> |
| 339 | |
| 340 | histogram divisions per hour. The default is 1, and |
| 341 | 0 suppresses histograms. Valid values are: |
| 342 | |
| 343 | 0, 1, 2, 3, 5, 10, 15, 20, 30 or 60. |
| 344 | |
| 345 | =item B<-ne> |
| 346 | |
| 347 | Don't display error information. |
| 348 | |
| 349 | =item B<-nr> |
| 350 | |
| 351 | Don't display relaying information. |
| 352 | |
| 353 | =item B<-nr>I</pattern/> |
| 354 | |
| 355 | Don't display relaying information that matches. |
| 356 | |
| 357 | =item B<-nt> |
| 358 | |
| 359 | Don't display transport information. |
| 360 | |
| 361 | =item B<-nt>I</pattern/> |
| 362 | |
| 363 | Don't display transport information that matches |
| 364 | |
| 365 | =item B<-q>I<list> |
| 366 | |
| 367 | List of times for queuing information single 0 item suppresses. |
| 368 | |
| 369 | =item B<-t>I<number> |
| 370 | |
| 371 | Display top <number> sources/destinations |
| 372 | default is 50, 0 suppresses top listing. |
| 373 | |
| 374 | =item B<-tnl> |
| 375 | |
| 376 | Omit local sources/destinations in top listing. |
| 377 | |
| 378 | =item B<-t_remote_users> |
| 379 | |
| 380 | Include remote users in the top source/destination listings. |
| 381 | |
| 382 | =item B<-include_original_destination> |
| 383 | |
| 384 | Include the original destination email addresses rather than just |
| 385 | using the final ones. |
| 386 | Useful for finding out which of your mailing lists are receiving mail. |
| 387 | |
| 388 | =item B<-show_dt>I<list> |
| 389 | |
| 390 | Show the delivery times (B<DT>)for all the messages. |
| 391 | |
| 392 | Exim must have been configured to use the +delivery_time logging option |
| 393 | for this option to work. |
| 394 | |
| 395 | I<list> is an optional list of times. Eg -show_dt1,2,4,8 will show |
| 396 | the number of messages with delivery times under 1 second, 2 seconds, 4 seconds, |
| 397 | 8 seconds, and over 8 seconds. |
| 398 | |
| 399 | =item B<-show_rt>I<list> |
| 400 | |
| 401 | Show the receipt times for all the messages. The receipt time is |
| 402 | defined as the Completed hh:mm:ss - queue_time_overall - the Receipt hh:mm:ss. |
| 403 | These figures will be skewed by pipelined messages so might not be that useful. |
| 404 | |
| 405 | Exim must have been configured to use the +queue_time_overall logging option |
| 406 | for this option to work. |
| 407 | |
| 408 | I<list> is an optional list of times. Eg -show_rt1,2,4,8 will show |
| 409 | the number of messages with receipt times under 1 second, 2 seconds, 4 seconds, |
| 410 | 8 seconds, and over 8 seconds. |
| 411 | |
| 412 | =item B<-byhost> |
| 413 | |
| 414 | Show results by sending host. This may be combined with |
| 415 | B<-bydomain> and/or B<-byemail> and/or B<-byedomain>. If none of these options |
| 416 | are specified, then B<-byhost> is assumed as a default. |
| 417 | |
| 418 | =item B<-bydomain> |
| 419 | |
| 420 | Show results by sending domain. |
| 421 | May be combined with B<-byhost> and/or B<-byemail> and/or B<-byedomain>. |
| 422 | |
| 423 | =item B<-byemail> |
| 424 | |
| 425 | Show results by sender's email address. |
| 426 | May be combined with B<-byhost> and/or B<-bydomain> and/or B<-byedomain>. |
| 427 | |
| 428 | =item B<-byemaildomain> or B<-byedomain> |
| 429 | |
| 430 | Show results by sender's email domain. |
| 431 | May be combined with B<-byhost> and/or B<-bydomain> and/or B<-byemail>. |
| 432 | |
| 433 | =item B<-pattern> I<Description> I</Pattern/> |
| 434 | |
| 435 | Look for the specified pattern and count the number of lines in which it appears. |
| 436 | This option can be specified multiple times. Eg: |
| 437 | |
| 438 | -pattern 'Refused connections' '/refused connection/' |
| 439 | |
| 440 | |
| 441 | =item B<-merge> |
| 442 | |
| 443 | This option allows eximstats to merge old eximstat reports together. Eg: |
| 444 | |
| 445 | eximstats mainlog.sun > report.sun.txt |
| 446 | eximstats mainlog.mon > report.mon.txt |
| 447 | eximstats mainlog.tue > report.tue.txt |
| 448 | eximstats mainlog.wed > report.web.txt |
| 449 | eximstats mainlog.thu > report.thu.txt |
| 450 | eximstats mainlog.fri > report.fri.txt |
| 451 | eximstats mainlog.sat > report.sat.txt |
| 452 | eximstats -merge report.*.txt > weekly_report.txt |
| 453 | eximstats -merge -html report.*.txt > weekly_report.html |
| 454 | |
| 455 | =over 4 |
| 456 | |
| 457 | =item * |
| 458 | |
| 459 | You can merge text or html reports and output the results as text or html. |
| 460 | |
| 461 | =item * |
| 462 | |
| 463 | You can use all the normal eximstat output options, but only data |
| 464 | included in the original reports can be shown! |
| 465 | |
| 466 | =item * |
| 467 | |
| 468 | When merging reports, some loss of accuracy may occur in the top I<n> lists. |
| 469 | This will be towards the ends of the lists. |
| 470 | |
| 471 | =item * |
| 472 | |
| 473 | The order of items in the top I<n> lists may vary when the data volumes |
| 474 | round to the same value. |
| 475 | |
| 476 | =back |
| 477 | |
| 478 | =item B<-charts> |
| 479 | |
| 480 | Create graphical charts to be displayed in HTML output. |
| 481 | Only valid in combination with I<-html>. |
| 482 | |
| 483 | This requires the following modules which can be obtained |
| 484 | from http://www.cpan.org/modules/01modules.index.html |
| 485 | |
| 486 | =over 4 |
| 487 | |
| 488 | =item GD |
| 489 | |
| 490 | =item GDTextUtil |
| 491 | |
| 492 | =item GDGraph |
| 493 | |
| 494 | =back |
| 495 | |
| 496 | To install these, download and unpack them, then use the normal perl installation procedure: |
| 497 | |
| 498 | perl Makefile.PL |
| 499 | make |
| 500 | make test |
| 501 | make install |
| 502 | |
| 503 | =item B<-chartdir>I <dir> |
| 504 | |
| 505 | Create the charts in the directory <dir> |
| 506 | |
| 507 | =item B<-chartrel>I <dir> |
| 508 | |
| 509 | Specify the relative directory for the "img src=" tags from where to include |
| 510 | the charts |
| 511 | |
| 512 | =item B<-emptyok> |
| 513 | |
| 514 | Specify that it's OK to not find any valid log lines. Without this |
| 515 | we will output an error message if we don't find any. |
| 516 | |
| 517 | =item B<-d> |
| 518 | |
| 519 | Debug flag. This outputs the eval()'d parser onto STDOUT which makes it |
| 520 | easier to trap errors in the eval section. Remember to add 1 to the line numbers to allow for the |
| 521 | title! |
| 522 | |
| 523 | =back |
| 524 | |
| 525 | =head1 DESCRIPTION |
| 526 | |
| 527 | Eximstats parses exim mainlog and syslog files to output a statistical |
| 528 | analysis of the messages processed. By default, a text |
| 529 | analysis is generated, but you can request other output formats |
| 530 | using flags. See the help (B<-help>) to learn |
| 531 | about how to create charts from the tables. |
| 532 | |
| 533 | =head1 AUTHOR |
| 534 | |
| 535 | There is a web site at http://www.exim.org - this contains details of the |
| 536 | mailing list exim-users@exim.org. |
| 537 | |
| 538 | =head1 TO DO |
| 539 | |
| 540 | This program does not perfectly handle messages whose received |
| 541 | and delivered log lines are in different files, which can happen |
| 542 | when you have multiple mail servers and a message cannot be |
| 543 | immeadiately delivered. Fixing this could be tricky... |
| 544 | |
| 545 | Merging of xls files is not (yet) possible. Be free to implement :) |
| 546 | |
| 547 | =cut |
| 548 | |
| 549 | use integer; |
| 550 | use strict; |
| 551 | use IO::File; |
| 552 | |
| 553 | # use Time::Local; # PH/FANF |
| 554 | use POSIX; |
| 555 | |
| 556 | use vars qw($HAVE_GD_Graph_pie $HAVE_GD_Graph_linespoints $HAVE_Spreadsheet_WriteExcel); |
| 557 | eval { require GD::Graph::pie; }; |
| 558 | $HAVE_GD_Graph_pie = $@ ? 0 : 1; |
| 559 | eval { require GD::Graph::linespoints; }; |
| 560 | $HAVE_GD_Graph_linespoints = $@ ? 0 : 1; |
| 561 | eval { require Spreadsheet::WriteExcel; }; |
| 562 | $HAVE_Spreadsheet_WriteExcel = $@ ? 0 : 1; |
| 563 | |
| 564 | |
| 565 | ################################################## |
| 566 | # Static data # |
| 567 | ################################################## |
| 568 | # 'use vars' instead of 'our' as perl5.005 is still in use out there! |
| 569 | use vars qw(@tab62 @days_per_month $gig); |
| 570 | use vars qw($VERSION); |
| 571 | use vars qw($COLUMN_WIDTHS); |
| 572 | use vars qw($WEEK $DAY $HOUR $MINUTE); |
| 573 | |
| 574 | |
| 575 | @tab62 = |
| 576 | (0,1,2,3,4,5,6,7,8,9,0,0,0,0,0,0, # 0-9 |
| 577 | 0,10,11,12,13,14,15,16,17,18,19,20, # A-K |
| 578 | 21,22,23,24,25,26,27,28,29,30,31,32, # L-W |
| 579 | 33,34,35, 0, 0, 0, 0, 0, # X-Z |
| 580 | 0,36,37,38,39,40,41,42,43,44,45,46, # a-k |
| 581 | 47,48,49,50,51,52,53,54,55,56,57,58, # l-w |
| 582 | 59,60,61); # x-z |
| 583 | |
| 584 | @days_per_month = (0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334); |
| 585 | $gig = 1024 * 1024 * 1024; |
| 586 | $VERSION = '1.50'; |
| 587 | |
| 588 | # How much space do we allow for the Hosts/Domains/Emails/Edomains column headers? |
| 589 | $COLUMN_WIDTHS = 8; |
| 590 | |
| 591 | $MINUTE = 60; |
| 592 | $HOUR = 60 * $MINUTE; |
| 593 | $DAY = 24 * $HOUR; |
| 594 | $WEEK = 7 * $DAY; |
| 595 | |
| 596 | # Declare global variables. |
| 597 | use vars qw($total_received_data $total_received_data_gigs $total_received_count); |
| 598 | use vars qw($total_delivered_data $total_delivered_data_gigs $total_delivered_messages $total_delivered_addresses); |
| 599 | use vars qw(%timestamp2time); #Hash of timestamp => time. |
| 600 | use vars qw($last_timestamp $last_time); #The last time convertion done. |
| 601 | use vars qw($last_date $date_seconds); #The last date convertion done. |
| 602 | use vars qw($last_offset $offset_seconds); #The last time offset convertion done. |
| 603 | use vars qw($localtime_offset); |
| 604 | use vars qw($i); #General loop counter. |
| 605 | use vars qw($debug); #Debug mode? |
| 606 | use vars qw($ntopchart); #How many entries should make it into the chart? |
| 607 | use vars qw($gddirectory); #Where to put files from GD::Graph |
| 608 | |
| 609 | # SpamAssassin variables |
| 610 | use vars qw($spam_score $spam_score_gigs); |
| 611 | use vars qw($ham_score $ham_score_gigs); |
| 612 | use vars qw(%ham_count_by_ip %spam_count_by_ip); |
| 613 | use vars qw(%rejected_count_by_ip %rejected_count_by_reason); |
| 614 | use vars qw(%temporarily_rejected_count_by_ip %temporarily_rejected_count_by_reason); |
| 615 | |
| 616 | #For use in Speadsheed::WriteExcel |
| 617 | use vars qw($workbook $ws_global $ws_relayed $ws_errors); |
| 618 | use vars qw($row $col $row_hist $col_hist); |
| 619 | use vars qw($run_hist); |
| 620 | use vars qw($f_default $f_header1 $f_header2 $f_header2_m $f_headertab $f_percent); #Format Header |
| 621 | |
| 622 | # Output FileHandles |
| 623 | use vars qw($txt_fh $htm_fh $xls_fh); |
| 624 | |
| 625 | $ntopchart = 5; |
| 626 | |
| 627 | # The following are parameters whose values are |
| 628 | # set by command line switches: |
| 629 | use vars qw($show_errors $show_relay $show_transport $transport_pattern); |
| 630 | use vars qw($topcount $local_league_table $include_remote_users $do_local_domain); |
| 631 | use vars qw($hist_opt $hist_interval $hist_number $volume_rounding $emptyOK); |
| 632 | use vars qw($relay_pattern @queue_times @user_patterns @user_descriptions); |
| 633 | use vars qw(@rcpt_times @delivery_times); |
| 634 | use vars qw($include_original_destination); |
| 635 | use vars qw($txt_fh $htm_fh $xls_fh); |
| 636 | |
| 637 | use vars qw(%do_sender); #Do sender by Host, Domain, Email, and/or Edomain tables. |
| 638 | use vars qw($charts $chartrel $chartdir $charts_option_specified); |
| 639 | use vars qw($merge_reports); #Merge old reports ? |
| 640 | |
| 641 | # The following are modified in the parse() routine, and |
| 642 | # referred to in the print_*() routines. |
| 643 | use vars qw($delayed_count $relayed_unshown $begin $end); |
| 644 | use vars qw(%messages @message); |
| 645 | use vars qw(%received_count %received_data %received_data_gigs); |
| 646 | use vars qw(%delivered_messages %delivered_data %delivered_data_gigs %delivered_addresses); |
| 647 | use vars qw(%received_count_user %received_data_user %received_data_gigs_user); |
| 648 | use vars qw(%delivered_messages_user %delivered_addresses_user %delivered_data_user %delivered_data_gigs_user); |
| 649 | use vars qw(%delivered_messages_local_domain %delivered_addresses_local_domain %delivered_data_local_domain %delivered_data_gigs_local_domain); |
| 650 | use vars qw(%transported_count %transported_data %transported_data_gigs); |
| 651 | use vars qw(%relayed %errors_count $message_errors); |
| 652 | use vars qw(@qt_all_bin @qt_remote_bin); |
| 653 | use vars qw($qt_all_overflow $qt_remote_overflow); |
| 654 | use vars qw(@dt_all_bin @dt_remote_bin %rcpt_times_bin); |
| 655 | use vars qw($dt_all_overflow $dt_remote_overflow %rcpt_times_overflow); |
| 656 | use vars qw(@received_interval_count @delivered_interval_count); |
| 657 | use vars qw(@user_pattern_totals @user_pattern_interval_count); |
| 658 | |
| 659 | use vars qw(%report_totals); |
| 660 | |
| 661 | # Enumerations |
| 662 | use vars qw($SIZE $FROM_HOST $FROM_ADDRESS $ARRIVAL_TIME $REMOTE_DELIVERED $PROTOCOL); |
| 663 | use vars qw($DELAYED $HAD_ERROR); |
| 664 | $SIZE = 0; |
| 665 | $FROM_HOST = 1; |
| 666 | $FROM_ADDRESS = 2; |
| 667 | $ARRIVAL_TIME = 3; |
| 668 | $REMOTE_DELIVERED = 4; |
| 669 | $DELAYED = 5; |
| 670 | $HAD_ERROR = 6; |
| 671 | $PROTOCOL = 7; |
| 672 | |
| 673 | |
| 674 | |
| 675 | ################################################## |
| 676 | # Subroutines # |
| 677 | ################################################## |
| 678 | |
| 679 | ####################################################################### |
| 680 | # get_filehandle($file,\%output_files); |
| 681 | # Return a filehandle writing to $file. |
| 682 | # |
| 683 | # If %output_files is defined, check that $output_files{$file} |
| 684 | # doesn't exist and die if it does, or set it if it doesn't. |
| 685 | ####################################################################### |
| 686 | sub get_filehandle { |
| 687 | my($file,$output_files_href) = @_; |
| 688 | |
| 689 | $file = '-' if ($file eq ''); |
| 690 | |
| 691 | if (defined $output_files_href) { |
| 692 | die "You can only output to '$file' once! Use -h for help.\n" if exists $output_files_href->{$file}; |
| 693 | $output_files_href->{$file} = 1; |
| 694 | } |
| 695 | |
| 696 | if ($file eq '-') { |
| 697 | return \*STDOUT; |
| 698 | } |
| 699 | |
| 700 | if (-e $file) { |
| 701 | unlink $file or die "Failed to rm $file: $!"; |
| 702 | } |
| 703 | |
| 704 | my $fh = new IO::File $file, O_WRONLY|O_CREAT|O_EXCL; |
| 705 | die "new IO::File $file failed: $!" unless (defined $fh); |
| 706 | return $fh; |
| 707 | } |
| 708 | |
| 709 | |
| 710 | ####################################################################### |
| 711 | # volume_rounded(); |
| 712 | # |
| 713 | # $rounded_volume = volume_rounded($bytes,$gigabytes); |
| 714 | # |
| 715 | # Given a data size in bytes, round it to KB, MB, or GB |
| 716 | # as appropriate. |
| 717 | # |
| 718 | # Eg 12000 => 12KB, 15000000 => 14GB, etc. |
| 719 | # |
| 720 | # Note: I've experimented with Math::BigInt and it results in a 33% |
| 721 | # performance degredation as opposed to storing numbers split into |
| 722 | # bytes and gigabytes. |
| 723 | ####################################################################### |
| 724 | sub volume_rounded { |
| 725 | my($x,$g) = @_; |
| 726 | $x = 0 unless $x; |
| 727 | $g = 0 unless $g; |
| 728 | my($rounded); |
| 729 | |
| 730 | while ($x > $gig) { |
| 731 | $g++; |
| 732 | $x -= $gig; |
| 733 | } |
| 734 | |
| 735 | if ($volume_rounding) { |
| 736 | # Values < 1 GB |
| 737 | if ($g <= 0) { |
| 738 | if ($x < 10000) { |
| 739 | $rounded = sprintf("%6d", $x); |
| 740 | } |
| 741 | elsif ($x < 10000000) { |
| 742 | $rounded = sprintf("%4dKB", ($x + 512)/1024); |
| 743 | } |
| 744 | else { |
| 745 | $rounded = sprintf("%4dMB", ($x + 512*1024)/(1024*1024)); |
| 746 | } |
| 747 | } |
| 748 | # Values between 1GB and 10GB are printed in MB |
| 749 | elsif ($g < 10) { |
| 750 | $rounded = sprintf("%4dMB", ($g * 1024) + ($x + 512*1024)/(1024*1024)); |
| 751 | } |
| 752 | else { |
| 753 | # Handle values over 10GB |
| 754 | $rounded = sprintf("%4dGB", $g + ($x + $gig/2)/$gig); |
| 755 | } |
| 756 | } |
| 757 | else { |
| 758 | # We don't want any rounding to be done. |
| 759 | # and we don't need broken formated output which on one hand avoids numbers from |
| 760 | # being interpreted as string by Spreadsheed Calculators, on the other hand |
| 761 | # breaks if more than 4 digits! -> flexible length instead of fixed length |
| 762 | # Format the return value at the output routine! -fh |
| 763 | #$rounded = sprintf("%d", ($g * $gig) + $x); |
| 764 | no integer; |
| 765 | $rounded = sprintf("%.0f", ($g * $gig) + $x); |
| 766 | } |
| 767 | |
| 768 | return $rounded; |
| 769 | } |
| 770 | |
| 771 | |
| 772 | ####################################################################### |
| 773 | # un_round(); |
| 774 | # |
| 775 | # un_round($rounded_volume,\$bytes,\$gigabytes); |
| 776 | # |
| 777 | # Given a volume in KB, MB or GB, as generated by volume_rounded(), |
| 778 | # do the reverse transformation and convert it back into Bytes and Gigabytes. |
| 779 | # These are added to the $bytes and $gigabytes parameters. |
| 780 | # |
| 781 | # Given a data size in bytes, round it to KB, MB, or GB |
| 782 | # as appropriate. |
| 783 | # |
| 784 | # EG: 500 => (500,0), 14GB => (0,14), etc. |
| 785 | ####################################################################### |
| 786 | sub un_round { |
| 787 | my($rounded,$bytes_sref,$gigabytes_sref) = @_; |
| 788 | |
| 789 | if ($rounded =~ /(\d+)GB/) { |
| 790 | $$gigabytes_sref += $1; |
| 791 | } |
| 792 | elsif ($rounded =~ /(\d+)MB/) { |
| 793 | $$gigabytes_sref += $1 / 1024; |
| 794 | $$bytes_sref += (($1 % 1024 ) * 1024 * 1024); |
| 795 | } |
| 796 | elsif ($rounded =~ /(\d+)KB/) { |
| 797 | $$gigabytes_sref += $1 / (1024 * 1024); |
| 798 | $$bytes_sref += ($1 % (1024 * 1024) * 1024); |
| 799 | } |
| 800 | elsif ($rounded =~ /(\d+)/) { |
| 801 | # We need to turn off integer in case we are merging an -nvr report. |
| 802 | no integer; |
| 803 | $$gigabytes_sref += int($1 / $gig); |
| 804 | $$bytes_sref += $1 % $gig; |
| 805 | } |
| 806 | |
| 807 | #Now reduce the bytes down to less than 1GB. |
| 808 | add_volume($bytes_sref,$gigabytes_sref,0) if ($$bytes_sref > $gig); |
| 809 | } |
| 810 | |
| 811 | |
| 812 | ####################################################################### |
| 813 | # add_volume(); |
| 814 | # |
| 815 | # add_volume(\$bytes,\$gigs,$size); |
| 816 | # |
| 817 | # Add $size to $bytes/$gigs where this is a number split into |
| 818 | # bytes ($bytes) and gigabytes ($gigs). This is significantly |
| 819 | # faster than using Math::BigInt. |
| 820 | ####################################################################### |
| 821 | sub add_volume { |
| 822 | my($bytes_ref,$gigs_ref,$size) = @_; |
| 823 | $$bytes_ref = 0 if ! defined $$bytes_ref; |
| 824 | $$gigs_ref = 0 if ! defined $$gigs_ref; |
| 825 | $$bytes_ref += $size; |
| 826 | while ($$bytes_ref > $gig) { |
| 827 | $$gigs_ref++; |
| 828 | $$bytes_ref -= $gig; |
| 829 | } |
| 830 | } |
| 831 | |
| 832 | |
| 833 | ####################################################################### |
| 834 | # format_time(); |
| 835 | # |
| 836 | # $formatted_time = format_time($seconds); |
| 837 | # |
| 838 | # Given a time in seconds, break it down into |
| 839 | # weeks, days, hours, minutes, and seconds. |
| 840 | # |
| 841 | # Eg 12005 => 3h20m5s |
| 842 | ####################################################################### |
| 843 | sub format_time { |
| 844 | my($t) = pop @_; |
| 845 | my($s) = $t % 60; |
| 846 | $t /= 60; |
| 847 | my($m) = $t % 60; |
| 848 | $t /= 60; |
| 849 | my($h) = $t % 24; |
| 850 | $t /= 24; |
| 851 | my($d) = $t % 7; |
| 852 | my($w) = $t/7; |
| 853 | my($p) = ""; |
| 854 | $p .= "$w"."w" if $w > 0; |
| 855 | $p .= "$d"."d" if $d > 0; |
| 856 | $p .= "$h"."h" if $h > 0; |
| 857 | $p .= "$m"."m" if $m > 0; |
| 858 | $p .= "$s"."s" if $s > 0 || $p eq ""; |
| 859 | $p; |
| 860 | } |
| 861 | |
| 862 | |
| 863 | ####################################################################### |
| 864 | # unformat_time(); |
| 865 | # |
| 866 | # $seconds = unformat_time($formatted_time); |
| 867 | # |
| 868 | # Given a time in weeks, days, hours, minutes, or seconds, convert it to seconds. |
| 869 | # |
| 870 | # Eg 3h20m5s => 12005 |
| 871 | ####################################################################### |
| 872 | sub unformat_time { |
| 873 | my($formated_time) = pop @_; |
| 874 | my $time = 0; |
| 875 | |
| 876 | while ($formated_time =~ s/^(\d+)([wdhms]?)//) { |
| 877 | $time += $1 if ($2 eq '' || $2 eq 's'); |
| 878 | $time += $1 * 60 if ($2 eq 'm'); |
| 879 | $time += $1 * 60 * 60 if ($2 eq 'h'); |
| 880 | $time += $1 * 60 * 60 * 24 if ($2 eq 'd'); |
| 881 | $time += $1 * 60 * 60 * 24 * 7 if ($2 eq 'w'); |
| 882 | } |
| 883 | $time; |
| 884 | } |
| 885 | |
| 886 | |
| 887 | ####################################################################### |
| 888 | # seconds(); |
| 889 | # |
| 890 | # $time = seconds($timestamp); |
| 891 | # |
| 892 | # Given a time-of-day timestamp, convert it into a time() value using |
| 893 | # POSIX::mktime. We expect the timestamp to be of the form |
| 894 | # "$year-$mon-$day $hour:$min:$sec", with month going from 1 to 12, |
| 895 | # and the year to be absolute (we do the necessary conversions). The |
| 896 | # timestamp may be followed with an offset from UTC like "+$hh$mm"; if the |
| 897 | # offset is not present, and we have not been told that the log is in UTC |
| 898 | # (with the -utc option), then we adjust the time by the current local |
| 899 | # time offset so that it can be compared with the time recorded in message |
| 900 | # IDs, which is UTC. |
| 901 | # |
| 902 | # To improve performance, we only use mktime on the date ($year-$mon-$day), |
| 903 | # and only calculate it if the date is different to the previous time we |
| 904 | # came here. We then add on seconds for the '$hour:$min:$sec'. |
| 905 | # |
| 906 | # We also store the results of the last conversion done, and only |
| 907 | # recalculate if the date is different. |
| 908 | # |
| 909 | # We used to have the '-cache' flag which would store the results of the |
| 910 | # mktime() call. However, the current way of just using mktime() on the |
| 911 | # date obsoletes this. |
| 912 | ####################################################################### |
| 913 | sub seconds { |
| 914 | my($timestamp) = @_; |
| 915 | |
| 916 | # Is the timestamp the same as the last one? |
| 917 | return $last_time if ($last_timestamp eq $timestamp); |
| 918 | |
| 919 | return 0 unless ($timestamp =~ /^((\d{4})\-(\d\d)-(\d\d))\s(\d\d):(\d\d):(\d\d)( ([+-])(\d\d)(\d\d))?/o); |
| 920 | |
| 921 | unless ($last_date eq $1) { |
| 922 | $last_date = $1; |
| 923 | my(@timestamp) = (0,0,0,$4,$3,$2); |
| 924 | $timestamp[5] -= 1900; |
| 925 | $timestamp[4]--; |
| 926 | $date_seconds = mktime(@timestamp); |
| 927 | } |
| 928 | my $time = $date_seconds + ($5 * 3600) + ($6 * 60) + $7; |
| 929 | |
| 930 | # SC. Use cacheing. Also note we want seconds not minutes. |
| 931 | #my($this_offset) = ($10 * 60 + $11) * ($9 . "1") if defined $8; |
| 932 | if (defined $8 && ($8 ne $last_offset)) { |
| 933 | $last_offset = $8; |
| 934 | $offset_seconds = ($10 * 60 + $11) * 60; |
| 935 | $offset_seconds = -$offset_seconds if ($9 eq '-'); |
| 936 | } |
| 937 | |
| 938 | |
| 939 | if (defined $7) { |
| 940 | #$time -= $this_offset; |
| 941 | $time -= $offset_seconds; |
| 942 | } elsif (defined $localtime_offset) { |
| 943 | $time -= $localtime_offset; |
| 944 | } |
| 945 | |
| 946 | # Store the last timestamp received. |
| 947 | $last_timestamp = $timestamp; |
| 948 | $last_time = $time; |
| 949 | |
| 950 | $time; |
| 951 | } |
| 952 | |
| 953 | |
| 954 | ####################################################################### |
| 955 | # id_seconds(); |
| 956 | # |
| 957 | # $time = id_seconds($message_id); |
| 958 | # |
| 959 | # Given a message ID, convert it into a time() value. |
| 960 | ####################################################################### |
| 961 | sub id_seconds { |
| 962 | my($sub_id) = substr((pop @_), 0, 6); |
| 963 | my($s) = 0; |
| 964 | my(@c) = split(//, $sub_id); |
| 965 | while($#c >= 0) { $s = $s * 62 + $tab62[ord(shift @c) - ord('0')] } |
| 966 | $s; |
| 967 | } |
| 968 | |
| 969 | ####################################################################### |
| 970 | # wdhms_seconds(); |
| 971 | # |
| 972 | # $seconds = wdhms_seconds($string); |
| 973 | # |
| 974 | # Convert a string in a week/day/hour/minute/second format (eg 4h10s) |
| 975 | # into seconds. |
| 976 | ####################################################################### |
| 977 | sub wdhms_seconds { |
| 978 | if ($_[0] =~ /^(?:(\d+)w)?(?:(\d+)d)?(?:(\d+)h)?(?:(\d+)m)?(?:(\d+)s)?/) { |
| 979 | return((($1||0) * $WEEK) + (($2||0) * $DAY) + (($3||0) * $HOUR) + (($4||0) * $MINUTE) + ($5||0)); |
| 980 | } |
| 981 | return undef; |
| 982 | } |
| 983 | |
| 984 | ####################################################################### |
| 985 | # queue_time(); |
| 986 | # |
| 987 | # $queued = queue_time($completed_tod, $arrival_time, $id); |
| 988 | # |
| 989 | # Given the completed time of day and either the arrival time |
| 990 | # (preferred), or the message ID, calculate how long the message has |
| 991 | # been on the queue. |
| 992 | # |
| 993 | ####################################################################### |
| 994 | sub queue_time { |
| 995 | my($completed_tod, $arrival_time, $id) = @_; |
| 996 | |
| 997 | # Note: id_seconds() benchmarks as 42% slower than seconds() |
| 998 | # and computing the time accounts for a significant portion of |
| 999 | # the run time. |
| 1000 | if (defined $arrival_time) { |
| 1001 | return(seconds($completed_tod) - seconds($arrival_time)); |
| 1002 | } |
| 1003 | else { |
| 1004 | return(seconds($completed_tod) - id_seconds($id)); |
| 1005 | } |
| 1006 | } |
| 1007 | |
| 1008 | |
| 1009 | ####################################################################### |
| 1010 | # calculate_localtime_offset(); |
| 1011 | # |
| 1012 | # $localtime_offset = calculate_localtime_offset(); |
| 1013 | # |
| 1014 | # Calculate the the localtime offset from gmtime in seconds. |
| 1015 | # |
| 1016 | # $localtime = time() + $localtime_offset. |
| 1017 | # |
| 1018 | # These are the same semantics as ISO 8601 and RFC 2822 timezone offsets. |
| 1019 | # (West is negative, East is positive.) |
| 1020 | ####################################################################### |
| 1021 | |
| 1022 | # $localtime = gmtime() + $localtime_offset. OLD COMMENT |
| 1023 | # This subroutine commented out as it's not currently in use. |
| 1024 | |
| 1025 | #sub calculate_localtime_offset { |
| 1026 | # # Pick an arbitrary date, convert it to localtime & gmtime, and return the difference. |
| 1027 | # my (@sample_date) = (0,0,0,5,5,100); |
| 1028 | # my $localtime = timelocal(@sample_date); |
| 1029 | # my $gmtime = timegm(@sample_date); |
| 1030 | # my $offset = $localtime - $gmtime; |
| 1031 | # return $offset; |
| 1032 | #} |
| 1033 | |
| 1034 | sub calculate_localtime_offset { |
| 1035 | # Assume that the offset at the moment is valid across the whole |
| 1036 | # period covered by the logs that we're analysing. This may not |
| 1037 | # be true around the time the clocks change in spring or autumn. |
| 1038 | my $utc = time; |
| 1039 | # mktime works on local time and gmtime works in UTC |
| 1040 | my $local = mktime(gmtime($utc)); |
| 1041 | return $local - $utc; |
| 1042 | } |
| 1043 | |
| 1044 | |
| 1045 | |
| 1046 | ####################################################################### |
| 1047 | # print_duration_table(); |
| 1048 | # |
| 1049 | # print_duration_table($title, $message_type, \@times, \@values, $overflow); |
| 1050 | # |
| 1051 | # Print a table showing how long a particular step took for |
| 1052 | # the messages. The parameters are: |
| 1053 | # $title Eg "Time spent on the queue" |
| 1054 | # $message_type Eg "Remote" |
| 1055 | # \@times The maximum time a message took for it to increment |
| 1056 | # the corresponding @values counter. |
| 1057 | # \@values An array of message counters. |
| 1058 | # $overflow The number of messages which exceeded the maximum |
| 1059 | # time. |
| 1060 | ####################################################################### |
| 1061 | sub print_duration_table { |
| 1062 | no integer; |
| 1063 | my($title, $message_type, $times_aref, $values_aref, $overflow) = @_; |
| 1064 | my(@chartdatanames); |
| 1065 | my(@chartdatavals); |
| 1066 | |
| 1067 | my $printed_one = 0; |
| 1068 | my $cumulative_percent = 0; |
| 1069 | |
| 1070 | my $queue_total = $overflow; |
| 1071 | map {$queue_total += $_} @$values_aref; |
| 1072 | |
| 1073 | my $temp = "$title: $message_type"; |
| 1074 | |
| 1075 | |
| 1076 | my $txt_format = "%5s %4s %6d %5.1f%% %5.1f%%\n"; |
| 1077 | 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"; |
| 1078 | |
| 1079 | # write header |
| 1080 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; |
| 1081 | if ($htm_fh) { |
| 1082 | print $htm_fh "<hr><a name=\"$title $message_type\"></a><h2>$temp</h2>\n"; |
| 1083 | print $htm_fh "<table border=0 width=\"100%\"><tr><td><table border=1>\n"; |
| 1084 | print $htm_fh "<tr><th>Time</th><th>Messages</th><th>Percentage</th><th>Cumulative Percentage</th>\n"; |
| 1085 | } |
| 1086 | if ($xls_fh) { |
| 1087 | $ws_global->write($row++, $col, "$title: ".$message_type, $f_header2); |
| 1088 | my @content=("Time", "Messages", "Percentage", "Cumulative Percentage"); |
| 1089 | &set_worksheet_line($ws_global, $row++, 1, \@content, $f_headertab); |
| 1090 | } |
| 1091 | |
| 1092 | |
| 1093 | for ($i = 0; $i <= $#$times_aref; ++$i) { |
| 1094 | if ($$values_aref[$i] > 0) |
| 1095 | { |
| 1096 | my $percent = ($values_aref->[$i] * 100)/$queue_total; |
| 1097 | $cumulative_percent += $percent; |
| 1098 | |
| 1099 | my @content=($printed_one? " " : "Under", |
| 1100 | format_time($times_aref->[$i]), |
| 1101 | $values_aref->[$i], $percent, $cumulative_percent); |
| 1102 | |
| 1103 | if ($htm_fh) { |
| 1104 | printf $htm_fh ($htm_format, @content); |
| 1105 | if (!defined($values_aref->[$i])) { |
| 1106 | print $htm_fh "Not defined"; |
| 1107 | } |
| 1108 | } |
| 1109 | if ($txt_fh) { |
| 1110 | printf $txt_fh ($txt_format, @content); |
| 1111 | if (!defined($times_aref->[$i])) { |
| 1112 | print $txt_fh "Not defined"; |
| 1113 | } |
| 1114 | } |
| 1115 | if ($xls_fh) |
| 1116 | { |
| 1117 | no integer; |
| 1118 | &set_worksheet_line($ws_global, $row, 0, [@content[0,1,2]], $f_default); |
| 1119 | &set_worksheet_line($ws_global, $row++, 3, [$content[3]/100,$content[4]/100], $f_percent); |
| 1120 | |
| 1121 | if (!defined($times_aref->[$i])) { |
| 1122 | $col=0; |
| 1123 | $ws_global->write($row++, $col, "Not defined" ); |
| 1124 | } |
| 1125 | } |
| 1126 | |
| 1127 | push(@chartdatanames, |
| 1128 | ($printed_one? "" : "Under") . format_time($times_aref->[$i])); |
| 1129 | push(@chartdatavals, $$values_aref[$i]); |
| 1130 | $printed_one = 1; |
| 1131 | } |
| 1132 | } |
| 1133 | |
| 1134 | if ($overflow && $overflow > 0) { |
| 1135 | my $percent = ($overflow * 100)/$queue_total; |
| 1136 | $cumulative_percent += $percent; |
| 1137 | |
| 1138 | my @content = ("Over ", format_time($times_aref->[-1]), |
| 1139 | $overflow, $percent, $cumulative_percent); |
| 1140 | |
| 1141 | printf $txt_fh ($txt_format, @content) if $txt_fh; |
| 1142 | printf $htm_fh ($htm_format, @content) if $htm_fh; |
| 1143 | if ($xls_fh) |
| 1144 | { |
| 1145 | &set_worksheet_line($ws_global, $row, 0, [@content[0,1,2]], $f_default); |
| 1146 | &set_worksheet_line($ws_global, $row++, 3, [$content[3]/100,$content[4]/100], $f_percent); |
| 1147 | } |
| 1148 | |
| 1149 | } |
| 1150 | |
| 1151 | push(@chartdatanames, "Over " . format_time($times_aref->[-1])); |
| 1152 | push(@chartdatavals, $overflow); |
| 1153 | |
| 1154 | #printf("Unknown %6d\n", $queue_unknown) if $queue_unknown > 0; |
| 1155 | if ($htm_fh) { |
| 1156 | print $htm_fh "</table></td><td>"; |
| 1157 | |
| 1158 | if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) { |
| 1159 | my @data = ( |
| 1160 | \@chartdatanames, |
| 1161 | \@chartdatavals |
| 1162 | ); |
| 1163 | my $graph = GD::Graph::pie->new(200, 200); |
| 1164 | my $pngname = "$title-$message_type.png"; |
| 1165 | $pngname =~ s/[^\w\-\.]/_/; |
| 1166 | |
| 1167 | my $graph_title = "$title ($message_type)"; |
| 1168 | $graph->set(title => $graph_title) if (length($graph_title) < 21); |
| 1169 | |
| 1170 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 1171 | if ($gd) { |
| 1172 | open(IMG, ">$chartdir/$pngname") or die "Could not write $chartdir/$pngname: $!\n"; |
| 1173 | binmode IMG; |
| 1174 | print IMG $gd->png; |
| 1175 | close IMG; |
| 1176 | print $htm_fh "<img src=\"$chartrel/$pngname\">"; |
| 1177 | } |
| 1178 | } |
| 1179 | print $htm_fh "</td></tr></table>\n"; |
| 1180 | } |
| 1181 | |
| 1182 | if ($xls_fh) |
| 1183 | { |
| 1184 | $row++; |
| 1185 | } |
| 1186 | print $txt_fh "\n" if $txt_fh; |
| 1187 | print $htm_fh "\n" if $htm_fh; |
| 1188 | |
| 1189 | } |
| 1190 | |
| 1191 | |
| 1192 | ####################################################################### |
| 1193 | # print_histogram(); |
| 1194 | # |
| 1195 | # print_histogram('Deliveries|Messages received|$pattern', $unit, @interval_count); |
| 1196 | # |
| 1197 | # Print a histogram of the messages delivered/received per time slot |
| 1198 | # (hour by default). |
| 1199 | ####################################################################### |
| 1200 | sub print_histogram { |
| 1201 | my($text, $unit, @interval_count) = @_; |
| 1202 | my(@chartdatanames); |
| 1203 | my(@chartdatavals); |
| 1204 | my($maxd) = 0; |
| 1205 | |
| 1206 | # save first row of print_histogram for xls output |
| 1207 | if (!$run_hist) { |
| 1208 | $row_hist = $row; |
| 1209 | } |
| 1210 | else { |
| 1211 | $row = $row_hist; |
| 1212 | } |
| 1213 | |
| 1214 | for ($i = 0; $i < $hist_number; $i++) |
| 1215 | { $maxd = $interval_count[$i] if $interval_count[$i] > $maxd; } |
| 1216 | |
| 1217 | my $scale = int(($maxd + 25)/50); |
| 1218 | $scale = 1 if $scale == 0; |
| 1219 | |
| 1220 | if ($scale != 1) { |
| 1221 | if ($unit !~ s/y$/ies/) { |
| 1222 | $unit .= 's'; |
| 1223 | } |
| 1224 | } |
| 1225 | |
| 1226 | # make and output title |
| 1227 | my $title = sprintf("$text per %s", |
| 1228 | ($hist_interval == 60)? "hour" : |
| 1229 | ($hist_interval == 1)? "minute" : "$hist_interval minutes"); |
| 1230 | |
| 1231 | my $txt_htm_title = $title . " (each dot is $scale $unit)"; |
| 1232 | |
| 1233 | printf $txt_fh ("%s\n%s\n\n", $txt_htm_title, "-" x length($txt_htm_title)) if $txt_fh; |
| 1234 | |
| 1235 | if ($htm_fh) { |
| 1236 | print $htm_fh "<hr><a name=\"$text\"></a><h2>$txt_htm_title</h2>\n"; |
| 1237 | print $htm_fh "<table border=0 width=\"100%\">\n"; |
| 1238 | print $htm_fh "<tr><td><pre>\n"; |
| 1239 | } |
| 1240 | |
| 1241 | if ($xls_fh) { |
| 1242 | $title =~ s/Messages/Msg/ ; |
| 1243 | $row += 2; |
| 1244 | $ws_global->write($row++, $col_hist+1, $title, $f_headertab); |
| 1245 | } |
| 1246 | |
| 1247 | |
| 1248 | my $hour = 0; |
| 1249 | my $minutes = 0; |
| 1250 | for ($i = 0; $i < $hist_number; $i++) { |
| 1251 | my $c = $interval_count[$i]; |
| 1252 | |
| 1253 | # If the interval is an hour (the maximum) print the starting and |
| 1254 | # ending hours as a label. Otherwise print the starting hour and |
| 1255 | # minutes, which take up the same space. |
| 1256 | |
| 1257 | my $temp; |
| 1258 | if ($hist_opt == 1) { |
| 1259 | $temp = sprintf("%02d-%02d", $hour, $hour + 1); |
| 1260 | |
| 1261 | print $txt_fh $temp if $txt_fh; |
| 1262 | print $htm_fh $temp if $htm_fh; |
| 1263 | |
| 1264 | if ($xls_fh) { |
| 1265 | if ($run_hist==0) { |
| 1266 | # only on first run |
| 1267 | $ws_global->write($row, 0, [$temp], $f_default); |
| 1268 | } |
| 1269 | } |
| 1270 | |
| 1271 | push(@chartdatanames, $temp); |
| 1272 | $hour++; |
| 1273 | } |
| 1274 | else { |
| 1275 | if ($minutes == 0) |
| 1276 | { $temp = sprintf("%02d:%02d", $hour, $minutes) } |
| 1277 | else |
| 1278 | { $temp = sprintf(" :%02d", $minutes) } |
| 1279 | |
| 1280 | print $txt_fh $temp if $txt_fh; |
| 1281 | print $htm_fh $temp if $htm_fh; |
| 1282 | if (($xls_fh) and ($run_hist==0)) { |
| 1283 | # only on first run |
| 1284 | $temp = sprintf("%02d:%02d", $hour, $minutes); |
| 1285 | $ws_global->write($row, 0, [$temp], $f_default); |
| 1286 | } |
| 1287 | |
| 1288 | push(@chartdatanames, $temp); |
| 1289 | $minutes += $hist_interval; |
| 1290 | if ($minutes >= 60) { |
| 1291 | $minutes = 0; |
| 1292 | $hour++; |
| 1293 | } |
| 1294 | } |
| 1295 | push(@chartdatavals, $c); |
| 1296 | |
| 1297 | printf $txt_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $txt_fh; |
| 1298 | printf $htm_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $htm_fh; |
| 1299 | $ws_global->write($row++, $col_hist+1, [$c], $f_default) if $xls_fh; |
| 1300 | |
| 1301 | } #end for |
| 1302 | |
| 1303 | printf $txt_fh "\n" if $txt_fh; |
| 1304 | printf $htm_fh "\n" if $htm_fh; |
| 1305 | |
| 1306 | if ($htm_fh) |
| 1307 | { |
| 1308 | print $htm_fh "</pre>\n"; |
| 1309 | print $htm_fh "</td><td>\n"; |
| 1310 | if ($HAVE_GD_Graph_linespoints && $charts && ($#chartdatavals > 0)) { |
| 1311 | # calculate the graph |
| 1312 | my @data = ( |
| 1313 | \@chartdatanames, |
| 1314 | \@chartdatavals |
| 1315 | ); |
| 1316 | my $graph = GD::Graph::linespoints->new(300, 300); |
| 1317 | $graph->set( |
| 1318 | x_label => 'Time', |
| 1319 | y_label => 'Amount', |
| 1320 | title => $text, |
| 1321 | x_labels_vertical => 1 |
| 1322 | ); |
| 1323 | my $pngname = "histogram_$text.png"; |
| 1324 | $pngname =~ s/[^\w\._]/_/g; |
| 1325 | |
| 1326 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 1327 | if ($gd) { |
| 1328 | open(IMG, ">$chartdir/$pngname") or die "Could not write $chartdir/$pngname: $!\n"; |
| 1329 | binmode IMG; |
| 1330 | print IMG $gd->png; |
| 1331 | close IMG; |
| 1332 | print $htm_fh "<img src=\"$chartrel/$pngname\">"; |
| 1333 | } |
| 1334 | } |
| 1335 | print $htm_fh "</td></tr></table>\n"; |
| 1336 | } |
| 1337 | |
| 1338 | $col_hist++; # where to continue next times |
| 1339 | |
| 1340 | $row+=2; # leave some space after history block |
| 1341 | $run_hist=1; # we have done this once or more |
| 1342 | } |
| 1343 | |
| 1344 | |
| 1345 | |
| 1346 | ####################################################################### |
| 1347 | # print_league_table(); |
| 1348 | # |
| 1349 | # print_league_table($league_table_type,\%message_count,\%address_count,\%message_data,\%message_data_gigs, $spreadsheet, $row_sref); |
| 1350 | # |
| 1351 | # Given hashes of message count, address count, and message data, |
| 1352 | # which are keyed by the table type (eg by the sending host), print a |
| 1353 | # league table showing the top $topcount (defaults to 50). |
| 1354 | ####################################################################### |
| 1355 | sub print_league_table { |
| 1356 | my($text,$m_count,$a_count,$m_data,$m_data_gigs,$spreadsheet, $row_sref) = @_; |
| 1357 | my($name) = ($topcount == 1)? "$text" : "$topcount ${text}s"; |
| 1358 | my($title) = "Top $name by message count"; |
| 1359 | my(@chartdatanames) = (); |
| 1360 | my(@chartdatavals) = (); |
| 1361 | my $chartotherval = 0; |
| 1362 | $text = ucfirst($text); |
| 1363 | |
| 1364 | # Align non-local addresses to the right (so all the .com's line up). |
| 1365 | # Local addresses are aligned on the left as they are userids. |
| 1366 | my $align = ($text !~ /local/i) ? 'right' : 'left'; |
| 1367 | |
| 1368 | |
| 1369 | ################################################ |
| 1370 | # Generate the printf formats and table headers. |
| 1371 | ################################################ |
| 1372 | my(@headers) = ('Messages'); |
| 1373 | #push(@headers,'Addresses') if defined $a_count; |
| 1374 | push(@headers,'Addresses') if defined $a_count && %$a_count; |
| 1375 | push(@headers,'Bytes','Average') if defined $m_data; |
| 1376 | |
| 1377 | my $txt_format = "%10s " x @headers . " %s\n"; |
| 1378 | my $txt_col_headers = sprintf $txt_format, @headers, $text; |
| 1379 | my $htm_format = "<tr>" . '<td align="right">%s</td>'x@headers . "<td align=\"$align\" nowrap>%s</td></tr>\n"; |
| 1380 | my $htm_col_headers = sprintf $htm_format, @headers, $text; |
| 1381 | $htm_col_headers =~ s/(<\/?)td/$1th/g; #Convert <td>'s to <th>'s for the header. |
| 1382 | |
| 1383 | |
| 1384 | ################################################ |
| 1385 | # Write the table headers |
| 1386 | ################################################ |
| 1387 | printf $txt_fh ("%s\n%s\n%s", $title, "-" x length($title),$txt_col_headers) if $txt_fh; |
| 1388 | |
| 1389 | if ($htm_fh) { |
| 1390 | print $htm_fh <<EoText; |
| 1391 | <hr><a name="$text count"></a><h2>$title</h2> |
| 1392 | <table border=0 width="100%"> |
| 1393 | <tr><td> |
| 1394 | <table border=1> |
| 1395 | EoText |
| 1396 | print $htm_fh $htm_col_headers |
| 1397 | } |
| 1398 | |
| 1399 | if ($xls_fh) { |
| 1400 | $spreadsheet->write(${$row_sref}++, 0, $title, $f_header2); |
| 1401 | $spreadsheet->write(${$row_sref}++, 0, [@headers, $text], $f_headertab); |
| 1402 | } |
| 1403 | |
| 1404 | |
| 1405 | # write content |
| 1406 | foreach my $key (top_n_sort($topcount,$m_count,$m_data_gigs,$m_data)) { |
| 1407 | |
| 1408 | # When displaying the average figures, we calculate the average of |
| 1409 | # the rounded data, as the user would calculate it. This reduces |
| 1410 | # the accuracy slightly, but we have to do it this way otherwise |
| 1411 | # when using -merge to convert results from text to HTML and |
| 1412 | # vice-versa discrepencies would occur. |
| 1413 | my $messages = $$m_count{$key}; |
| 1414 | my @content = ($messages); |
| 1415 | push(@content, $$a_count{$key}) if defined $a_count; |
| 1416 | if (defined $m_data) { |
| 1417 | my $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); |
| 1418 | my($data,$gigs) = (0,0); |
| 1419 | un_round($rounded_volume,\$data,\$gigs); |
| 1420 | my $rounded_average = volume_rounded($data/$messages,$gigs/$messages); |
| 1421 | push(@content, $rounded_volume, $rounded_average); |
| 1422 | } |
| 1423 | |
| 1424 | # write content |
| 1425 | printf $txt_fh ($txt_format, @content, $key) if $txt_fh; |
| 1426 | |
| 1427 | if ($htm_fh) { |
| 1428 | my $htmlkey = $key; |
| 1429 | $htmlkey =~ s/>/\>\;/g; |
| 1430 | $htmlkey =~ s/</\<\;/g; |
| 1431 | printf $htm_fh ($htm_format, @content, $htmlkey); |
| 1432 | } |
| 1433 | $spreadsheet->write(${$row_sref}++, 0, [@content, $key], $f_default) if $xls_fh; |
| 1434 | |
| 1435 | if (scalar @chartdatanames < $ntopchart) { |
| 1436 | push(@chartdatanames, $key); |
| 1437 | push(@chartdatavals, $$m_count{$key}); |
| 1438 | } |
| 1439 | else { |
| 1440 | $chartotherval += $$m_count{$key}; |
| 1441 | } |
| 1442 | } |
| 1443 | |
| 1444 | push(@chartdatanames, "Other"); |
| 1445 | push(@chartdatavals, $chartotherval); |
| 1446 | |
| 1447 | print $txt_fh "\n" if $txt_fh; |
| 1448 | if ($htm_fh) { |
| 1449 | print $htm_fh "</table>\n"; |
| 1450 | print $htm_fh "</td><td>\n"; |
| 1451 | if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) |
| 1452 | { |
| 1453 | # calculate the graph |
| 1454 | my @data = ( |
| 1455 | \@chartdatanames, |
| 1456 | \@chartdatavals |
| 1457 | ); |
| 1458 | my $graph = GD::Graph::pie->new(300, 300); |
| 1459 | $graph->set( |
| 1460 | x_label => 'Name', |
| 1461 | y_label => 'Amount', |
| 1462 | title => 'By count', |
| 1463 | ); |
| 1464 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 1465 | if ($gd) { |
| 1466 | my $temp = $text; |
| 1467 | $temp =~ s/ /_/g; |
| 1468 | open(IMG, ">$chartdir/${temp}_count.png") or die "Could not write $chartdir/${temp}_count.png: $!\n"; |
| 1469 | binmode IMG; |
| 1470 | print IMG $gd->png; |
| 1471 | close IMG; |
| 1472 | print $htm_fh "<img src=\"$chartrel/${temp}_count.png\">"; |
| 1473 | } |
| 1474 | } |
| 1475 | print $htm_fh "</td><td>\n"; |
| 1476 | print $htm_fh "</td></tr></table>\n\n"; |
| 1477 | } |
| 1478 | ++${$row_sref} if $xls_fh; |
| 1479 | |
| 1480 | |
| 1481 | if (defined $m_data) { |
| 1482 | # write header |
| 1483 | |
| 1484 | $title = "Top $name by volume"; |
| 1485 | |
| 1486 | printf $txt_fh ("%s\n%s\n%s", $title, "-" x length($title),$txt_col_headers) if $txt_fh; |
| 1487 | |
| 1488 | if ($htm_fh) { |
| 1489 | print $htm_fh <<EoText; |
| 1490 | <hr><a name="$text volume"></a><h2>$title</h2> |
| 1491 | <table border=0 width="100%"> |
| 1492 | <tr><td> |
| 1493 | <table border=1> |
| 1494 | EoText |
| 1495 | print $htm_fh $htm_col_headers; |
| 1496 | } |
| 1497 | if ($xls_fh) { |
| 1498 | $spreadsheet->write(${$row_sref}++, 0, $title, $f_header2); |
| 1499 | $spreadsheet->write(${$row_sref}++, 0, [@headers, $text], $f_headertab); |
| 1500 | } |
| 1501 | |
| 1502 | @chartdatanames = (); |
| 1503 | @chartdatavals = (); |
| 1504 | $chartotherval = 0; |
| 1505 | my $use_gig = 0; |
| 1506 | foreach my $key (top_n_sort($topcount,$m_data_gigs,$m_data,$m_count)) { |
| 1507 | # The largest volume will be the first (top of the list). |
| 1508 | # If it has at least 1 gig, then just use gigabytes to avoid |
| 1509 | # risking an integer overflow when generating the pie charts. |
| 1510 | if ($$m_data_gigs{$key}) { |
| 1511 | $use_gig = 1; |
| 1512 | } |
| 1513 | |
| 1514 | my $messages = $$m_count{$key}; |
| 1515 | my @content = ($messages); |
| 1516 | push(@content, $$a_count{$key}) if defined $a_count; |
| 1517 | my $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); |
| 1518 | my($data ,$gigs) = (0,0); |
| 1519 | un_round($rounded_volume,\$data,\$gigs); |
| 1520 | my $rounded_average = volume_rounded($data/$messages,$gigs/$messages); |
| 1521 | push(@content, $rounded_volume, $rounded_average ); |
| 1522 | |
| 1523 | # write content |
| 1524 | printf $txt_fh ($txt_format, @content, $key) if $txt_fh; |
| 1525 | if ($htm_fh) { |
| 1526 | my $htmlkey = $key; |
| 1527 | $htmlkey =~ s/>/\>\;/g; |
| 1528 | $htmlkey =~ s/</\<\;/g; |
| 1529 | printf $htm_fh ($htm_format, @content, $htmlkey); |
| 1530 | } |
| 1531 | $spreadsheet->write(${$row_sref}++, 0, [@content, $key], $f_default) if $xls_fh; |
| 1532 | |
| 1533 | |
| 1534 | if (scalar @chartdatanames < $ntopchart) { |
| 1535 | if ($use_gig) { |
| 1536 | if ($$m_data_gigs{$key}) { |
| 1537 | push(@chartdatanames, $key); |
| 1538 | push(@chartdatavals, $$m_data_gigs{$key}); |
| 1539 | } |
| 1540 | } |
| 1541 | else { |
| 1542 | push(@chartdatanames, $key); |
| 1543 | push(@chartdatavals, $$m_data{$key}); |
| 1544 | } |
| 1545 | } |
| 1546 | else { |
| 1547 | $chartotherval += ($use_gig) ? $$m_data_gigs{$key} : $$m_data{$key}; |
| 1548 | } |
| 1549 | } |
| 1550 | push(@chartdatanames, "Other"); |
| 1551 | push(@chartdatavals, $chartotherval); |
| 1552 | |
| 1553 | print $txt_fh "\n" if $txt_fh; |
| 1554 | if ($htm_fh) { |
| 1555 | print $htm_fh "</table>\n"; |
| 1556 | print $htm_fh "</td><td>\n"; |
| 1557 | if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals > 0)) { |
| 1558 | # calculate the graph |
| 1559 | my @data = ( |
| 1560 | \@chartdatanames, |
| 1561 | \@chartdatavals |
| 1562 | ); |
| 1563 | my $graph = GD::Graph::pie->new(300, 300); |
| 1564 | $graph->set( |
| 1565 | x_label => 'Name', |
| 1566 | y_label => 'Volume' , |
| 1567 | title => 'By Volume', |
| 1568 | ); |
| 1569 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 1570 | if ($gd) { |
| 1571 | my $temp = $text; |
| 1572 | $temp =~ s/ /_/g; |
| 1573 | open(IMG, ">$chartdir/${temp}_volume.png") or die "Could not write $chartdir/${temp}_volume.png: $!\n"; |
| 1574 | binmode IMG; |
| 1575 | print IMG $gd->png; |
| 1576 | close IMG; |
| 1577 | print $htm_fh "<img src=\"$chartrel/${temp}_volume.png\">"; |
| 1578 | } |
| 1579 | } |
| 1580 | print $htm_fh "</td><td>\n"; |
| 1581 | print $htm_fh "</td></tr></table>\n\n"; |
| 1582 | } |
| 1583 | |
| 1584 | ++${$row_sref} if $xls_fh; |
| 1585 | } |
| 1586 | } |
| 1587 | |
| 1588 | |
| 1589 | ####################################################################### |
| 1590 | # top_n_sort(); |
| 1591 | # |
| 1592 | # @sorted_keys = top_n_sort($n,$href1,$href2,$href3); |
| 1593 | # |
| 1594 | # Given a hash which has numerical values, return the sorted $n keys which |
| 1595 | # point to the top values. The second and third hashes are used as |
| 1596 | # tiebreakers. They all must have the same keys. |
| 1597 | # |
| 1598 | # The idea behind this routine is that when you only want to see the |
| 1599 | # top n members of a set, rather than sorting the entire set and then |
| 1600 | # plucking off the top n, sort through the stack as you go, discarding |
| 1601 | # any member which is lower than your current n'th highest member. |
| 1602 | # |
| 1603 | # This proves to be an order of magnitude faster for large hashes. |
| 1604 | # On 200,000 lines of mainlog it benchmarked 9 times faster. |
| 1605 | # On 700,000 lines of mainlog it benchmarked 13.8 times faster. |
| 1606 | # |
| 1607 | # We assume the values are > 0. |
| 1608 | ####################################################################### |
| 1609 | sub top_n_sort { |
| 1610 | my($n,$href1,$href2,$href3) = @_; |
| 1611 | |
| 1612 | # PH's original sort was: |
| 1613 | # |
| 1614 | # foreach $key (sort |
| 1615 | # { |
| 1616 | # $$m_count{$b} <=> $$m_count{$a} || |
| 1617 | # $$m_data_gigs{$b} <=> $$m_data_gigs{$a} || |
| 1618 | # $$m_data{$b} <=> $$m_data{$a} || |
| 1619 | # $a cmp $b |
| 1620 | # } |
| 1621 | # keys %{$m_count}) |
| 1622 | # |
| 1623 | |
| 1624 | #We use a key of '_' to represent non-existant values, as null keys are valid. |
| 1625 | #'_' is not a valid domain, edomain, host, or email. |
| 1626 | my(@top_n_keys) = ('_') x $n; |
| 1627 | my($minimum_value1,$minimum_value2,$minimum_value3) = (0,0,0); |
| 1628 | my $top_n_key = ''; |
| 1629 | my $n_minus_1 = $n - 1; |
| 1630 | my $n_minus_2 = $n - 2; |
| 1631 | |
| 1632 | # Create a dummy hash incase the user has not provided us with |
| 1633 | # tiebreaker hashes. |
| 1634 | my(%dummy_hash); |
| 1635 | $href2 = \%dummy_hash unless defined $href2; |
| 1636 | $href3 = \%dummy_hash unless defined $href3; |
| 1637 | |
| 1638 | # Pick out the top $n keys. |
| 1639 | my($key,$value1,$value2,$value3,$i,$comparison,$insert_position); |
| 1640 | while (($key,$value1) = each %$href1) { |
| 1641 | |
| 1642 | #print STDERR "key $key ($value1,",$href2->{$key},",",$href3->{$key},") <=> ($minimum_value1,$minimum_value2,$minimum_value3)\n"; |
| 1643 | |
| 1644 | # Check to see that the new value is bigger than the lowest of the |
| 1645 | # top n keys that we're keeping. We test the main key first, because |
| 1646 | # for the majority of cases we can skip creating dummy hash values |
| 1647 | # should the user have not provided real tie-breaking hashes. |
| 1648 | next unless $value1 >= $minimum_value1; |
| 1649 | |
| 1650 | # Create a dummy hash entry for the key if required. |
| 1651 | # Note that setting the dummy_hash value sets it for both href2 & |
| 1652 | # href3. Also note that currently we are guarenteed to have a real |
| 1653 | # value for href3 if a real value for href2 exists so don't need to |
| 1654 | # test for it as well. |
| 1655 | $dummy_hash{$key} = 0 unless exists $href2->{$key}; |
| 1656 | |
| 1657 | $comparison = $value1 <=> $minimum_value1 || |
| 1658 | $href2->{$key} <=> $minimum_value2 || |
| 1659 | $href3->{$key} <=> $minimum_value3 || |
| 1660 | $top_n_key cmp $key; |
| 1661 | next unless ($comparison == 1); |
| 1662 | |
| 1663 | # As we will be using these values a few times, extract them into scalars. |
| 1664 | $value2 = $href2->{$key}; |
| 1665 | $value3 = $href3->{$key}; |
| 1666 | |
| 1667 | # This key is bigger than the bottom n key, so the lowest position we |
| 1668 | # will insert it into is $n minus 1 (the bottom of the list). |
| 1669 | $insert_position = $n_minus_1; |
| 1670 | |
| 1671 | # Now go through the list, stopping when we find a key that we're |
| 1672 | # bigger than, or we come to the penultimate position - we've |
| 1673 | # already tested bigger than the last. |
| 1674 | # |
| 1675 | # Note: we go top down as the list starts off empty. |
| 1676 | # Note: stepping through the list in this way benchmarks nearly |
| 1677 | # three times faster than doing a sort() on the reduced list. |
| 1678 | # I assume this is because the list is already in order, and |
| 1679 | # we get a performance boost from not having to do hash lookups |
| 1680 | # on the new key. |
| 1681 | for ($i = 0; $i < $n_minus_1; $i++) { |
| 1682 | $top_n_key = $top_n_keys[$i]; |
| 1683 | if ( ($top_n_key eq '_') || |
| 1684 | ( ($value1 <=> $href1->{$top_n_key} || |
| 1685 | $value2 <=> $href2->{$top_n_key} || |
| 1686 | $value3 <=> $href3->{$top_n_key} || |
| 1687 | $top_n_key cmp $key) == 1 |
| 1688 | ) |
| 1689 | ) { |
| 1690 | $insert_position = $i; |
| 1691 | last; |
| 1692 | } |
| 1693 | } |
| 1694 | |
| 1695 | # Remove the last element, then insert the new one. |
| 1696 | $#top_n_keys = $n_minus_2; |
| 1697 | splice(@top_n_keys,$insert_position,0,$key); |
| 1698 | |
| 1699 | # Extract our new minimum values. |
| 1700 | $top_n_key = $top_n_keys[$n_minus_1]; |
| 1701 | if ($top_n_key ne '_') { |
| 1702 | $minimum_value1 = $href1->{$top_n_key}; |
| 1703 | $minimum_value2 = $href2->{$top_n_key}; |
| 1704 | $minimum_value3 = $href3->{$top_n_key}; |
| 1705 | } |
| 1706 | } |
| 1707 | |
| 1708 | # Return the top n list, grepping out non-existant values, just in case |
| 1709 | # we didn't have that many values. |
| 1710 | return(grep(!/^_$/,@top_n_keys)); |
| 1711 | } |
| 1712 | |
| 1713 | |
| 1714 | |
| 1715 | ####################################################################### |
| 1716 | # html_header(); |
| 1717 | # |
| 1718 | # $header = html_header($title); |
| 1719 | # |
| 1720 | # Print our HTML header and start the <body> block. |
| 1721 | ####################################################################### |
| 1722 | sub html_header { |
| 1723 | my($title) = @_; |
| 1724 | my $text = << "EoText"; |
| 1725 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"> |
| 1726 | <html> |
| 1727 | <head> |
| 1728 | <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-15"> |
| 1729 | <title>$title</title> |
| 1730 | </head> |
| 1731 | <body bgcolor="white"> |
| 1732 | <h1>$title</h1> |
| 1733 | EoText |
| 1734 | return $text; |
| 1735 | } |
| 1736 | |
| 1737 | |
| 1738 | |
| 1739 | ####################################################################### |
| 1740 | # help(); |
| 1741 | # |
| 1742 | # help(); |
| 1743 | # |
| 1744 | # Display usage instructions and exit. |
| 1745 | ####################################################################### |
| 1746 | sub help { |
| 1747 | print << "EoText"; |
| 1748 | |
| 1749 | eximstats Version $VERSION |
| 1750 | |
| 1751 | Usage: |
| 1752 | eximstats [Output] [Options] mainlog1 mainlog2 ... |
| 1753 | eximstats -merge -html [Options] report.1.html ... > weekly_rep.html |
| 1754 | |
| 1755 | Examples: |
| 1756 | eximstats -html=eximstats.html mainlog1 mainlog2 ... |
| 1757 | eximstats mainlog1 mainlog2 ... > report.txt |
| 1758 | |
| 1759 | Parses exim mainlog or syslog files and generates a statistical analysis |
| 1760 | of the messages processed. |
| 1761 | |
| 1762 | Valid output types are: |
| 1763 | -txt[=<file>] plain text (default unless no other type is specified) |
| 1764 | -html[=<file>] HTML |
| 1765 | -xls[=<file>] Excel |
| 1766 | With no type and file given, defaults to -txt and STDOUT. |
| 1767 | |
| 1768 | Valid options are: |
| 1769 | -h<number> histogram divisions per hour. The default is 1, and |
| 1770 | 0 suppresses histograms. Other valid values are: |
| 1771 | 2, 3, 5, 10, 15, 20, 30 or 60. |
| 1772 | -ne don't display error information |
| 1773 | -nr don't display relaying information |
| 1774 | -nr/pattern/ don't display relaying information that matches |
| 1775 | -nt don't display transport information |
| 1776 | -nt/pattern/ don't display transport information that matches |
| 1777 | -nvr don't do volume rounding. Display in bytes, not KB/MB/GB. |
| 1778 | -t<number> display top <number> sources/destinations |
| 1779 | default is 50, 0 suppresses top listing |
| 1780 | -tnl omit local sources/destinations in top listing |
| 1781 | -t_remote_users show top user sources/destinations from non-local domains |
| 1782 | -q<list> list of times for queuing information. -q0 suppresses. |
| 1783 | -show_rt<list> Show the receipt times for all the messages. |
| 1784 | -show_dt<list> Show the delivery times for all the messages. |
| 1785 | <list> is an optional list of times in seconds. |
| 1786 | Eg -show_rt1,2,4,8. |
| 1787 | |
| 1788 | -include_original_destination show both the final and original |
| 1789 | destinations in the results rather than just the final ones. |
| 1790 | |
| 1791 | -byhost show results by sending host (default unless bydomain or |
| 1792 | byemail is specified) |
| 1793 | -bydomain show results by sending domain. |
| 1794 | -byemail show results by sender's email address |
| 1795 | -byedomain show results by sender's email domain |
| 1796 | -bylocaldomain show results by local domain |
| 1797 | |
| 1798 | -pattern "Description" /pattern/ |
| 1799 | Count lines matching specified patterns and show them in |
| 1800 | the results. It can be specified multiple times. Eg: |
| 1801 | -pattern 'Refused connections' '/refused connection/' |
| 1802 | |
| 1803 | -merge merge previously generated reports into a new report |
| 1804 | |
| 1805 | -charts Create charts (this requires the GD::Graph modules). |
| 1806 | Only valid with -html. |
| 1807 | -chartdir <dir> Create the charts' png files in the directory <dir> |
| 1808 | -chartrel <dir> Specify the relative directory for the "img src=" tags |
| 1809 | from where to include the charts in the html file |
| 1810 | -chartdir and -chartrel default to '.' |
| 1811 | |
| 1812 | -emptyok It is OK if there is no valid input, don't print an error. |
| 1813 | |
| 1814 | -d Debug mode - dump the eval'ed parser onto STDERR. |
| 1815 | |
| 1816 | EoText |
| 1817 | |
| 1818 | exit 1; |
| 1819 | } |
| 1820 | |
| 1821 | |
| 1822 | |
| 1823 | ####################################################################### |
| 1824 | # generate_parser(); |
| 1825 | # |
| 1826 | # $parser = generate_parser(); |
| 1827 | # |
| 1828 | # This subroutine generates the parsing routine which will be |
| 1829 | # used to parse the mainlog. We take the base operation, and remove bits not in use. |
| 1830 | # This improves performance depending on what bits you take out or add. |
| 1831 | # |
| 1832 | # I've tested using study(), but this does not improve performance. |
| 1833 | # |
| 1834 | # We store our parsing routing in a variable, and process it looking for #IFDEF (Expression) |
| 1835 | # or #IFNDEF (Expression) statements and corresponding #ENDIF (Expression) statements. If |
| 1836 | # the expression evaluates to true, then it is included/excluded accordingly. |
| 1837 | ####################################################################### |
| 1838 | sub generate_parser { |
| 1839 | my $parser = ' |
| 1840 | my($ip,$host,$email,$edomain,$domain,$thissize,$size,$old,$new); |
| 1841 | my($tod,$m_hour,$m_min,$id,$flag,$extra,$length); |
| 1842 | my($seconds,$queued,$rcpt_time,$local_domain); |
| 1843 | my $rej_id = 0; |
| 1844 | while (<$fh>) { |
| 1845 | |
| 1846 | # Convert syslog lines to mainlog format. |
| 1847 | if (! /^\\d{4}/) { |
| 1848 | next unless s/^.*? exim\\b.*?: //; |
| 1849 | } |
| 1850 | |
| 1851 | $length = length($_); |
| 1852 | next if ($length < 38); |
| 1853 | next unless /^(\\d{4}\\-\\d\\d-\\d\\d\\s(\\d\\d):(\\d\\d):\\d\\d( [-+]\\d\\d\\d\\d)?)( \\[\\d+\\])?/o; |
| 1854 | |
| 1855 | ($tod,$m_hour,$m_min) = ($1,$2,$3); |
| 1856 | |
| 1857 | # PH - watch for GMT offsets in the timestamp. |
| 1858 | if (defined($4)) { |
| 1859 | $extra = 6; |
| 1860 | next if ($length < 44); |
| 1861 | } |
| 1862 | else { |
| 1863 | $extra = 0; |
| 1864 | } |
| 1865 | |
| 1866 | # PH - watch for PID added after the timestamp. |
| 1867 | if (defined($5)) { |
| 1868 | $extra += length($5); |
| 1869 | next if ($length < 38 + $extra); |
| 1870 | } |
| 1871 | |
| 1872 | $id = substr($_, 20 + $extra, 16); |
| 1873 | $flag = substr($_, 37 + $extra, 2); |
| 1874 | |
| 1875 | if ($flag !~ /^([<>=*-]+|SA)$/ && /rejected|refused|dropped/) { |
| 1876 | $flag = "Re"; |
| 1877 | $extra -= 3; |
| 1878 | } |
| 1879 | |
| 1880 | # Rejects can have no MSGID... |
| 1881 | if ($flag eq "Re" && $id !~ /^[-0-9a-zA-Z]+$/) { |
| 1882 | $id = "reject:" . ++$rej_id; |
| 1883 | $extra -= 17; |
| 1884 | } |
| 1885 | '; |
| 1886 | |
| 1887 | # Watch for user specified patterns. |
| 1888 | my $user_pattern_index = 0; |
| 1889 | foreach (@user_patterns) { |
| 1890 | $user_pattern_totals[$user_pattern_index] = 0; |
| 1891 | $parser .= " if ($_) {\n"; |
| 1892 | $parser .= " \$user_pattern_totals[$user_pattern_index]++;\n"; |
| 1893 | $parser .= " \$user_pattern_interval_count[$user_pattern_index][(\$m_hour*60 + \$m_min)/$hist_interval]++;\n" if ($hist_opt > 0); |
| 1894 | $parser .= " }\n"; |
| 1895 | $user_pattern_index++; |
| 1896 | } |
| 1897 | |
| 1898 | $parser .= ' |
| 1899 | next unless ($flag =~ /<=|=>|->|==|\\*\\*|Co|SA|Re/); |
| 1900 | |
| 1901 | #Strip away the timestamp, ID and flag to speed up later pattern matches. |
| 1902 | #The flags include Co (Completed), Re (Rejected), and SA (SpamAssassin). |
| 1903 | $_ = substr($_, 40 + $extra); # PH |
| 1904 | |
| 1905 | # Alias @message to the array of information about the message. |
| 1906 | # This minimises the number of calls to hash functions. |
| 1907 | $messages{$id} = [] unless exists $messages{$id}; |
| 1908 | *message = $messages{$id}; |
| 1909 | |
| 1910 | |
| 1911 | # JN - Skip over certain transports as specified via the "-nt/.../" command |
| 1912 | # line switch (where ... is a perl style regular expression). This is |
| 1913 | # required so that transports that skew stats such as SpamAssassin can be |
| 1914 | # ignored. |
| 1915 | #IFDEF ($transport_pattern) |
| 1916 | if (/\\sT=(\\S+)/) { |
| 1917 | next if ($1 =~ /$transport_pattern/o) ; |
| 1918 | } |
| 1919 | #ENDIF ($transport_pattern) |
| 1920 | |
| 1921 | |
| 1922 | |
| 1923 | # Do some pattern matches to get the host and IP address. |
| 1924 | # We expect lines to be of the form "H=[IpAddr]" or "H=Host [IpAddr]" or |
| 1925 | # "H=Host (UnverifiedHost) [IpAddr]" or "H=(UnverifiedHost) [IpAddr]". |
| 1926 | # We do 2 separate matches to keep the matches simple and fast. |
| 1927 | # Host is local unless otherwise specified. |
| 1928 | $ip = (/\\bH=.*?(\\[[^]]+\\])/) ? $1 |
| 1929 | # 2008-03-31 06:25:22 Connection from [213.246.33.217]:39456 refused: too many connections from that IP address // .hs |
| 1930 | : (/Connection from (\[\S+\])/) ? $1 |
| 1931 | # 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 |
| 1932 | : (/SMTP call from .*?(\[\S+\])/) ? $1 |
| 1933 | : "local"; |
| 1934 | $host = (/\\bH=(\\S+)/) ? $1 : "local"; |
| 1935 | |
| 1936 | $domain = "localdomain"; #Domain is localdomain unless otherwise specified. |
| 1937 | |
| 1938 | #IFDEF ($do_sender{Domain}) |
| 1939 | if ($host =~ /^\\[/ || $host =~ /^[\\d\\.]+$/) { |
| 1940 | # Host is just an IP address. |
| 1941 | $domain = $host; |
| 1942 | } |
| 1943 | elsif ($host =~ /^(\\(?)[^\\.]+\\.([^\\.]+\\..*)/) { |
| 1944 | # Remove the host portion from the DNS name. We ensure that we end up |
| 1945 | # with at least xxx.yyy. $host can be "(x.y.z)" or "x.y.z". |
| 1946 | $domain = lc("$1.$2"); |
| 1947 | $domain =~ s/^\\.//; #Remove preceding dot. |
| 1948 | } |
| 1949 | #ENDIF ($do_sender{Domain}) |
| 1950 | |
| 1951 | #IFDEF ($do_sender{Email}) |
| 1952 | #IFDEF ($include_original_destination) |
| 1953 | # Catch both "a@b.com <c@d.com>" and "e@f.com" |
| 1954 | #$email = (/^(\S+) (<(\S*?)>)?/) ? $3 || $1 : ""; |
| 1955 | $email = (/^(\S+ (<[^@>]+@?[^>]*>)?)/) ? $1 : ""; |
| 1956 | chomp($email); |
| 1957 | #ENDIF ($include_original_destination) |
| 1958 | |
| 1959 | #IFNDEF ($include_original_destination) |
| 1960 | $email = (/^(\S+)/) ? $1 : ""; |
| 1961 | #ENDIF ($include_original_destination) |
| 1962 | #ENDIF ($do_sender{Email}) |
| 1963 | |
| 1964 | #IFDEF ($do_sender{Edomain}) |
| 1965 | if (/^(<>|blackhole)/) { |
| 1966 | $edomain = $1; |
| 1967 | } |
| 1968 | #IFDEF ($include_original_destination) |
| 1969 | elsif (/^(\S+ (<\S*?\\@(\S+?)>)?)/) { |
| 1970 | $edomain = $1; |
| 1971 | chomp($edomain); |
| 1972 | $edomain =~ s/@(\S+?)>/"@" . lc($1) . ">"/e; |
| 1973 | } |
| 1974 | #ENDIF ($include_original_destination) |
| 1975 | #IFNDEF ($include_original_destination) |
| 1976 | elsif (/^\S*?\\@(\S+)/) { |
| 1977 | $edomain = lc($1); |
| 1978 | } |
| 1979 | #ENDIF ($include_original_destination) |
| 1980 | else { |
| 1981 | $edomain = ""; |
| 1982 | } |
| 1983 | |
| 1984 | #ENDIF ($do_sender{Edomain}) |
| 1985 | |
| 1986 | if ($tod lt $begin) { |
| 1987 | $begin = $tod; |
| 1988 | } |
| 1989 | elsif ($tod gt $end) { |
| 1990 | $end = $tod; |
| 1991 | } |
| 1992 | |
| 1993 | |
| 1994 | if ($flag eq "<=") { |
| 1995 | $thissize = (/\\sS=(\\d+)( |$)/) ? $1 : 0; |
| 1996 | $message[$SIZE] = $thissize; |
| 1997 | $message[$PROTOCOL] = (/ P=(\S+)/) ? $1 : undef; |
| 1998 | |
| 1999 | #IFDEF ($show_relay) |
| 2000 | if ($host ne "local") { |
| 2001 | # Save incoming information in case it becomes interesting |
| 2002 | # later, when delivery lines are read. |
| 2003 | my($from) = /^(\\S+)/; |
| 2004 | $message[$FROM_HOST] = "$host$ip"; |
| 2005 | $message[$FROM_ADDRESS] = $from; |
| 2006 | } |
| 2007 | #ENDIF ($show_relay) |
| 2008 | |
| 2009 | #IFDEF ($local_league_table || $include_remote_users) |
| 2010 | if (/\sU=(\\S+)/) { |
| 2011 | my $user = $1; |
| 2012 | |
| 2013 | #IFDEF ($local_league_table && $include_remote_users) |
| 2014 | { #Store both local and remote users. |
| 2015 | #ENDIF ($local_league_table && $include_remote_users) |
| 2016 | |
| 2017 | #IFDEF ($local_league_table && ! $include_remote_users) |
| 2018 | if ($host eq "local") { #Store local users only. |
| 2019 | #ENDIF ($local_league_table && ! $include_remote_users) |
| 2020 | |
| 2021 | #IFDEF ($include_remote_users && ! $local_league_table) |
| 2022 | if ($host ne "local") { #Store remote users only. |
| 2023 | #ENDIF ($include_remote_users && ! $local_league_table) |
| 2024 | |
| 2025 | ++$received_count_user{$user}; |
| 2026 | add_volume(\\$received_data_user{$user},\\$received_data_gigs_user{$user},$thissize); |
| 2027 | } |
| 2028 | } |
| 2029 | #ENDIF ($local_league_table || $include_remote_users) |
| 2030 | |
| 2031 | #IFDEF ($do_sender{Host}) |
| 2032 | ++$received_count{Host}{$host}; |
| 2033 | add_volume(\\$received_data{Host}{$host},\\$received_data_gigs{Host}{$host},$thissize); |
| 2034 | #ENDIF ($do_sender{Host}) |
| 2035 | |
| 2036 | #IFDEF ($do_sender{Domain}) |
| 2037 | if ($domain) { |
| 2038 | ++$received_count{Domain}{$domain}; |
| 2039 | add_volume(\\$received_data{Domain}{$domain},\\$received_data_gigs{Domain}{$domain},$thissize); |
| 2040 | } |
| 2041 | #ENDIF ($do_sender{Domain}) |
| 2042 | |
| 2043 | #IFDEF ($do_sender{Email}) |
| 2044 | ++$received_count{Email}{$email}; |
| 2045 | add_volume(\\$received_data{Email}{$email},\\$received_data_gigs{Email}{$email},$thissize); |
| 2046 | #ENDIF ($do_sender{Email}) |
| 2047 | |
| 2048 | #IFDEF ($do_sender{Edomain}) |
| 2049 | ++$received_count{Edomain}{$edomain}; |
| 2050 | add_volume(\\$received_data{Edomain}{$edomain},\\$received_data_gigs{Edomain}{$edomain},$thissize); |
| 2051 | #ENDIF ($do_sender{Edomain}) |
| 2052 | |
| 2053 | ++$total_received_count; |
| 2054 | add_volume(\\$total_received_data,\\$total_received_data_gigs,$thissize); |
| 2055 | |
| 2056 | #IFDEF ($#queue_times >= 0 || $#rcpt_times >= 0) |
| 2057 | $message[$ARRIVAL_TIME] = $tod; |
| 2058 | #ENDIF ($#queue_times >= 0 || $#rcpt_times >= 0) |
| 2059 | |
| 2060 | #IFDEF ($hist_opt > 0) |
| 2061 | $received_interval_count[($m_hour*60 + $m_min)/$hist_interval]++; |
| 2062 | #ENDIF ($hist_opt > 0) |
| 2063 | } |
| 2064 | |
| 2065 | elsif ($flag eq "=>") { |
| 2066 | $size = $message[$SIZE] || 0; |
| 2067 | if ($host ne "local") { |
| 2068 | $message[$REMOTE_DELIVERED] = 1; |
| 2069 | |
| 2070 | |
| 2071 | #IFDEF ($show_relay) |
| 2072 | # Determine relaying address if either only one address listed, |
| 2073 | # or two the same. If they are different, it implies a forwarding |
| 2074 | # or aliasing, which is not relaying. Note that for multi-aliased |
| 2075 | # addresses, there may be a further address between the first |
| 2076 | # and last. |
| 2077 | |
| 2078 | if (defined $message[$FROM_HOST]) { |
| 2079 | if (/^(\\S+)(?:\\s+\\([^)]\\))?\\s+<([^>]+)>/) { |
| 2080 | ($old,$new) = ($1,$2); |
| 2081 | } |
| 2082 | else { |
| 2083 | $old = $new = ""; |
| 2084 | } |
| 2085 | |
| 2086 | if ("\\L$new" eq "\\L$old") { |
| 2087 | ($old) = /^(\\S+)/ if $old eq ""; |
| 2088 | my $key = "H=\\L$message[$FROM_HOST]\\E A=\\L$message[$FROM_ADDRESS]\\E => " . |
| 2089 | "H=\\L$host\\E$ip A=\\L$old\\E"; |
| 2090 | if (!defined $relay_pattern || $key !~ /$relay_pattern/o) { |
| 2091 | $relayed{$key} = 0 if !defined $relayed{$key}; |
| 2092 | ++$relayed{$key}; |
| 2093 | } |
| 2094 | else { |
| 2095 | ++$relayed_unshown; |
| 2096 | } |
| 2097 | } |
| 2098 | } |
| 2099 | #ENDIF ($show_relay) |
| 2100 | |
| 2101 | } |
| 2102 | |
| 2103 | #IFDEF ($local_league_table || $include_remote_users) |
| 2104 | #IFDEF ($local_league_table && $include_remote_users) |
| 2105 | { #Store both local and remote users. |
| 2106 | #ENDIF ($local_league_table && $include_remote_users) |
| 2107 | |
| 2108 | #IFDEF ($local_league_table && ! $include_remote_users) |
| 2109 | if ($host eq "local") { #Store local users only. |
| 2110 | #ENDIF ($local_league_table && ! $include_remote_users) |
| 2111 | |
| 2112 | #IFDEF ($include_remote_users && ! $local_league_table) |
| 2113 | if ($host ne "local") { #Store remote users only. |
| 2114 | #ENDIF ($include_remote_users && ! $local_league_table) |
| 2115 | |
| 2116 | if (my($user) = split((/\\s</)? " <" : " ", $_)) { |
| 2117 | #IFDEF ($include_original_destination) |
| 2118 | { |
| 2119 | #ENDIF ($include_original_destination) |
| 2120 | #IFNDEF ($include_original_destination) |
| 2121 | if ($user =~ /^[\\/|]/) { |
| 2122 | #ENDIF ($include_original_destination) |
| 2123 | #my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/; |
| 2124 | my($parent) = $_ =~ / (<.+?>) /; #DT 1.54 |
| 2125 | if (defined $parent) { |
| 2126 | $user = "$user $parent"; |
| 2127 | #IFDEF ($do_local_domain) |
| 2128 | if ($parent =~ /\\@(.+)>/) { |
| 2129 | $local_domain = lc($1); |
| 2130 | ++$delivered_messages_local_domain{$local_domain}; |
| 2131 | ++$delivered_addresses_local_domain{$local_domain}; |
| 2132 | add_volume(\\$delivered_data_local_domain{$local_domain},\\$delivered_data_gigs_local_domain{$local_domain},$size); |
| 2133 | } |
| 2134 | #ENDIF ($do_local_domain) |
| 2135 | } |
| 2136 | } |
| 2137 | ++$delivered_messages_user{$user}; |
| 2138 | ++$delivered_addresses_user{$user}; |
| 2139 | add_volume(\\$delivered_data_user{$user},\\$delivered_data_gigs_user{$user},$size); |
| 2140 | } |
| 2141 | } |
| 2142 | #ENDIF ($local_league_table || $include_remote_users) |
| 2143 | |
| 2144 | #IFDEF ($do_sender{Host}) |
| 2145 | $delivered_messages{Host}{$host}++; |
| 2146 | $delivered_addresses{Host}{$host}++; |
| 2147 | add_volume(\\$delivered_data{Host}{$host},\\$delivered_data_gigs{Host}{$host},$size); |
| 2148 | #ENDIF ($do_sender{Host}) |
| 2149 | #IFDEF ($do_sender{Domain}) |
| 2150 | if ($domain) { |
| 2151 | ++$delivered_messages{Domain}{$domain}; |
| 2152 | ++$delivered_addresses{Domain}{$domain}; |
| 2153 | add_volume(\\$delivered_data{Domain}{$domain},\\$delivered_data_gigs{Domain}{$domain},$size); |
| 2154 | } |
| 2155 | #ENDIF ($do_sender{Domain}) |
| 2156 | #IFDEF ($do_sender{Email}) |
| 2157 | ++$delivered_messages{Email}{$email}; |
| 2158 | ++$delivered_addresses{Email}{$email}; |
| 2159 | add_volume(\\$delivered_data{Email}{$email},\\$delivered_data_gigs{Email}{$email},$size); |
| 2160 | #ENDIF ($do_sender{Email}) |
| 2161 | #IFDEF ($do_sender{Edomain}) |
| 2162 | ++$delivered_messages{Edomain}{$edomain}; |
| 2163 | ++$delivered_addresses{Edomain}{$edomain}; |
| 2164 | add_volume(\\$delivered_data{Edomain}{$edomain},\\$delivered_data_gigs{Edomain}{$edomain},$size); |
| 2165 | #ENDIF ($do_sender{Edomain}) |
| 2166 | |
| 2167 | ++$total_delivered_messages; |
| 2168 | ++$total_delivered_addresses; |
| 2169 | add_volume(\\$total_delivered_data,\\$total_delivered_data_gigs,$size); |
| 2170 | |
| 2171 | #IFDEF ($show_transport) |
| 2172 | my $transport = (/\\sT=(\\S+)/) ? $1 : ":blackhole:"; |
| 2173 | ++$transported_count{$transport}; |
| 2174 | add_volume(\\$transported_data{$transport},\\$transported_data_gigs{$transport},$size); |
| 2175 | #ENDIF ($show_transport) |
| 2176 | |
| 2177 | #IFDEF ($hist_opt > 0) |
| 2178 | $delivered_interval_count[($m_hour*60 + $m_min)/$hist_interval]++; |
| 2179 | #ENDIF ($hist_opt > 0) |
| 2180 | |
| 2181 | #IFDEF ($#delivery_times > 0) |
| 2182 | if (/ DT=(\S+)/) { |
| 2183 | $seconds = wdhms_seconds($1); |
| 2184 | for ($i = 0; $i <= $#delivery_times; $i++) { |
| 2185 | if ($seconds < $delivery_times[$i]) { |
| 2186 | ++$dt_all_bin[$i]; |
| 2187 | ++$dt_remote_bin[$i] if $message[$REMOTE_DELIVERED]; |
| 2188 | last; |
| 2189 | } |
| 2190 | } |
| 2191 | if ($i > $#delivery_times) { |
| 2192 | ++$dt_all_overflow; |
| 2193 | ++$dt_remote_overflow if $message[$REMOTE_DELIVERED]; |
| 2194 | } |
| 2195 | } |
| 2196 | #ENDIF ($#delivery_times > 0) |
| 2197 | |
| 2198 | } |
| 2199 | |
| 2200 | elsif ($flag eq "->") { |
| 2201 | |
| 2202 | #IFDEF ($local_league_table || $include_remote_users) |
| 2203 | #IFDEF ($local_league_table && $include_remote_users) |
| 2204 | { #Store both local and remote users. |
| 2205 | #ENDIF ($local_league_table && $include_remote_users) |
| 2206 | |
| 2207 | #IFDEF ($local_league_table && ! $include_remote_users) |
| 2208 | if ($host eq "local") { #Store local users only. |
| 2209 | #ENDIF ($local_league_table && ! $include_remote_users) |
| 2210 | |
| 2211 | #IFDEF ($include_remote_users && ! $local_league_table) |
| 2212 | if ($host ne "local") { #Store remote users only. |
| 2213 | #ENDIF ($include_remote_users && ! $local_league_table) |
| 2214 | |
| 2215 | if (my($user) = split((/\\s</)? " <" : " ", $_)) { |
| 2216 | #IFDEF ($include_original_destination) |
| 2217 | { |
| 2218 | #ENDIF ($include_original_destination) |
| 2219 | #IFNDEF ($include_original_destination) |
| 2220 | if ($user =~ /^[\\/|]/) { |
| 2221 | #ENDIF ($include_original_destination) |
| 2222 | #my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/; |
| 2223 | my($parent) = $_ =~ / (<.+?>) /; #DT 1.54 |
| 2224 | $user = "$user $parent" if defined $parent; |
| 2225 | } |
| 2226 | ++$delivered_addresses_user{$user}; |
| 2227 | } |
| 2228 | } |
| 2229 | #ENDIF ($local_league_table || $include_remote_users) |
| 2230 | |
| 2231 | #IFDEF ($do_sender{Host}) |
| 2232 | $delivered_addresses{Host}{$host}++; |
| 2233 | #ENDIF ($do_sender{Host}) |
| 2234 | #IFDEF ($do_sender{Domain}) |
| 2235 | if ($domain) { |
| 2236 | ++$delivered_addresses{Domain}{$domain}; |
| 2237 | } |
| 2238 | #ENDIF ($do_sender{Domain}) |
| 2239 | #IFDEF ($do_sender{Email}) |
| 2240 | ++$delivered_addresses{Email}{$email}; |
| 2241 | #ENDIF ($do_sender{Email}) |
| 2242 | #IFDEF ($do_sender{Edomain}) |
| 2243 | ++$delivered_addresses{Edomain}{$edomain}; |
| 2244 | #ENDIF ($do_sender{Edomain}) |
| 2245 | |
| 2246 | ++$total_delivered_addresses; |
| 2247 | } |
| 2248 | |
| 2249 | elsif ($flag eq "==" && defined($message[$SIZE]) && !defined($message[$DELAYED])) { |
| 2250 | ++$delayed_count; |
| 2251 | $message[$DELAYED] = 1; |
| 2252 | } |
| 2253 | |
| 2254 | elsif ($flag eq "**") { |
| 2255 | if (defined ($message[$SIZE])) { |
| 2256 | unless (defined $message[$HAD_ERROR]) { |
| 2257 | ++$message_errors; |
| 2258 | $message[$HAD_ERROR] = 1; |
| 2259 | } |
| 2260 | } |
| 2261 | |
| 2262 | #IFDEF ($show_errors) |
| 2263 | ++$errors_count{$_}; |
| 2264 | #ENDIF ($show_errors) |
| 2265 | |
| 2266 | } |
| 2267 | |
| 2268 | elsif ($flag eq "Co") { |
| 2269 | #Completed? |
| 2270 | #IFDEF ($#queue_times >= 0) |
| 2271 | $queued = queue_time($tod, $message[$ARRIVAL_TIME], $id); |
| 2272 | |
| 2273 | for ($i = 0; $i <= $#queue_times; $i++) { |
| 2274 | if ($queued < $queue_times[$i]) { |
| 2275 | ++$qt_all_bin[$i]; |
| 2276 | ++$qt_remote_bin[$i] if $message[$REMOTE_DELIVERED]; |
| 2277 | last; |
| 2278 | } |
| 2279 | } |
| 2280 | if ($i > $#queue_times) { |
| 2281 | ++$qt_all_overflow; |
| 2282 | ++$qt_remote_overflow if $message[$REMOTE_DELIVERED]; |
| 2283 | } |
| 2284 | #ENDIF ($#queue_times >= 0) |
| 2285 | |
| 2286 | #IFDEF ($#rcpt_times >= 0) |
| 2287 | if (/ QT=(\S+)/) { |
| 2288 | $seconds = wdhms_seconds($1); |
| 2289 | #Calculate $queued if not previously calculated above. |
| 2290 | #IFNDEF ($#queue_times >= 0) |
| 2291 | $queued = queue_time($tod, $message[$ARRIVAL_TIME], $id); |
| 2292 | #ENDIF ($#queue_times >= 0) |
| 2293 | $rcpt_time = $seconds - $queued; |
| 2294 | my($protocol); |
| 2295 | |
| 2296 | if (defined $message[$PROTOCOL]) { |
| 2297 | $protocol = $message[$PROTOCOL]; |
| 2298 | |
| 2299 | # Create the bin if its not already defined. |
| 2300 | unless (exists $rcpt_times_bin{$protocol}) { |
| 2301 | initialise_rcpt_times($protocol); |
| 2302 | } |
| 2303 | } |
| 2304 | |
| 2305 | |
| 2306 | for ($i = 0; $i <= $#rcpt_times; ++$i) { |
| 2307 | if ($rcpt_time < $rcpt_times[$i]) { |
| 2308 | ++$rcpt_times_bin{all}[$i]; |
| 2309 | ++$rcpt_times_bin{$protocol}[$i] if defined $protocol; |
| 2310 | last; |
| 2311 | } |
| 2312 | } |
| 2313 | |
| 2314 | if ($i > $#rcpt_times) { |
| 2315 | ++$rcpt_times_overflow{all}; |
| 2316 | ++$rcpt_times_overflow{$protocol} if defined $protocol; |
| 2317 | } |
| 2318 | } |
| 2319 | #ENDIF ($#rcpt_times >= 0) |
| 2320 | |
| 2321 | delete($messages{$id}); |
| 2322 | } |
| 2323 | elsif ($flag eq "SA") { |
| 2324 | $ip = (/From.*?(\\[[^]]+\\])/ || /\\((local)\\)/) ? $1 : ""; |
| 2325 | #SpamAssassin message |
| 2326 | if (/Action: ((permanently|temporarily) rejected message|flagged as Spam but accepted): score=(\d+\.\d)/) { |
| 2327 | #add_volume(\\$spam_score,\\$spam_score_gigs,$3); |
| 2328 | ++$spam_count_by_ip{$ip}; |
| 2329 | } elsif (/Action: scanned but message isn\'t spam: score=(-?\d+\.\d)/) { |
| 2330 | #add_volume(\\$ham_score,\\$ham_score_gigs,$1); |
| 2331 | ++$ham_count_by_ip{$ip}; |
| 2332 | } elsif (/(Not running SA because SAEximRunCond expanded to false|check skipped due to message size)/) { |
| 2333 | ++$ham_count_by_ip{$ip}; |
| 2334 | } |
| 2335 | } |
| 2336 | |
| 2337 | # Look for Reject messages or blackholed messages (deliveries |
| 2338 | # without a transport) |
| 2339 | if ($flag eq "Re" || ($flag eq "=>" && ! /\\sT=\\S+/)) { |
| 2340 | # Correct the IP address for rejects: |
| 2341 | # rejected EHLO from my.test.net [10.0.0.5]: syntactically invalid argument(s): |
| 2342 | # rejected EHLO from [10.0.0.6]: syntactically invalid argument(s): |
| 2343 | $ip = $1 if ($ip eq "local" && /^rejected [HE][HE]LO from .*?(\[.+?\]):/); |
| 2344 | if (/SpamAssassin/) { |
| 2345 | ++$rejected_count_by_reason{"Rejected by SpamAssassin"}; |
| 2346 | ++$rejected_count_by_ip{$ip}; |
| 2347 | } |
| 2348 | elsif ( |
| 2349 | /(temporarily rejected [A-Z]*) .*?(: .*?)(:|\s*$)/ |
| 2350 | ) { |
| 2351 | ++$temporarily_rejected_count_by_reason{"\u$1$2"}; |
| 2352 | ++$temporarily_rejected_count_by_ip{$ip}; |
| 2353 | } |
| 2354 | elsif ( |
| 2355 | /(temporarily refused connection)/ |
| 2356 | ) { |
| 2357 | ++$temporarily_rejected_count_by_reason{"\u$1"}; |
| 2358 | ++$temporarily_rejected_count_by_ip{$ip}; |
| 2359 | } |
| 2360 | elsif ( |
| 2361 | /(listed at [^ ]+)/ || |
| 2362 | /(Forged IP detected in HELO)/ || |
| 2363 | /(Invalid domain or IP given in HELO\/EHLO)/ || |
| 2364 | /(unqualified recipient rejected)/ || |
| 2365 | /(closed connection (after|in response) .*?)\s*$/ || |
| 2366 | /(sender rejected)/ || |
| 2367 | # 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. |
| 2368 | # 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. |
| 2369 | / rejected after DATA: (.*)/ || |
| 2370 | /.DATA ACL discarded recipients.: (.*)/ || |
| 2371 | /rejected after DATA: (unqualified address not permitted)/ || |
| 2372 | /(VRFY rejected)/ || |
| 2373 | # /(sender verify (defer|fail))/i || |
| 2374 | /(too many recipients)/ || |
| 2375 | /(refused relay.*?) to/ || |
| 2376 | /(rejected by non-SMTP ACL: .*)/ || |
| 2377 | /(rejected by local_scan.*)/ || |
| 2378 | # SMTP call from %s dropped: too many syntax or protocol errors (last command was "%s" |
| 2379 | # SMTP call from %s dropped: too many nonmail commands |
| 2380 | /(dropped: too many ((nonmail|unrecognized) commands|syntax or protocol errors))/ || |
| 2381 | |
| 2382 | # local_scan() function crashed with signal %d - message temporarily rejected |
| 2383 | # local_scan() function timed out - message temporarily rejected |
| 2384 | /(local_scan.. function .* - message temporarily rejected)/ || |
| 2385 | # SMTP protocol synchronization error (input sent without waiting for greeting): rejected connection from %s |
| 2386 | /(SMTP protocol .*?(error|violation))/ || |
| 2387 | /(message too big)/ |
| 2388 | ) { |
| 2389 | ++$rejected_count_by_reason{"\u$1"}; |
| 2390 | ++$rejected_count_by_ip{$ip}; |
| 2391 | } |
| 2392 | elsif (/rejected [HE][HE]LO from [^:]*: syntactically invalid argument/) { |
| 2393 | ++$rejected_count_by_reason{"Rejected HELO/EHLO: syntactically invalid argument"}; |
| 2394 | ++$rejected_count_by_ip{$ip}; |
| 2395 | } |
| 2396 | elsif (/response to "RCPT TO.*? was: (.*)/) { |
| 2397 | ++$rejected_count_by_reason{"Response to RCPT TO was: $1"}; |
| 2398 | ++$rejected_count_by_ip{$ip}; |
| 2399 | } |
| 2400 | elsif ( |
| 2401 | /(lookup of host )\S+ (failed)/ || |
| 2402 | |
| 2403 | # rejected from <%s>%s%s%s%s: message too big: |
| 2404 | /(rejected [A-Z]*) .*?(: .*?)(:|\s*$)/ || |
| 2405 | # refused connection from %s (host_reject_connection) |
| 2406 | # refused connection from %s (tcp wrappers) |
| 2407 | /(refused connection )from.*? (\(.*)/ || |
| 2408 | |
| 2409 | # 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 |
| 2410 | # 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. |
| 2411 | /(error from remote mailer after .*?:).*(: .*?)(:|\s*$)/ || |
| 2412 | |
| 2413 | # a.b.c F=<a@b.c> rejected after DATA: "@" or "." expected after "Undisclosed-Recipient": failing address in "To" header is: <Undisclosed-Recipient:;> |
| 2414 | /rejected after DATA: ("." or "." expected).*?(: failing address in .*? header)/ || |
| 2415 | |
| 2416 | # connection from %s refused load average = %.2f |
| 2417 | /(Connection )from.*? (refused: load average)/ || |
| 2418 | # connection from %s refused (IP options) |
| 2419 | # Connection from %s refused: too many connections |
| 2420 | # connection from %s refused |
| 2421 | /([Cc]onnection )from.*? (refused.*)/ || |
| 2422 | # [10.0.0.1]: connection refused |
| 2423 | /: (Connection refused)()/ |
| 2424 | ) { |
| 2425 | ++$rejected_count_by_reason{"\u$1$2"}; |
| 2426 | ++$rejected_count_by_ip{$ip}; |
| 2427 | } |
| 2428 | elsif ( |
| 2429 | # 2008-03-31 06:25:22 H=mail.densitron.com [216.70.140.224]:45386 temporarily rejected connection in "connect" ACL: too fast reconnects // .hs |
| 2430 | # 2008-03-31 06:25:22 H=mail.densitron.com [216.70.140.224]:45386 temporarily rejected connection in "connect" ACL // .hs |
| 2431 | /(temporarily rejected connection in .*?ACL:?.*)/ |
| 2432 | ) { |
| 2433 | ++$temporarily_rejected_count_by_ip{$ip}; |
| 2434 | ++$temporarily_rejected_count_by_reason{"\u$1"}; |
| 2435 | } |
| 2436 | else { |
| 2437 | ++$rejected_count_by_reason{Unknown}; |
| 2438 | ++$rejected_count_by_ip{$ip}; |
| 2439 | print STDERR "Unknown rejection: $_" if $debug; |
| 2440 | } |
| 2441 | } |
| 2442 | }'; |
| 2443 | |
| 2444 | # We now do a 'C preprocessor style operation on our parser |
| 2445 | # to remove bits not in use. |
| 2446 | my(%defines_in_operation,$removing_lines,$processed_parser); |
| 2447 | foreach (split (/\n/,$parser)) { |
| 2448 | if ((/^\s*#\s*IFDEF\s*\((.*?)\)/i && ! eval $1) || |
| 2449 | (/^\s*#\s*IFNDEF\s*\((.*?)\)/i && eval $1) ) { |
| 2450 | $defines_in_operation{$1} = 1; |
| 2451 | $removing_lines = 1; |
| 2452 | } |
| 2453 | |
| 2454 | # Convert constants. |
| 2455 | while (/(\$[A-Z][A-Z_]*)\b/) { |
| 2456 | my $constant = eval $1; |
| 2457 | s/(\$[A-Z][A-Z_]*)\b/$constant/; |
| 2458 | } |
| 2459 | |
| 2460 | $processed_parser .= $_."\n" unless $removing_lines; |
| 2461 | |
| 2462 | if (/^\s*#\s*ENDIF\s*\((.*?)\)/i) { |
| 2463 | delete $defines_in_operation{$1}; |
| 2464 | unless (keys %defines_in_operation) { |
| 2465 | $removing_lines = 0; |
| 2466 | } |
| 2467 | } |
| 2468 | } |
| 2469 | print STDERR "# START OF PARSER:$processed_parser\n# END OF PARSER\n\n" if $debug; |
| 2470 | |
| 2471 | return $processed_parser; |
| 2472 | } |
| 2473 | |
| 2474 | |
| 2475 | |
| 2476 | ####################################################################### |
| 2477 | # parse(); |
| 2478 | # |
| 2479 | # parse($parser,\*FILEHANDLE); |
| 2480 | # |
| 2481 | # This subroutine accepts a parser and a filehandle from main and parses each |
| 2482 | # line. We store the results into global variables. |
| 2483 | ####################################################################### |
| 2484 | sub parse { |
| 2485 | my($parser,$fh) = @_; |
| 2486 | |
| 2487 | if ($merge_reports) { |
| 2488 | parse_old_eximstat_reports($fh); |
| 2489 | } |
| 2490 | else { |
| 2491 | eval $parser; |
| 2492 | die ($@) if $@; |
| 2493 | } |
| 2494 | |
| 2495 | } |
| 2496 | |
| 2497 | |
| 2498 | |
| 2499 | ####################################################################### |
| 2500 | # print_header(); |
| 2501 | # |
| 2502 | # print_header(); |
| 2503 | # |
| 2504 | # Print our headers and contents. |
| 2505 | ####################################################################### |
| 2506 | sub print_header { |
| 2507 | |
| 2508 | |
| 2509 | my $title = "Exim statistics from $begin to $end"; |
| 2510 | |
| 2511 | print $txt_fh "\n$title\n" if $txt_fh; |
| 2512 | if ($htm_fh) { |
| 2513 | print $htm_fh html_header($title); |
| 2514 | print $htm_fh "<ul>\n"; |
| 2515 | print $htm_fh "<li><a href=\"#Grandtotal\">Grand total summary</a>\n"; |
| 2516 | print $htm_fh "<li><a href=\"#Patterns\">User Specified Patterns</a>\n" if @user_patterns; |
| 2517 | print $htm_fh "<li><a href=\"#Transport\">Deliveries by Transport</a>\n" if $show_transport; |
| 2518 | if ($hist_opt) { |
| 2519 | print $htm_fh "<li><a href=\"#Messages received\">Messages received per hour</a>\n"; |
| 2520 | print $htm_fh "<li><a href=\"#Deliveries\">Deliveries per hour</a>\n"; |
| 2521 | } |
| 2522 | |
| 2523 | if ($#queue_times >= 0) { |
| 2524 | print $htm_fh "<li><a href=\"#Time spent on the queue all messages\">Time spent on the queue: all messages</a>\n"; |
| 2525 | print $htm_fh "<li><a href=\"#Time spent on the queue messages with at least one remote delivery\">Time spent on the queue: messages with at least one remote delivery</a>\n"; |
| 2526 | } |
| 2527 | |
| 2528 | if ($#delivery_times >= 0) { |
| 2529 | print $htm_fh "<li><a href=\"#Delivery times all messages\">Delivery times: all messages</a>\n"; |
| 2530 | print $htm_fh "<li><a href=\"#Delivery times messages with at least one remote delivery\">Delivery times: messages with at least one remote delivery</a>\n"; |
| 2531 | } |
| 2532 | |
| 2533 | if ($#rcpt_times >= 0) { |
| 2534 | print $htm_fh "<li><a href=\"#Receipt times all messages\">Receipt times</a>\n"; |
| 2535 | } |
| 2536 | |
| 2537 | print $htm_fh "<li><a href=\"#Relayed messages\">Relayed messages</a>\n" if $show_relay; |
| 2538 | if ($topcount) { |
| 2539 | 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; |
| 2540 | foreach ('Host','Domain','Email','Edomain') { |
| 2541 | next unless $do_sender{$_}; |
| 2542 | print $htm_fh "<li><a href=\"#Sending \l$_ count\">Top $topcount sending \l${_}s by message count</a>\n"; |
| 2543 | print $htm_fh "<li><a href=\"#Sending \l$_ volume\">Top $topcount sending \l${_}s by volume</a>\n"; |
| 2544 | } |
| 2545 | if (($local_league_table || $include_remote_users) && %received_count_user) { |
| 2546 | print $htm_fh "<li><a href=\"#Local sender count\">Top $topcount local senders by message count</a>\n"; |
| 2547 | print $htm_fh "<li><a href=\"#Local sender volume\">Top $topcount local senders by volume</a>\n"; |
| 2548 | } |
| 2549 | foreach ('Host','Domain','Email','Edomain') { |
| 2550 | next unless $do_sender{$_}; |
| 2551 | print $htm_fh "<li><a href=\"#$_ destination count\">Top $topcount \l$_ destinations by message count</a>\n"; |
| 2552 | print $htm_fh "<li><a href=\"#$_ destination volume\">Top $topcount \l$_ destinations by volume</a>\n"; |
| 2553 | } |
| 2554 | if (($local_league_table || $include_remote_users) && %delivered_messages_user) { |
| 2555 | print $htm_fh "<li><a href=\"#Local destination count\">Top $topcount local destinations by message count</a>\n"; |
| 2556 | print $htm_fh "<li><a href=\"#Local destination volume\">Top $topcount local destinations by volume</a>\n"; |
| 2557 | } |
| 2558 | if (($local_league_table || $include_remote_users) && %delivered_messages_local_domain) { |
| 2559 | print $htm_fh "<li><a href=\"#Local domain destination count\">Top $topcount local domain destinations by message count</a>\n"; |
| 2560 | print $htm_fh "<li><a href=\"#Local domain destination volume\">Top $topcount local domain destinations by volume</a>\n"; |
| 2561 | } |
| 2562 | |
| 2563 | print $htm_fh "<li><a href=\"#Rejected ip count\">Top $topcount rejected ips by message count</a>\n" if %rejected_count_by_ip; |
| 2564 | 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; |
| 2565 | 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; |
| 2566 | |
| 2567 | } |
| 2568 | print $htm_fh "<li><a href=\"#errors\">List of errors</a>\n" if %errors_count; |
| 2569 | print $htm_fh "</ul>\n<hr>\n"; |
| 2570 | } |
| 2571 | if ($xls_fh) |
| 2572 | { |
| 2573 | $ws_global->write($row++, $col+0, "Exim Statistics", $f_header1); |
| 2574 | &set_worksheet_line($ws_global, $row, $col, ["from:", $begin, "to:", $end], $f_default); |
| 2575 | $row+=2; |
| 2576 | } |
| 2577 | } |
| 2578 | |
| 2579 | |
| 2580 | ####################################################################### |
| 2581 | # print_grandtotals(); |
| 2582 | # |
| 2583 | # print_grandtotals(); |
| 2584 | # |
| 2585 | # Print the grand totals. |
| 2586 | ####################################################################### |
| 2587 | sub print_grandtotals { |
| 2588 | |
| 2589 | # Get the sender by headings and results. This is complicated as we can have |
| 2590 | # different numbers of columns. |
| 2591 | my($sender_txt_header,$sender_txt_format,$sender_html_format); |
| 2592 | my(@received_totals,@delivered_totals); |
| 2593 | my($row_tablehead, $row_max); |
| 2594 | my(@col_headers) = ('TOTAL', 'Volume', 'Messages', 'Addresses'); |
| 2595 | |
| 2596 | foreach ('Host','Domain','Email','Edomain') { |
| 2597 | next unless $do_sender{$_}; |
| 2598 | if ($merge_reports) { |
| 2599 | push(@received_totals, get_report_total($report_totals{Received},"${_}s")); |
| 2600 | push(@delivered_totals,get_report_total($report_totals{Delivered},"${_}s")); |
| 2601 | } |
| 2602 | else { |
| 2603 | push(@received_totals,scalar(keys %{$received_data{$_}})); |
| 2604 | push(@delivered_totals,scalar(keys %{$delivered_data{$_}})); |
| 2605 | } |
| 2606 | $sender_txt_header .= " " x ($COLUMN_WIDTHS - length($_)) . $_ . 's'; |
| 2607 | $sender_html_format .= "<td align=\"right\">%s</td>"; |
| 2608 | $sender_txt_format .= " " x ($COLUMN_WIDTHS - 5) . "%6s"; |
| 2609 | push(@col_headers,"${_}s"); |
| 2610 | } |
| 2611 | |
| 2612 | my $txt_format1 = " %-16s %9s %6d %6s $sender_txt_format"; |
| 2613 | my $txt_format2 = " %6d %4.1f%% %6d %4.1f%%", |
| 2614 | 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"; |
| 2615 | my $htm_format2 = "<td align=\"right\">%d</td><td align=\"right\">%4.1f%%</td><td align=\"right\">%d</td><td align=\"right\">%4.1f%%</td>"; |
| 2616 | |
| 2617 | if ($txt_fh) { |
| 2618 | my $sender_spaces = " " x length($sender_txt_header); |
| 2619 | print $txt_fh "\n"; |
| 2620 | print $txt_fh "Grand total summary\n"; |
| 2621 | print $txt_fh "-------------------\n"; |
| 2622 | print $txt_fh " $sender_spaces At least one address\n"; |
| 2623 | print $txt_fh " TOTAL Volume Messages Addresses $sender_txt_header Delayed Failed\n"; |
| 2624 | } |
| 2625 | if ($htm_fh) { |
| 2626 | print $htm_fh "<a name=\"Grandtotal\"></a>\n"; |
| 2627 | print $htm_fh "<h2>Grand total summary</h2>\n"; |
| 2628 | print $htm_fh "<table border=1>\n"; |
| 2629 | print $htm_fh "<tr><th>" . join('</th><th>',@col_headers) . "</th><th colspan=2>At least one addr<br>Delayed</th><th colspan=2>At least one addr<br>Failed</th>\n"; |
| 2630 | } |
| 2631 | if ($xls_fh) { |
| 2632 | $ws_global->write($row++, 0, "Grand total summary", $f_header2); |
| 2633 | $ws_global->write($row, 0, \@col_headers, $f_header2); |
| 2634 | $ws_global->merge_range($row, scalar(@col_headers), $row, scalar(@col_headers)+1, "At least one addr Delayed", $f_header2_m); |
| 2635 | $ws_global->merge_range($row, scalar(@col_headers)+2, $row, scalar(@col_headers)+3, "At least one addr Failed", $f_header2_m); |
| 2636 | #$ws_global->write(++$row, scalar(@col_headers), ['Total','Percent','Total','Percent'], $f_header2); |
| 2637 | } |
| 2638 | |
| 2639 | |
| 2640 | my($volume,$failed_count); |
| 2641 | if ($merge_reports) { |
| 2642 | $volume = volume_rounded($report_totals{Received}{Volume}, $report_totals{Received}{'Volume-gigs'}); |
| 2643 | $total_received_count = get_report_total($report_totals{Received},'Messages'); |
| 2644 | $failed_count = get_report_total($report_totals{Received},'Failed'); |
| 2645 | $delayed_count = get_report_total($report_totals{Received},'Delayed'); |
| 2646 | } |
| 2647 | else { |
| 2648 | $volume = volume_rounded($total_received_data, $total_received_data_gigs); |
| 2649 | $failed_count = $message_errors; |
| 2650 | } |
| 2651 | |
| 2652 | { |
| 2653 | no integer; |
| 2654 | |
| 2655 | my @content=( |
| 2656 | $volume,$total_received_count,'', |
| 2657 | @received_totals, |
| 2658 | $delayed_count, |
| 2659 | ($total_received_count) ? ($delayed_count*100/$total_received_count) : 0, |
| 2660 | $failed_count, |
| 2661 | ($total_received_count) ? ($failed_count*100/$total_received_count) : 0 |
| 2662 | ); |
| 2663 | |
| 2664 | printf $txt_fh ("$txt_format1$txt_format2\n", 'Received', @content) if $txt_fh; |
| 2665 | printf $htm_fh ("$htm_format1$htm_format2\n", 'Received', @content) if $htm_fh; |
| 2666 | if ($xls_fh) { |
| 2667 | $ws_global->write(++$row, 0, 'Received', $f_default); |
| 2668 | for (my $i=0; $i < scalar(@content); $i++) { |
| 2669 | if ($i == 4 || $i == 6) { |
| 2670 | $ws_global->write($row, $i+1, $content[$i]/100, $f_percent); |
| 2671 | } |
| 2672 | else { |
| 2673 | $ws_global->write($row, $i+1, $content[$i], $f_default); |
| 2674 | } |
| 2675 | } |
| 2676 | } |
| 2677 | } |
| 2678 | |
| 2679 | if ($merge_reports) { |
| 2680 | $volume = volume_rounded($report_totals{Delivered}{Volume}, $report_totals{Delivered}{'Volume-gigs'}); |
| 2681 | $total_delivered_messages = get_report_total($report_totals{Delivered},'Messages'); |
| 2682 | $total_delivered_addresses = get_report_total($report_totals{Delivered},'Addresses'); |
| 2683 | } |
| 2684 | else { |
| 2685 | $volume = volume_rounded($total_delivered_data, $total_delivered_data_gigs); |
| 2686 | } |
| 2687 | |
| 2688 | my @content=($volume, $total_delivered_messages, $total_delivered_addresses, @delivered_totals); |
| 2689 | printf $txt_fh ("$txt_format1\n", 'Delivered', @content) if $txt_fh; |
| 2690 | printf $htm_fh ("$htm_format1\n", 'Delivered', @content) if $htm_fh; |
| 2691 | |
| 2692 | if ($xls_fh) { |
| 2693 | $ws_global->write(++$row, 0, 'Delivered', $f_default); |
| 2694 | for (my $i=0; $i < scalar(@content); $i++) { |
| 2695 | $ws_global->write($row, $i+1, $content[$i], $f_default); |
| 2696 | } |
| 2697 | } |
| 2698 | |
| 2699 | if ($merge_reports) { |
| 2700 | foreach ('Rejects', 'Temp Rejects', 'Ham', 'Spam') { |
| 2701 | my $messages = get_report_total($report_totals{$_},'Messages'); |
| 2702 | my $addresses = get_report_total($report_totals{$_},'Addresses'); |
| 2703 | if ($messages) { |
| 2704 | @content = ($_, '', $messages, ''); |
| 2705 | push(@content,get_report_total($report_totals{$_},'Hosts')) if $do_sender{Host}; |
| 2706 | #These rows do not have entries for the following columns (if specified) |
| 2707 | foreach ('Domain','Email','Edomain') { |
| 2708 | push(@content,'') if $do_sender{$_}; |
| 2709 | } |
| 2710 | |
| 2711 | printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; |
| 2712 | printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; |
| 2713 | $ws_global->write(++$row, 0, \@content) if $xls_fh; |
| 2714 | } |
| 2715 | } |
| 2716 | } |
| 2717 | else { |
| 2718 | foreach my $total_aref (['Rejects',\%rejected_count_by_ip], |
| 2719 | ['Temp Rejects',\%temporarily_rejected_count_by_ip], |
| 2720 | ['Ham',\%ham_count_by_ip], |
| 2721 | ['Spam',\%spam_count_by_ip]) { |
| 2722 | #Count the number of messages of this type. |
| 2723 | my $messages = 0; |
| 2724 | map {$messages += $_} values %{$total_aref->[1]}; |
| 2725 | |
| 2726 | if ($messages > 0) { |
| 2727 | @content = ($total_aref->[0], '', $messages, ''); |
| 2728 | |
| 2729 | #Count the number of distict IPs for the Hosts column. |
| 2730 | push(@content,scalar(keys %{$total_aref->[1]})) if $do_sender{Host}; |
| 2731 | |
| 2732 | #These rows do not have entries for the following columns (if specified) |
| 2733 | foreach ('Domain','Email','Edomain') { |
| 2734 | push(@content,'') if $do_sender{$_}; |
| 2735 | } |
| 2736 | |
| 2737 | printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; |
| 2738 | printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; |
| 2739 | $ws_global->write(++$row, 0, \@content) if $xls_fh; |
| 2740 | } |
| 2741 | } |
| 2742 | } |
| 2743 | |
| 2744 | printf $txt_fh "\n" if $txt_fh; |
| 2745 | printf $htm_fh "</table>\n" if $htm_fh; |
| 2746 | ++$row; |
| 2747 | } |
| 2748 | |
| 2749 | |
| 2750 | ####################################################################### |
| 2751 | # print_user_patterns() |
| 2752 | # |
| 2753 | # print_user_patterns(); |
| 2754 | # |
| 2755 | # Print the counts of user specified patterns. |
| 2756 | ####################################################################### |
| 2757 | sub print_user_patterns { |
| 2758 | my $txt_format1 = " %-18s %6d"; |
| 2759 | my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%d</td>"; |
| 2760 | |
| 2761 | if ($txt_fh) { |
| 2762 | print $txt_fh "User Specified Patterns\n"; |
| 2763 | print $txt_fh "-----------------------"; |
| 2764 | print $txt_fh "\n Total\n"; |
| 2765 | } |
| 2766 | if ($htm_fh) { |
| 2767 | print $htm_fh "<hr><a name=\"Patterns\"></a><h2>User Specified Patterns</h2>\n"; |
| 2768 | print $htm_fh "<table border=0 width=\"100%\">\n"; |
| 2769 | print $htm_fh "<tr><td>\n"; |
| 2770 | print $htm_fh "<table border=1>\n"; |
| 2771 | print $htm_fh "<tr><th> </th><th>Total</th>\n"; |
| 2772 | } |
| 2773 | if ($xls_fh) { |
| 2774 | $ws_global->write($row++, $col, "User Specified Patterns", $f_header2); |
| 2775 | &set_worksheet_line($ws_global, $row++, 1, ["Total"], $f_headertab); |
| 2776 | } |
| 2777 | |
| 2778 | |
| 2779 | my($key); |
| 2780 | if ($merge_reports) { |
| 2781 | # We are getting our data from previous reports. |
| 2782 | foreach $key (@user_descriptions) { |
| 2783 | my $count = get_report_total($report_totals{patterns}{$key},'Total'); |
| 2784 | printf $txt_fh ("$txt_format1\n",$key,$count) if $txt_fh; |
| 2785 | printf $htm_fh ("$htm_format1\n",$key,$count) if $htm_fh; |
| 2786 | if ($xls_fh) |
| 2787 | { |
| 2788 | &set_worksheet_line($ws_global, $row++, 0, [$key,$count], $f_default); |
| 2789 | } |
| 2790 | } |
| 2791 | } |
| 2792 | else { |
| 2793 | # We are getting our data from mainlog files. |
| 2794 | my $user_pattern_index = 0; |
| 2795 | foreach $key (@user_descriptions) { |
| 2796 | printf $txt_fh ("$txt_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $txt_fh; |
| 2797 | printf $htm_fh ("$htm_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $htm_fh; |
| 2798 | $ws_global->write($row++, 0, [$key,$user_pattern_totals[$user_pattern_index]]) if $xls_fh; |
| 2799 | $user_pattern_index++; |
| 2800 | } |
| 2801 | } |
| 2802 | print $txt_fh "\n" if $txt_fh; |
| 2803 | print $htm_fh "</table>\n\n" if $htm_fh; |
| 2804 | if ($xls_fh) |
| 2805 | { |
| 2806 | ++$row; |
| 2807 | } |
| 2808 | |
| 2809 | if ($hist_opt > 0) { |
| 2810 | my $user_pattern_index = 0; |
| 2811 | foreach $key (@user_descriptions) { |
| 2812 | print_histogram($key, 'occurence', @{$user_pattern_interval_count[$user_pattern_index]}); |
| 2813 | $user_pattern_index++; |
| 2814 | } |
| 2815 | } |
| 2816 | } |
| 2817 | |
| 2818 | ####################################################################### |
| 2819 | # print_rejects() |
| 2820 | # |
| 2821 | # print_rejects(); |
| 2822 | # |
| 2823 | # Print statistics about rejected mail. |
| 2824 | ####################################################################### |
| 2825 | sub print_rejects { |
| 2826 | my($format1,$reason); |
| 2827 | |
| 2828 | my $txt_format1 = " %-40s %6d"; |
| 2829 | my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%d</td>"; |
| 2830 | |
| 2831 | if ($txt_fh) { |
| 2832 | print $txt_fh "Rejected mail by reason\n"; |
| 2833 | print $txt_fh "-----------------------"; |
| 2834 | print $txt_fh "\n Total\n"; |
| 2835 | } |
| 2836 | if ($htm_fh) { |
| 2837 | print $htm_fh "<hr><a name=\"patterns\"></a><h2>Rejected mail by reason</h2>\n"; |
| 2838 | print $htm_fh "<table border=0 width=\"100%\"><tr><td><table border=1>\n"; |
| 2839 | print $htm_fh "<tr><th> </th><th>Total</th>\n"; |
| 2840 | } |
| 2841 | if ($xls_fh) { |
| 2842 | $ws_global->write($row++, $col, "Rejected mail by reason", $f_header2); |
| 2843 | &set_worksheet_line($ws_global, $row++, 1, ["Total"], $f_headertab); |
| 2844 | } |
| 2845 | |
| 2846 | |
| 2847 | my $href = ($merge_reports) ? $report_totals{rejected_mail_by_reason} : \%rejected_count_by_reason; |
| 2848 | my(@chartdatanames, @chartdatavals_count); |
| 2849 | |
| 2850 | foreach $reason (top_n_sort($topcount, $href, undef, undef)) { |
| 2851 | printf $txt_fh ("$txt_format1\n",$reason,$href->{$reason}) if $txt_fh; |
| 2852 | printf $htm_fh ("$htm_format1\n",$reason,$href->{$reason}) if $htm_fh; |
| 2853 | set_worksheet_line($ws_global, $row++, 0, [$reason,$href->{$reason}], $f_default) if $xls_fh; |
| 2854 | push(@chartdatanames, $reason); |
| 2855 | push(@chartdatavals_count, $href->{$reason}); |
| 2856 | } |
| 2857 | |
| 2858 | $row++ if $xls_fh; |
| 2859 | print $txt_fh "\n" if $txt_fh; |
| 2860 | |
| 2861 | if ($htm_fh) { |
| 2862 | print $htm_fh "</tr></table></td><td>"; |
| 2863 | if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_count > 0)) { |
| 2864 | # calculate the graph |
| 2865 | my @data = ( |
| 2866 | \@chartdatanames, |
| 2867 | \@chartdatavals_count |
| 2868 | ); |
| 2869 | my $graph = GD::Graph::pie->new(200, 200); |
| 2870 | $graph->set( |
| 2871 | x_label => 'Rejection Reasons', |
| 2872 | y_label => 'Messages', |
| 2873 | title => 'By count', |
| 2874 | ); |
| 2875 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 2876 | if ($gd) { |
| 2877 | open(IMG, ">$chartdir/rejections_count.png") or die "Could not write $chartdir/rejections_count.png: $!\n"; |
| 2878 | binmode IMG; |
| 2879 | print IMG $gd->png; |
| 2880 | close IMG; |
| 2881 | print $htm_fh "<img src=\"$chartrel/rejections_count.png\">"; |
| 2882 | } |
| 2883 | } |
| 2884 | print $htm_fh "</td></tr></table>\n\n"; |
| 2885 | } |
| 2886 | } |
| 2887 | |
| 2888 | |
| 2889 | |
| 2890 | |
| 2891 | |
| 2892 | ####################################################################### |
| 2893 | # print_transport(); |
| 2894 | # |
| 2895 | # print_transport(); |
| 2896 | # |
| 2897 | # Print totals by transport. |
| 2898 | ####################################################################### |
| 2899 | sub print_transport { |
| 2900 | my(@chartdatanames); |
| 2901 | my(@chartdatavals_count); |
| 2902 | my(@chartdatavals_vol); |
| 2903 | no integer; #Lose this for charting the data. |
| 2904 | |
| 2905 | my $txt_format1 = " %-18s %6s %6d"; |
| 2906 | my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%s</td><td align=\"right\">%d</td>"; |
| 2907 | |
| 2908 | if ($txt_fh) { |
| 2909 | print $txt_fh "Deliveries by transport\n"; |
| 2910 | print $txt_fh "-----------------------"; |
| 2911 | print $txt_fh "\n Volume Messages\n"; |
| 2912 | } |
| 2913 | if ($htm_fh) { |
| 2914 | print $htm_fh "<hr><a name=\"Transport\"></a><h2>Deliveries by Transport</h2>\n"; |
| 2915 | print $htm_fh "<table border=0 width=\"100%\"><tr><td><table border=1>\n"; |
| 2916 | print $htm_fh "<tr><th> </th><th>Volume</th><th>Messages</th>\n"; |
| 2917 | } |
| 2918 | if ($xls_fh) { |
| 2919 | $ws_global->write(++$row, $col, "Deliveries by transport", $f_header2); |
| 2920 | $ws_global->write(++$row, 1, ["Volume", "Messages"], $f_headertab); |
| 2921 | } |
| 2922 | |
| 2923 | my($key); |
| 2924 | if ($merge_reports) { |
| 2925 | # We are getting our data from previous reports. |
| 2926 | foreach $key (sort keys %{$report_totals{transport}}) { |
| 2927 | my $count = get_report_total($report_totals{transport}{$key},'Messages'); |
| 2928 | my @content=($key, volume_rounded($report_totals{transport}{$key}{Volume}, |
| 2929 | $report_totals{transport}{$key}{'Volume-gigs'}), $count); |
| 2930 | push(@chartdatanames, $key); |
| 2931 | push(@chartdatavals_count, $count); |
| 2932 | push(@chartdatavals_vol, $report_totals{transport}{$key}{'Volume-gigs'}*$gig + $report_totals{transport}{$key}{Volume} ); |
| 2933 | printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; |
| 2934 | printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; |
| 2935 | $ws_global->write(++$row, 0, \@content) if $xls_fh; |
| 2936 | } |
| 2937 | } |
| 2938 | else { |
| 2939 | # We are getting our data from mainlog files. |
| 2940 | foreach $key (sort keys %transported_data) { |
| 2941 | my @content=($key, volume_rounded($transported_data{$key},$transported_data_gigs{$key}), |
| 2942 | $transported_count{$key}); |
| 2943 | push(@chartdatanames, $key); |
| 2944 | push(@chartdatavals_count, $transported_count{$key}); |
| 2945 | push(@chartdatavals_vol, $transported_data_gigs{$key}*$gig + $transported_data{$key}); |
| 2946 | printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; |
| 2947 | printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; |
| 2948 | $ws_global->write(++$row, 0, \@content) if $xls_fh; |
| 2949 | } |
| 2950 | } |
| 2951 | print $txt_fh "\n" if $txt_fh; |
| 2952 | if ($htm_fh) { |
| 2953 | print $htm_fh "</tr></table></td><td>"; |
| 2954 | |
| 2955 | if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_count > 0)) |
| 2956 | { |
| 2957 | # calculate the graph |
| 2958 | my @data = ( |
| 2959 | \@chartdatanames, |
| 2960 | \@chartdatavals_count |
| 2961 | ); |
| 2962 | my $graph = GD::Graph::pie->new(200, 200); |
| 2963 | $graph->set( |
| 2964 | x_label => 'Transport', |
| 2965 | y_label => 'Messages', |
| 2966 | title => 'By count', |
| 2967 | ); |
| 2968 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 2969 | if ($gd) { |
| 2970 | open(IMG, ">$chartdir/transports_count.png") or die "Could not write $chartdir/transports_count.png: $!\n"; |
| 2971 | binmode IMG; |
| 2972 | print IMG $gd->png; |
| 2973 | close IMG; |
| 2974 | print $htm_fh "<img src=\"$chartrel/transports_count.png\">"; |
| 2975 | } |
| 2976 | } |
| 2977 | print $htm_fh "</td><td>"; |
| 2978 | |
| 2979 | if ($HAVE_GD_Graph_pie && $charts && ($#chartdatavals_vol > 0)) { |
| 2980 | my @data = ( |
| 2981 | \@chartdatanames, |
| 2982 | \@chartdatavals_vol |
| 2983 | ); |
| 2984 | my $graph = GD::Graph::pie->new(200, 200); |
| 2985 | $graph->set( |
| 2986 | title => 'By volume', |
| 2987 | ); |
| 2988 | my $gd = $graph->plot(\@data) or warn($graph->error); |
| 2989 | if ($gd) { |
| 2990 | open(IMG, ">$chartdir/transports_vol.png") or die "Could not write $chartdir/transports_vol.png: $!\n"; |
| 2991 | binmode IMG; |
| 2992 | print IMG $gd->png; |
| 2993 | close IMG; |
| 2994 | print $htm_fh "<img src=\"$chartrel/transports_vol.png\">"; |
| 2995 | } |
| 2996 | } |
| 2997 | |
| 2998 | print $htm_fh "</td></tr></table>\n\n"; |
| 2999 | } |
| 3000 | } |
| 3001 | |
| 3002 | |
| 3003 | |
| 3004 | ####################################################################### |
| 3005 | # print_relay(); |
| 3006 | # |
| 3007 | # print_relay(); |
| 3008 | # |
| 3009 | # Print our totals by relay. |
| 3010 | ####################################################################### |
| 3011 | sub print_relay { |
| 3012 | my $row_print_relay=1; |
| 3013 | my $temp = "Relayed messages"; |
| 3014 | print $htm_fh "<hr><a name=\"$temp\"></a><h2>$temp</h2>\n" if $htm_fh; |
| 3015 | if (scalar(keys %relayed) > 0 || $relayed_unshown > 0) { |
| 3016 | my $shown = 0; |
| 3017 | my $spacing = ""; |
| 3018 | my $txt_format = "%7d %s\n => %s\n"; |
| 3019 | my $htm_format = "<tr><td align=\"right\">%d</td><td>%s</td><td>%s</td>\n"; |
| 3020 | |
| 3021 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; |
| 3022 | if ($htm_fh) { |
| 3023 | print $htm_fh "<table border=1>\n"; |
| 3024 | print $htm_fh "<tr><th>Count</th><th>From</th><th>To</th>\n"; |
| 3025 | } |
| 3026 | if ($xls_fh) { |
| 3027 | $ws_relayed->write($row_print_relay++, $col, $temp, $f_header2); |
| 3028 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, ["Count", "From", "To"], $f_headertab); |
| 3029 | } |
| 3030 | |
| 3031 | |
| 3032 | my($key); |
| 3033 | foreach $key (sort keys %relayed) { |
| 3034 | my $count = $relayed{$key}; |
| 3035 | $shown += $count; |
| 3036 | $key =~ s/[HA]=//g; |
| 3037 | my($one,$two) = split(/=> /, $key); |
| 3038 | my @content=($count, $one, $two); |
| 3039 | printf $txt_fh ($txt_format, @content) if $txt_fh; |
| 3040 | printf $htm_fh ($htm_format, @content) if $htm_fh; |
| 3041 | if ($xls_fh) |
| 3042 | { |
| 3043 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, \@content); |
| 3044 | } |
| 3045 | $spacing = "\n"; |
| 3046 | } |
| 3047 | |
| 3048 | print $htm_fh "</table>\n<p>\n" if $htm_fh; |
| 3049 | print $txt_fh "${spacing}Total: $shown (plus $relayed_unshown unshown)\n\n" if $txt_fh; |
| 3050 | print $htm_fh "${spacing}Total: $shown (plus $relayed_unshown unshown)\n\n" if $htm_fh; |
| 3051 | if ($xls_fh) |
| 3052 | { |
| 3053 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, [$shown, "Sum of shown" ]); |
| 3054 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, [$relayed_unshown, "unshown"]); |
| 3055 | $row_print_relay++; |
| 3056 | } |
| 3057 | } |
| 3058 | else { |
| 3059 | print $txt_fh "No relayed messages\n-------------------\n\n" if $txt_fh; |
| 3060 | print $htm_fh "No relayed messages\n\n" if $htm_fh; |
| 3061 | if ($xls_fh) |
| 3062 | { |
| 3063 | $row_print_relay++; |
| 3064 | } |
| 3065 | } |
| 3066 | } |
| 3067 | |
| 3068 | |
| 3069 | |
| 3070 | ####################################################################### |
| 3071 | # print_errors(); |
| 3072 | # |
| 3073 | # print_errors(); |
| 3074 | # |
| 3075 | # Print our errors. In HTML, we display them as a list rather than a table - |
| 3076 | # Netscape doesn't like large tables! |
| 3077 | ####################################################################### |
| 3078 | sub print_errors { |
| 3079 | my $total_errors = 0; |
| 3080 | $row=1; |
| 3081 | |
| 3082 | if (scalar(keys %errors_count) != 0) { |
| 3083 | my $temp = "List of errors"; |
| 3084 | my $htm_format = "<li>%d - %s\n"; |
| 3085 | |
| 3086 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; |
| 3087 | if ($htm_fh) { |
| 3088 | print $htm_fh "<hr><a name=\"errors\"></a><h2>$temp</h2>\n"; |
| 3089 | print $htm_fh "<ul><li><b>Count - Error</b>\n"; |
| 3090 | } |
| 3091 | if ($xls_fh) |
| 3092 | { |
| 3093 | $ws_errors->write($row++, 0, $temp, $f_header2); |
| 3094 | &set_worksheet_line($ws_errors, $row++, 0, ["Count", "Error"], $f_headertab); |
| 3095 | } |
| 3096 | |
| 3097 | |
| 3098 | my($key); |
| 3099 | foreach $key (sort keys %errors_count) { |
| 3100 | my $text = $key; |
| 3101 | chomp($text); |
| 3102 | $text =~ s/\s\s+/ /g; #Convert multiple spaces to a single space. |
| 3103 | $total_errors += $errors_count{$key}; |
| 3104 | |
| 3105 | if ($txt_fh) { |
| 3106 | printf $txt_fh ("%5d ", $errors_count{$key}); |
| 3107 | my $text_remaining = $text; |
| 3108 | while (length($text_remaining) > 65) { |
| 3109 | my($first,$rest) = $text_remaining =~ /(.{50}\S*)\s+(.+)/; |
| 3110 | last if !$first; |
| 3111 | printf $txt_fh ("%s\n\t ", $first); |
| 3112 | $text_remaining = $rest; |
| 3113 | } |
| 3114 | printf $txt_fh ("%s\n\n", $text_remaining); |
| 3115 | } |
| 3116 | |
| 3117 | if ($htm_fh) { |
| 3118 | |
| 3119 | #Translate HTML tag characters. Sergey Sholokh. |
| 3120 | $text =~ s/\</\<\;/g; |
| 3121 | $text =~ s/\>/\>\;/g; |
| 3122 | |
| 3123 | printf $htm_fh ($htm_format,$errors_count{$key},$text); |
| 3124 | } |
| 3125 | if ($xls_fh) |
| 3126 | { |
| 3127 | &set_worksheet_line($ws_errors, $row++, 0, [$errors_count{$key},$text]); |
| 3128 | } |
| 3129 | } |
| 3130 | |
| 3131 | $temp = "Errors encountered: $total_errors"; |
| 3132 | |
| 3133 | if ($txt_fh) { |
| 3134 | print $txt_fh $temp, "\n"; |
| 3135 | print $txt_fh "-" x length($temp),"\n"; |
| 3136 | } |
| 3137 | if ($htm_fh) { |
| 3138 | print $htm_fh "</ul>\n<p>\n"; |
| 3139 | print $htm_fh $temp, "\n"; |
| 3140 | } |
| 3141 | if ($xls_fh) |
| 3142 | { |
| 3143 | &set_worksheet_line($ws_errors, $row++, 0, [$total_errors, "Sum of Errors encountered"]); |
| 3144 | } |
| 3145 | } |
| 3146 | |
| 3147 | } |
| 3148 | |
| 3149 | |
| 3150 | ####################################################################### |
| 3151 | # parse_old_eximstat_reports(); |
| 3152 | # |
| 3153 | # parse_old_eximstat_reports($fh); |
| 3154 | # |
| 3155 | # Parse old eximstat output so we can merge daily stats to weekly stats and weekly to monthly etc. |
| 3156 | # |
| 3157 | # To test that the merging still works after changes, do something like the following. |
| 3158 | # All the diffs should produce no output. |
| 3159 | # |
| 3160 | # options='-bydomain -byemail -byhost -byedomain' |
| 3161 | # options="$options -show_rt1,2,4 -show_dt 1,2,4" |
| 3162 | # options="$options -pattern 'Completed Messages' /Completed/" |
| 3163 | # options="$options -pattern 'Received Messages' /<=/" |
| 3164 | # |
| 3165 | # ./eximstats $options mainlog > mainlog.txt |
| 3166 | # ./eximstats $options -merge mainlog.txt > mainlog.2.txt |
| 3167 | # diff mainlog.txt mainlog.2.txt |
| 3168 | # |
| 3169 | # ./eximstats $options -html mainlog > mainlog.html |
| 3170 | # ./eximstats $options -merge -html mainlog.txt > mainlog.2.html |
| 3171 | # diff mainlog.html mainlog.2.html |
| 3172 | # |
| 3173 | # ./eximstats $options -merge mainlog.html > mainlog.3.txt |
| 3174 | # diff mainlog.txt mainlog.3.txt |
| 3175 | # |
| 3176 | # ./eximstats $options -merge -html mainlog.html > mainlog.3.html |
| 3177 | # diff mainlog.html mainlog.3.html |
| 3178 | # |
| 3179 | # ./eximstats $options -nvr mainlog > mainlog.nvr.txt |
| 3180 | # ./eximstats $options -merge mainlog.nvr.txt > mainlog.4.txt |
| 3181 | # diff mainlog.txt mainlog.4.txt |
| 3182 | # |
| 3183 | # # double_mainlog.txt should have twice the values that mainlog.txt has. |
| 3184 | # ./eximstats $options mainlog mainlog > double_mainlog.txt |
| 3185 | ####################################################################### |
| 3186 | sub parse_old_eximstat_reports { |
| 3187 | my($fh) = @_; |
| 3188 | |
| 3189 | my(%league_table_value_entered, %league_table_value_was_zero, %table_order); |
| 3190 | |
| 3191 | my(%user_pattern_index); |
| 3192 | my $user_pattern_index = 0; |
| 3193 | map {$user_pattern_index{$_} = $user_pattern_index++} @user_descriptions; |
| 3194 | my $user_pattern_keys = join('|', @user_descriptions); |
| 3195 | |
| 3196 | while (<$fh>) { |
| 3197 | PARSE_OLD_REPORT_LINE: |
| 3198 | if (/Exim statistics from ([\d\-]+ [\d:]+(\s+[\+\-]\d+)?) to ([\d\-]+ [\d:]+(\s+[\+\-]\d+)?)/) { |
| 3199 | $begin = $1 if ($1 lt $begin); |
| 3200 | $end = $3 if ($3 gt $end); |
| 3201 | } |
| 3202 | elsif (/Grand total summary/) { |
| 3203 | # Fill in $report_totals{Received|Delivered}{Volume|Messages|Addresses|Hosts|Domains|...|Delayed|DelayedPercent|Failed|FailedPercent} |
| 3204 | my(@fields, @delivered_fields); |
| 3205 | my $doing_table = 0; |
| 3206 | while (<$fh>) { |
| 3207 | $_ = html2txt($_); #Convert general HTML markup to text. |
| 3208 | s/At least one addr//g; #Another part of the HTML output we don't want. |
| 3209 | |
| 3210 | # TOTAL Volume Messages Addresses Hosts Domains Delayed Failed |
| 3211 | # Received 26MB 237 177 23 8 3.4% 28 11.8% |
| 3212 | # Delivered 13MB 233 250 99 88 |
| 3213 | if (/TOTAL\s+(.*?)\s*$/) { |
| 3214 | $doing_table = 1; |
| 3215 | @delivered_fields = split(/\s+/,$1); |
| 3216 | |
| 3217 | #Delayed and Failed have two columns each, so add the extra field names in. |
| 3218 | splice(@delivered_fields,-1,1,'DelayedPercent','Failed','FailedPercent'); |
| 3219 | |
| 3220 | # Addresses only figure in the Delivered row, so remove them from the |
| 3221 | # normal fields. |
| 3222 | @fields = grep !/Addresses/, @delivered_fields; |
| 3223 | } |
| 3224 | elsif (/(Received)\s+(.*?)\s*$/) { |
| 3225 | print STDERR "Parsing $_" if $debug; |
| 3226 | add_to_totals($report_totals{$1},\@fields,$2); |
| 3227 | } |
| 3228 | elsif (/(Delivered)\s+(.*?)\s*$/) { |
| 3229 | print STDERR "Parsing $_" if $debug; |
| 3230 | add_to_totals($report_totals{$1},\@delivered_fields,$2); |
| 3231 | my $data = $2; |
| 3232 | # If we're merging an old report which doesn't include addresses, |
| 3233 | # then use the Messages field instead. |
| 3234 | unless (grep(/Addresses/, @delivered_fields)) { |
| 3235 | my %tmp; |
| 3236 | line_to_hash(\%tmp,\@delivered_fields,$data); |
| 3237 | add_to_totals($report_totals{Delivered},['Addresses'],$tmp{Messages}); |
| 3238 | } |
| 3239 | } |
| 3240 | elsif (/(Temp Rejects|Rejects|Ham|Spam)\s+(.*?)\s*$/) { |
| 3241 | print STDERR "Parsing $_" if $debug; |
| 3242 | add_to_totals($report_totals{$1},['Messages','Hosts'],$2); |
| 3243 | } |
| 3244 | else { |
| 3245 | last if $doing_table; |
| 3246 | } |
| 3247 | } |
| 3248 | } |
| 3249 | |
| 3250 | elsif (/User Specified Patterns/i) { |
| 3251 | #User Specified Patterns |
| 3252 | #----------------------- |
| 3253 | # Total |
| 3254 | # Description 85 |
| 3255 | |
| 3256 | while (<$fh>) { last if (/Total/); } #Wait until we get the table headers. |
| 3257 | while (<$fh>) { |
| 3258 | print STDERR "Parsing $_" if $debug; |
| 3259 | $_ = html2txt($_); #Convert general HTML markup to text. |
| 3260 | if (/^\s*(.*?)\s+(\d+)\s*$/) { |
| 3261 | $report_totals{patterns}{$1} = {} unless (defined $report_totals{patterns}{$1}); |
| 3262 | add_to_totals($report_totals{patterns}{$1},['Total'],$2); |
| 3263 | } |
| 3264 | last if (/^\s*$/); #Finished if we have a blank line. |
| 3265 | } |
| 3266 | } |
| 3267 | |
| 3268 | elsif (/(^|<h2>)($user_pattern_keys) per /o) { |
| 3269 | # Parse User defined pattern histograms if they exist. |
| 3270 | parse_histogram($fh, $user_pattern_interval_count[$user_pattern_index{$2}] ); |
| 3271 | } |
| 3272 | |
| 3273 | |
| 3274 | elsif (/Deliveries by transport/i) { |
| 3275 | #Deliveries by transport |
| 3276 | #----------------------- |
| 3277 | # Volume Messages |
| 3278 | # :blackhole: 70KB 51 |
| 3279 | # address_pipe 655KB 1 |
| 3280 | # smtp 11MB 151 |
| 3281 | |
| 3282 | while (<$fh>) { last if (/Volume/); } #Wait until we get the table headers. |
| 3283 | while (<$fh>) { |
| 3284 | print STDERR "Parsing $_" if $debug; |
| 3285 | $_ = html2txt($_); #Convert general HTML markup to text. |
| 3286 | if (/(\S+)\s+(\d+\S*\s+\d+)/) { |
| 3287 | $report_totals{transport}{$1} = {} unless (defined $report_totals{transport}{$1}); |
| 3288 | add_to_totals($report_totals{transport}{$1},['Volume','Messages'],$2); |
| 3289 | } |
| 3290 | last if (/^\s*$/); #Finished if we have a blank line. |
| 3291 | } |
| 3292 | } |
| 3293 | elsif (/Messages received per/) { |
| 3294 | parse_histogram($fh, \@received_interval_count); |
| 3295 | } |
| 3296 | elsif (/Deliveries per/) { |
| 3297 | parse_histogram($fh, \@delivered_interval_count); |
| 3298 | } |
| 3299 | |
| 3300 | #elsif (/Time spent on the queue: (all messages|messages with at least one remote delivery)/) { |
| 3301 | elsif (/(Time spent on the queue|Delivery times|Receipt times): ((\S+) messages|messages with at least one remote delivery)((<[^>]*>)*\s*)$/) { |
| 3302 | #Time spent on the queue: all messages |
| 3303 | #------------------------------------- |
| 3304 | # |
| 3305 | #Under 1m 217 91.9% 91.9% |
| 3306 | # 5m 2 0.8% 92.8% |
| 3307 | # 3h 8 3.4% 96.2% |
| 3308 | # 6h 7 3.0% 99.2% |
| 3309 | # 12h 2 0.8% 100.0% |
| 3310 | |
| 3311 | # Set a pointer to the queue bin so we can use the same code |
| 3312 | # block for both all messages and remote deliveries. |
| 3313 | #my $bin_aref = ($1 eq 'all messages') ? \@qt_all_bin : \@qt_remote_bin; |
| 3314 | my($bin_aref, $times_aref, $overflow_sref); |
| 3315 | if ($1 eq 'Time spent on the queue') { |
| 3316 | $times_aref = \@queue_times; |
| 3317 | if ($2 eq 'all messages') { |
| 3318 | $bin_aref = \@qt_all_bin; |
| 3319 | $overflow_sref = \$qt_all_overflow; |
| 3320 | } |
| 3321 | else { |
| 3322 | $bin_aref = \@qt_remote_bin; |
| 3323 | $overflow_sref = \$qt_remote_overflow; |
| 3324 | } |
| 3325 | } |
| 3326 | elsif ($1 eq 'Delivery times') { |
| 3327 | $times_aref = \@delivery_times; |
| 3328 | if ($2 eq 'all messages') { |
| 3329 | $bin_aref = \@dt_all_bin; |
| 3330 | $overflow_sref = \$dt_all_overflow; |
| 3331 | } |
| 3332 | else { |
| 3333 | $bin_aref = \@dt_remote_bin; |
| 3334 | $overflow_sref = \$dt_remote_overflow; |
| 3335 | } |
| 3336 | } |
| 3337 | else { |
| 3338 | unless (exists $rcpt_times_bin{$3}) { |
| 3339 | initialise_rcpt_times($3); |
| 3340 | } |
| 3341 | $bin_aref = $rcpt_times_bin{$3}; |
| 3342 | $times_aref = \@rcpt_times; |
| 3343 | $overflow_sref = \$rcpt_times_overflow{$3}; |
| 3344 | } |
| 3345 | |
| 3346 | |
| 3347 | my ($blank_lines, $reached_table) = (0,0); |
| 3348 | while (<$fh>) { |
| 3349 | $_ = html2txt($_); #Convert general HTML markup to text. |
| 3350 | # The table is preceded by one blank line, and has one blank line |
| 3351 | # following it. As the table may be empty, the best way to determine |
| 3352 | # that we've finished it is to look for the second blank line. |
| 3353 | ++$blank_lines if /^\s*$/; |
| 3354 | last if ($blank_lines >=2); #Finished the table ? |
| 3355 | $reached_table = 1 if (/\d/); |
| 3356 | next unless $reached_table; |
| 3357 | my $previous_seconds_on_queue = 0; |
| 3358 | if (/^\s*(Under|Over|)\s+(\d+[smhdw])\s+(\d+)/) { |
| 3359 | print STDERR "Parsing $_" if $debug; |
| 3360 | my($modifier,$formated_time,$count) = ($1,$2,$3); |
| 3361 | my $seconds = unformat_time($formated_time); |
| 3362 | my $time_on_queue = ($seconds + $previous_seconds_on_queue) / 2; |
| 3363 | $previous_seconds_on_queue = $seconds; |
| 3364 | $time_on_queue = $seconds * 2 if ($modifier eq 'Over'); |
| 3365 | my($i); |
| 3366 | for ($i = 0; $i <= $#$times_aref; $i++) { |
| 3367 | if ($time_on_queue < $times_aref->[$i]) { |
| 3368 | $$bin_aref[$i] += $count; |
| 3369 | last; |
| 3370 | } |
| 3371 | } |
| 3372 | $$overflow_sref += $count if ($i > $#$times_aref); |
| 3373 | |
| 3374 | } |
| 3375 | } |
| 3376 | } |
| 3377 | |
| 3378 | elsif (/Relayed messages/) { |
| 3379 | #Relayed messages |
| 3380 | #---------------- |
| 3381 | # |
| 3382 | # 1 addr.domain.com [1.2.3.4] a.user@domain.com |
| 3383 | # => addr2.domain2.com [5.6.7.8] a2.user2@domain2.com |
| 3384 | # |
| 3385 | #<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> |
| 3386 | |
| 3387 | my $reached_table = 0; |
| 3388 | my($count,$sender); |
| 3389 | while (<$fh>) { |
| 3390 | unless ($reached_table) { |
| 3391 | last if (/No relayed messages/); |
| 3392 | $reached_table = 1 if (/^\s*\d/ || />\d+</); |
| 3393 | next unless $reached_table; |
| 3394 | } |
| 3395 | if (/>(\d+)<.td><td>(.*?) ?<.td><td>(.*?)</) { |
| 3396 | update_relayed($1,$2,$3); |
| 3397 | } |
| 3398 | elsif (/^\s*(\d+)\s+(.*?)\s*$/) { |
| 3399 | ($count,$sender) = ($1,$2); |
| 3400 | } |
| 3401 | elsif (/=>\s+(.*?)\s*$/) { |
| 3402 | update_relayed($count,$sender,$1); |
| 3403 | } |
| 3404 | else { |
| 3405 | last; #Finished the table ? |
| 3406 | } |
| 3407 | } |
| 3408 | } |
| 3409 | |
| 3410 | elsif (/Top (.*?) by (message count|volume)/) { |
| 3411 | #Top 50 sending hosts by message count |
| 3412 | #------------------------------------- |
| 3413 | # |
| 3414 | # 48 1468KB local |
| 3415 | # Could also have average values for HTML output. |
| 3416 | # 48 1468KB 30KB local |
| 3417 | |
| 3418 | my($category,$by_count_or_volume) = ($1,$2); |
| 3419 | |
| 3420 | #As we show 2 views of each table (by count and by volume), |
| 3421 | #most (but not all) entries will appear in both tables. |
| 3422 | #Set up a hash to record which entries we have already seen |
| 3423 | #and one to record which ones we are seeing for the first time. |
| 3424 | if ($by_count_or_volume =~ /count/) { |
| 3425 | undef %league_table_value_entered; |
| 3426 | undef %league_table_value_was_zero; |
| 3427 | undef %table_order; |
| 3428 | } |
| 3429 | |
| 3430 | #As this section processes multiple different table categories, |
| 3431 | #set up pointers to the hashes to be updated. |
| 3432 | my($messages_href,$addresses_href,$data_href,$data_gigs_href); |
| 3433 | if ($category =~ /local sender/) { |
| 3434 | $messages_href = \%received_count_user; |
| 3435 | $addresses_href = undef; |
| 3436 | $data_href = \%received_data_user; |
| 3437 | $data_gigs_href = \%received_data_gigs_user; |
| 3438 | } |
| 3439 | elsif ($category =~ /sending (\S+?)s?\b/) { |
| 3440 | #Top 50 sending (host|domain|email|edomain)s |
| 3441 | #Top sending (host|domain|email|edomain) |
| 3442 | $messages_href = \%{$received_count{"\u$1"}}; |
| 3443 | $data_href = \%{$received_data{"\u$1"}}; |
| 3444 | $data_gigs_href = \%{$received_data_gigs{"\u$1"}}; |
| 3445 | } |
| 3446 | elsif ($category =~ /local destination/) { |
| 3447 | $messages_href = \%delivered_messages_user; |
| 3448 | $addresses_href = \%delivered_addresses_user; |
| 3449 | $data_href = \%delivered_data_user; |
| 3450 | $data_gigs_href = \%delivered_data_gigs_user; |
| 3451 | } |
| 3452 | elsif ($category =~ /local domain destination/) { |
| 3453 | $messages_href = \%delivered_messages_local_domain; |
| 3454 | $addresses_href = \%delivered_addresses_local_domain; |
| 3455 | $data_href = \%delivered_data_local_domain; |
| 3456 | $data_gigs_href = \%delivered_data_gigs_local_domain; |
| 3457 | } |
| 3458 | elsif ($category =~ /(\S+) destination/) { |
| 3459 | #Top 50 (host|domain|email|edomain) destinations |
| 3460 | #Top (host|domain|email|edomain) destination |
| 3461 | $messages_href = \%{$delivered_messages{"\u$1"}}; |
| 3462 | $addresses_href = \%{$delivered_addresses{"\u$1"}}; |
| 3463 | $data_href = \%{$delivered_data{"\u$1"}}; |
| 3464 | $data_gigs_href = \%{$delivered_data_gigs{"\u$1"}}; |
| 3465 | } |
| 3466 | elsif ($category =~ /temporarily rejected ips/) { |
| 3467 | $messages_href = \%temporarily_rejected_count_by_ip; |
| 3468 | } |
| 3469 | elsif ($category =~ /rejected ips/) { |
| 3470 | $messages_href = \%rejected_count_by_ip; |
| 3471 | } |
| 3472 | elsif ($category =~ /non-rejected spamming ips/) { |
| 3473 | $messages_href = \%spam_count_by_ip; |
| 3474 | } |
| 3475 | elsif ($category =~ /mail temporary rejection reasons/) { |
| 3476 | $messages_href = \%temporarily_rejected_count_by_reason; |
| 3477 | } |
| 3478 | elsif ($category =~ /mail rejection reasons/) { |
| 3479 | $messages_href = \%rejected_count_by_reason; |
| 3480 | } |
| 3481 | |
| 3482 | my $reached_table = 0; |
| 3483 | my $row_re; |
| 3484 | while (<$fh>) { |
| 3485 | # Watch out for empty tables. |
| 3486 | goto PARSE_OLD_REPORT_LINE if (/<h2>/ or (/^\s*[a-zA-Z]/ && !/^\s*Messages/)); |
| 3487 | |
| 3488 | $_ = html2txt($_); #Convert general HTML markup to text. |
| 3489 | |
| 3490 | # Messages Addresses Bytes Average |
| 3491 | if (/^\s*Messages/) { |
| 3492 | my $pattern = '^\s*(\d+)'; |
| 3493 | $pattern .= (/Addresses/) ? '\s+(\d+)' : '()'; |
| 3494 | $pattern .= (/Bytes/) ? '\s+([\dKMGB]+)' : '()'; |
| 3495 | $pattern .= (/Average/) ? '\s+[\dKMGB]+' : ''; |
| 3496 | $pattern .= '\s+(.*?)\s*$'; |
| 3497 | $row_re = qr/$pattern/; |
| 3498 | $reached_table = 1; |
| 3499 | next; |
| 3500 | } |
| 3501 | next unless $reached_table; |
| 3502 | |
| 3503 | my($messages, $addresses, $rounded_volume, $entry); |
| 3504 | |
| 3505 | if (/$row_re/) { |
| 3506 | ($messages, $addresses, $rounded_volume, $entry) = ($1, $2, $3, $4); |
| 3507 | } |
| 3508 | else { |
| 3509 | #Else we have finished the table and we may need to do some |
| 3510 | #kludging to retain the order of the entries. |
| 3511 | |
| 3512 | if ($by_count_or_volume =~ /volume/) { |
| 3513 | #Add a few bytes to appropriate entries to preserve the order. |
| 3514 | foreach $rounded_volume (keys %table_order) { |
| 3515 | #For each rounded volume, we want to create a list which has things |
| 3516 | #ordered from the volume table at the front, and additional things |
| 3517 | #from the count table ordered at the back. |
| 3518 | @{$table_order{$rounded_volume}{volume}} = () unless defined $table_order{$rounded_volume}{volume}; |
| 3519 | @{$table_order{$rounded_volume}{'message count'}} = () unless defined $table_order{$rounded_volume}{'message count'}; |
| 3520 | my(@order,%mark); |
| 3521 | map {$mark{$_} = 1} @{$table_order{$rounded_volume}{volume}}; |
| 3522 | @order = @{$table_order{$rounded_volume}{volume}}; |
| 3523 | map {push(@order,$_)} grep(!$mark{$_},@{$table_order{$rounded_volume}{'message count'}}); |
| 3524 | |
| 3525 | my $bonus_bytes = $#order; |
| 3526 | $bonus_bytes = 511 if ($bonus_bytes > 511); #Don't go over the half-K boundary! |
| 3527 | while (@order and ($bonus_bytes > 0)) { |
| 3528 | my $entry = shift(@order); |
| 3529 | if ($league_table_value_was_zero{$entry}) { |
| 3530 | $$data_href{$entry} += $bonus_bytes; |
| 3531 | print STDERR "$category by $by_count_or_volume: added $bonus_bytes bonus bytes to $entry\n" if $debug; |
| 3532 | } |
| 3533 | $bonus_bytes--; |
| 3534 | } |
| 3535 | } |
| 3536 | } |
| 3537 | last; |
| 3538 | } |
| 3539 | |
| 3540 | # Store a new table entry. |
| 3541 | |
| 3542 | # Add the entry into the %table_order hash if it has a rounded |
| 3543 | # volume (KB/MB/GB). |
| 3544 | push(@{$table_order{$rounded_volume}{$by_count_or_volume}},$entry) if ($rounded_volume =~ /\D/); |
| 3545 | |
| 3546 | unless ($league_table_value_entered{$entry}) { |
| 3547 | $league_table_value_entered{$entry} = 1; |
| 3548 | unless ($$messages_href{$entry}) { |
| 3549 | $$messages_href{$entry} = 0; |
| 3550 | $$addresses_href{$entry} = 0; |
| 3551 | $$data_href{$entry} = 0; |
| 3552 | $$data_gigs_href{$entry} = 0; |
| 3553 | $league_table_value_was_zero{$entry} = 1; |
| 3554 | } |
| 3555 | |
| 3556 | $$messages_href{$entry} += $messages; |
| 3557 | |
| 3558 | # When adding the addresses, be aware that we could be merging |
| 3559 | # an old report which does not include addresses. In this case, |
| 3560 | # we add the messages instead. |
| 3561 | $$addresses_href{$entry} += ($addresses) ? $addresses : $messages; |
| 3562 | |
| 3563 | #Add the rounded value to the data and data_gigs hashes. |
| 3564 | un_round($rounded_volume,\$$data_href{$entry},\$$data_gigs_href{$entry}) if $rounded_volume; |
| 3565 | print STDERR "$category by $by_count_or_volume: added $messages,$rounded_volume to $entry\n" if $debug; |
| 3566 | } |
| 3567 | |
| 3568 | } |
| 3569 | } |
| 3570 | elsif (/List of errors/) { |
| 3571 | #List of errors |
| 3572 | #-------------- |
| 3573 | # |
| 3574 | # 1 07904931641@one2one.net R=external T=smtp: SMTP error |
| 3575 | # from remote mailer after RCPT TO:<07904931641@one2one.net>: |
| 3576 | # host mail.one2one.net [193.133.192.24]: 550 User unknown |
| 3577 | # |
| 3578 | #<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> |
| 3579 | |
| 3580 | |
| 3581 | my $reached_table = 0; |
| 3582 | my($count,$error,$blanks); |
| 3583 | while (<$fh>) { |
| 3584 | $reached_table = 1 if (/^( *|<li>)(\d+)/); |
| 3585 | next unless $reached_table; |
| 3586 | |
| 3587 | s/^<li>(\d+) -/$1/; #Convert an HTML line to a text line. |
| 3588 | $_ = html2txt($_); #Convert general HTML markup to text. |
| 3589 | |
| 3590 | if (/\t\s*(.*)/) { |
| 3591 | $error .= ' ' . $1; #Join a multiline error. |
| 3592 | } |
| 3593 | elsif (/^\s*(\d+)\s+(.*)/) { |
| 3594 | if ($error) { |
| 3595 | #Finished with a previous multiline error so save it. |
| 3596 | $errors_count{$error} = 0 unless $errors_count{$error}; |
| 3597 | $errors_count{$error} += $count; |
| 3598 | } |
| 3599 | ($count,$error) = ($1,$2); |
| 3600 | } |
| 3601 | elsif (/Errors encountered/) { |
| 3602 | if ($error) { |
| 3603 | #Finished the section, so save our stored last error. |
| 3604 | $errors_count{$error} = 0 unless $errors_count{$error}; |
| 3605 | $errors_count{$error} += $count; |
| 3606 | } |
| 3607 | last; |
| 3608 | } |
| 3609 | } |
| 3610 | } |
| 3611 | |
| 3612 | } |
| 3613 | } |
| 3614 | |
| 3615 | ####################################################################### |
| 3616 | # parse_histogram($fh, \@delivered_interval_count); |
| 3617 | # Parse a histogram into the provided array of counters. |
| 3618 | ####################################################################### |
| 3619 | sub parse_histogram { |
| 3620 | my($fh, $counters_aref) = @_; |
| 3621 | |
| 3622 | # Messages received per hour (each dot is 2 messages) |
| 3623 | #--------------------------------------------------- |
| 3624 | # |
| 3625 | #00-01 106 ..................................................... |
| 3626 | #01-02 103 ................................................... |
| 3627 | |
| 3628 | my $reached_table = 0; |
| 3629 | while (<$fh>) { |
| 3630 | $reached_table = 1 if (/^00/); |
| 3631 | next unless $reached_table; |
| 3632 | print STDERR "Parsing $_" if $debug; |
| 3633 | if (/^(\d+):(\d+)\s+(\d+)/) { #hh:mm start time format ? |
| 3634 | $$counters_aref[($1*60 + $2)/$hist_interval] += $3 if $hist_opt; |
| 3635 | } |
| 3636 | elsif (/^(\d+)-(\d+)\s+(\d+)/) { #hh-hh start-end time format ? |
| 3637 | $$counters_aref[($1*60)/$hist_interval] += $3 if $hist_opt; |
| 3638 | } |
| 3639 | else { #Finished the table ? |
| 3640 | last; |
| 3641 | } |
| 3642 | } |
| 3643 | } |
| 3644 | |
| 3645 | |
| 3646 | ####################################################################### |
| 3647 | # update_relayed(); |
| 3648 | # |
| 3649 | # update_relayed($count,$sender,$recipient); |
| 3650 | # |
| 3651 | # Adds an entry into the %relayed hash. Currently only used when |
| 3652 | # merging reports. |
| 3653 | ####################################################################### |
| 3654 | sub update_relayed { |
| 3655 | my($count,$sender,$recipient) = @_; |
| 3656 | |
| 3657 | #When generating the key, put in the 'H=' and 'A=' which can be used |
| 3658 | #in searches. |
| 3659 | my $key = "H=$sender => H=$recipient"; |
| 3660 | $key =~ s/ ([^=\s]+\@\S+|<>)/ A=$1/g; |
| 3661 | if (!defined $relay_pattern || $key !~ /$relay_pattern/o) { |
| 3662 | $relayed{$key} = 0 if !defined $relayed{$key}; |
| 3663 | $relayed{$key} += $count; |
| 3664 | } |
| 3665 | else { |
| 3666 | $relayed_unshown += $count; |
| 3667 | } |
| 3668 | } |
| 3669 | |
| 3670 | |
| 3671 | ####################################################################### |
| 3672 | # add_to_totals(); |
| 3673 | # |
| 3674 | # add_to_totals(\%totals,\@keys,$values); |
| 3675 | # |
| 3676 | # Given a line of space seperated values, add them into the provided hash using @keys |
| 3677 | # as the hash keys. |
| 3678 | # |
| 3679 | # If the value contains a '%', then the value is set rather than added. Otherwise, we |
| 3680 | # convert the value to bytes and gigs. The gigs get added to I<Key>-gigs. |
| 3681 | ####################################################################### |
| 3682 | sub add_to_totals { |
| 3683 | my($totals_href,$keys_aref,$values) = @_; |
| 3684 | my(@values) = split(/\s+/,$values); |
| 3685 | |
| 3686 | for(my $i = 0; $i < @values && $i < @$keys_aref; ++$i) { |
| 3687 | my $key = $keys_aref->[$i]; |
| 3688 | if ($values[$i] =~ /%/) { |
| 3689 | $$totals_href{$key} = $values[$i]; |
| 3690 | } |
| 3691 | else { |
| 3692 | $$totals_href{$key} = 0 unless ($$totals_href{$key}); |
| 3693 | $$totals_href{"$key-gigs"} = 0 unless ($$totals_href{"$key-gigs"}); |
| 3694 | un_round($values[$i], \$$totals_href{$key}, \$$totals_href{"$key-gigs"}); |
| 3695 | print STDERR "Added $values[$i] to $key - $$totals_href{$key} , " . $$totals_href{"$key-gigs"} . "GB.\n" if $debug; |
| 3696 | } |
| 3697 | } |
| 3698 | } |
| 3699 | |
| 3700 | |
| 3701 | ####################################################################### |
| 3702 | # line_to_hash(); |
| 3703 | # |
| 3704 | # line_to_hash(\%hash,\@keys,$line); |
| 3705 | # |
| 3706 | # Given a line of space seperated values, set them into the provided hash |
| 3707 | # using @keys as the hash keys. |
| 3708 | ####################################################################### |
| 3709 | sub line_to_hash { |
| 3710 | my($href,$keys_aref,$values) = @_; |
| 3711 | my(@values) = split(/\s+/,$values); |
| 3712 | for(my $i = 0; $i < @values && $i < @$keys_aref; ++$i) { |
| 3713 | $$href{$keys_aref->[$i]} = $values[$i]; |
| 3714 | } |
| 3715 | } |
| 3716 | |
| 3717 | |
| 3718 | ####################################################################### |
| 3719 | # get_report_total(); |
| 3720 | # |
| 3721 | # $total = get_report_total(\%hash,$key); |
| 3722 | # |
| 3723 | # If %hash contains values split into Units and Gigs, we calculate and return |
| 3724 | # |
| 3725 | # $hash{$key} + 1024*1024*1024 * $hash{"${key}-gigs"} |
| 3726 | ####################################################################### |
| 3727 | sub get_report_total { |
| 3728 | no integer; |
| 3729 | my($hash_ref,$key) = @_; |
| 3730 | if ($$hash_ref{"${key}-gigs"}) { |
| 3731 | return $$hash_ref{$key} + $gig * $$hash_ref{"${key}-gigs"}; |
| 3732 | } |
| 3733 | return $$hash_ref{$key} || 0; |
| 3734 | } |
| 3735 | |
| 3736 | ####################################################################### |
| 3737 | # html2txt(); |
| 3738 | # |
| 3739 | # $text_line = html2txt($html_line); |
| 3740 | # |
| 3741 | # Convert a line from html to text. Currently we just convert HTML tags to spaces |
| 3742 | # and convert >, <, and tags back. |
| 3743 | ####################################################################### |
| 3744 | sub html2txt { |
| 3745 | ($_) = @_; |
| 3746 | |
| 3747 | # Convert HTML tags to spacing. Note that the reports may contain <Userid> and |
| 3748 | # <Userid@Domain> words, so explicitly specify the HTML tags we will remove |
| 3749 | # (the ones used by this program). If someone is careless enough to have their |
| 3750 | # Userid the same as an HTML tag, there's not much we can do about it. |
| 3751 | s/<\/?(html|head|title|body|h\d|ul|li|a\s+|table|tr|td|th|pre|hr|p|br)\b.*?>/ /g; |
| 3752 | |
| 3753 | s/\<\;/\</og; #Convert '<' to '<'. |
| 3754 | s/\>\;/\>/og; #Convert '>' to '>'. |
| 3755 | s/\ \;/ /og; #Convert ' ' to ' '. |
| 3756 | return($_); |
| 3757 | } |
| 3758 | |
| 3759 | ####################################################################### |
| 3760 | # get_next_arg(); |
| 3761 | # |
| 3762 | # $arg = get_next_arg(); |
| 3763 | # |
| 3764 | # Because eximstats arguments are often passed as variables, |
| 3765 | # we can't rely on shell parsing to deal with quotes. This |
| 3766 | # subroutine returns $ARGV[1] and does a shift. If $ARGV[1] |
| 3767 | # starts with a quote (' or "), and doesn't end in one, then |
| 3768 | # we append the next argument to it and shift again. We repeat |
| 3769 | # until we've got all of the argument. |
| 3770 | # |
| 3771 | # This isn't perfect as all white space gets reduced to one space, |
| 3772 | # but it's as good as we can get! If it's esential that spacing |
| 3773 | # be preserved precisely, then you get that by not using shell |
| 3774 | # variables. |
| 3775 | ####################################################################### |
| 3776 | sub get_next_arg { |
| 3777 | my $arg = ''; |
| 3778 | my $matched_pattern = 0; |
| 3779 | while ($ARGV[1]) { |
| 3780 | $arg .= ' ' if $arg; |
| 3781 | $arg .= $ARGV[1]; shift(@ARGV); |
| 3782 | if ($arg !~ /^['"]/) { |
| 3783 | $matched_pattern = 1; |
| 3784 | last; |
| 3785 | } |
| 3786 | if ($arg =~ s/^(['"])(.*)\1$/$2/) { |
| 3787 | $matched_pattern = 1; |
| 3788 | last; |
| 3789 | } |
| 3790 | } |
| 3791 | die "Mismatched argument quotes - <$arg>.\n" unless $matched_pattern; |
| 3792 | return $arg; |
| 3793 | } |
| 3794 | |
| 3795 | ####################################################################### |
| 3796 | # set_worksheet_line($ws_global, $startrow, $startcol, \@content, $format); |
| 3797 | # |
| 3798 | # set values to a sequence of cells in a row. |
| 3799 | # |
| 3800 | ####################################################################### |
| 3801 | sub set_worksheet_line { |
| 3802 | my ($worksheet, $row, $col, $content, $format) = @_; |
| 3803 | |
| 3804 | foreach my $token (@$content) |
| 3805 | { |
| 3806 | $worksheet->write($row, $col++, $token, $format ); |
| 3807 | } |
| 3808 | |
| 3809 | } |
| 3810 | |
| 3811 | ####################################################################### |
| 3812 | # @rcpt_times = parse_time_list($string); |
| 3813 | # |
| 3814 | # Parse a comma seperated list of time values in seconds given by |
| 3815 | # the user and fill an array. |
| 3816 | # |
| 3817 | # Return a default list if $string is undefined. |
| 3818 | # Return () if $string eq '0'. |
| 3819 | ####################################################################### |
| 3820 | sub parse_time_list { |
| 3821 | my($string) = @_; |
| 3822 | if (! defined $string) { |
| 3823 | return(60, 5*60, 15*60, 30*60, 60*60, 3*60*60, 6*60*60, 12*60*60, 24*60*60); |
| 3824 | } |
| 3825 | my(@times) = split(/,/, $string); |
| 3826 | foreach my $q (@times) { $q = eval($q) + 0 } |
| 3827 | @times = sort { $a <=> $b } @times; |
| 3828 | @times = () if ($#times == 0 && $times[0] == 0); |
| 3829 | return(@times); |
| 3830 | } |
| 3831 | |
| 3832 | |
| 3833 | ####################################################################### |
| 3834 | # initialise_rcpt_times($protocol); |
| 3835 | # Initialise an array of rcpt_times to 0 for the specified protocol. |
| 3836 | ####################################################################### |
| 3837 | sub initialise_rcpt_times { |
| 3838 | my($protocol) = @_; |
| 3839 | for (my $i = 0; $i <= $#rcpt_times; ++$i) { |
| 3840 | $rcpt_times_bin{$protocol}[$i] = 0; |
| 3841 | } |
| 3842 | $rcpt_times_overflow{$protocol} = 0; |
| 3843 | } |
| 3844 | |
| 3845 | |
| 3846 | ################################################## |
| 3847 | # Main Program # |
| 3848 | ################################################## |
| 3849 | |
| 3850 | |
| 3851 | $last_timestamp = ''; |
| 3852 | $last_date = ''; |
| 3853 | $show_errors = 1; |
| 3854 | $show_relay = 1; |
| 3855 | $show_transport = 1; |
| 3856 | $topcount = 50; |
| 3857 | $local_league_table = 1; |
| 3858 | $include_remote_users = 0; |
| 3859 | $include_original_destination = 0; |
| 3860 | $hist_opt = 1; |
| 3861 | $volume_rounding = 1; |
| 3862 | $localtime_offset = calculate_localtime_offset(); # PH/FANF |
| 3863 | |
| 3864 | $charts = 0; |
| 3865 | $charts_option_specified = 0; |
| 3866 | $chartrel = "."; |
| 3867 | $chartdir = "."; |
| 3868 | |
| 3869 | @queue_times = parse_time_list(); |
| 3870 | @rcpt_times = (); |
| 3871 | @delivery_times = (); |
| 3872 | |
| 3873 | $last_offset = ''; |
| 3874 | $offset_seconds = 0; |
| 3875 | |
| 3876 | $row=1; |
| 3877 | $col=0; |
| 3878 | $col_hist=0; |
| 3879 | $run_hist=0; |
| 3880 | my(%output_files); # What output files have been specified? |
| 3881 | |
| 3882 | # Decode options |
| 3883 | |
| 3884 | while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') { |
| 3885 | if ($ARGV[0] =~ /^\-h(\d+)$/) { $hist_opt = $1 } |
| 3886 | elsif ($ARGV[0] =~ /^\-ne$/) { $show_errors = 0 } |
| 3887 | elsif ($ARGV[0] =~ /^\-nr(.?)(.*)\1$/) { |
| 3888 | if ($1 eq "") { $show_relay = 0 } else { $relay_pattern = $2 } |
| 3889 | } |
| 3890 | elsif ($ARGV[0] =~ /^\-q([,\d\+\-\*\/]+)$/) { @queue_times = parse_time_list($1) } |
| 3891 | elsif ($ARGV[0] =~ /^-nt$/) { $show_transport = 0 } |
| 3892 | elsif ($ARGV[0] =~ /^\-nt(.?)(.*)\1$/) |
| 3893 | { |
| 3894 | if ($1 eq "") { $show_transport = 0 } else { $transport_pattern = $2 } |
| 3895 | } |
| 3896 | elsif ($ARGV[0] =~ /^-t(\d+)$/) { $topcount = $1 } |
| 3897 | elsif ($ARGV[0] =~ /^-tnl$/) { $local_league_table = 0 } |
| 3898 | elsif ($ARGV[0] =~ /^-txt=?(\S*)$/) { $txt_fh = get_filehandle($1,\%output_files) } |
| 3899 | elsif ($ARGV[0] =~ /^-html=?(\S*)$/) { $htm_fh = get_filehandle($1,\%output_files) } |
| 3900 | elsif ($ARGV[0] =~ /^-xls=?(\S*)$/) { |
| 3901 | if ($HAVE_Spreadsheet_WriteExcel) { |
| 3902 | $xls_fh = get_filehandle($1,\%output_files); |
| 3903 | } |
| 3904 | else { |
| 3905 | warn "WARNING: CPAN Module Spreadsheet::WriteExcel not installed. Obtain from www.cpan.org\n"; |
| 3906 | } |
| 3907 | } |
| 3908 | elsif ($ARGV[0] =~ /^-merge$/) { $merge_reports = 1 } |
| 3909 | elsif ($ARGV[0] =~ /^-charts$/) { |
| 3910 | $charts = 1; |
| 3911 | warn "WARNING: CPAN Module GD::Graph::pie not installed. Obtain from www.cpan.org\n" unless $HAVE_GD_Graph_pie; |
| 3912 | warn "WARNING: CPAN Module GD::Graph::linespoints not installed. Obtain from www.cpan.org\n" unless $HAVE_GD_Graph_linespoints; |
| 3913 | } |
| 3914 | elsif ($ARGV[0] =~ /^-chartdir$/) { $chartdir = $ARGV[1]; shift; $charts_option_specified = 1; } |
| 3915 | elsif ($ARGV[0] =~ /^-chartrel$/) { $chartrel = $ARGV[1]; shift; $charts_option_specified = 1; } |
| 3916 | elsif ($ARGV[0] =~ /^-include_original_destination$/) { $include_original_destination = 1 } |
| 3917 | elsif ($ARGV[0] =~ /^-cache$/) { } #Not currently used. |
| 3918 | elsif ($ARGV[0] =~ /^-byhost$/) { $do_sender{Host} = 1 } |
| 3919 | elsif ($ARGV[0] =~ /^-bydomain$/) { $do_sender{Domain} = 1 } |
| 3920 | elsif ($ARGV[0] =~ /^-byemail$/) { $do_sender{Email} = 1 } |
| 3921 | elsif ($ARGV[0] =~ /^-byemaildomain$/) { $do_sender{Edomain} = 1 } |
| 3922 | elsif ($ARGV[0] =~ /^-byedomain$/) { $do_sender{Edomain} = 1 } |
| 3923 | elsif ($ARGV[0] =~ /^-bylocaldomain$/) { $do_local_domain = 1 } |
| 3924 | elsif ($ARGV[0] =~ /^-emptyok$/) { $emptyOK = 1 } |
| 3925 | elsif ($ARGV[0] =~ /^-nvr$/) { $volume_rounding = 0 } |
| 3926 | elsif ($ARGV[0] =~ /^-show_rt([,\d\+\-\*\/]+)?$/) { @rcpt_times = parse_time_list($1) } |
| 3927 | elsif ($ARGV[0] =~ /^-show_dt([,\d\+\-\*\/]+)?$/) { @delivery_times = parse_time_list($1) } |
| 3928 | elsif ($ARGV[0] =~ /^-d$/) { $debug = 1 } |
| 3929 | elsif ($ARGV[0] =~ /^--?h(elp)?$/){ help() } |
| 3930 | elsif ($ARGV[0] =~ /^-t_remote_users$/) { $include_remote_users = 1 } |
| 3931 | elsif ($ARGV[0] =~ /^-pattern$/) |
| 3932 | { |
| 3933 | push(@user_descriptions,get_next_arg()); |
| 3934 | push(@user_patterns,get_next_arg()); |
| 3935 | } |
| 3936 | elsif ($ARGV[0] =~ /^-utc$/) |
| 3937 | { |
| 3938 | # We don't need this value if the log is in UTC. |
| 3939 | $localtime_offset = undef; |
| 3940 | } |
| 3941 | else |
| 3942 | { |
| 3943 | print STDERR "Eximstats: Unknown or malformed option $ARGV[0]\n"; |
| 3944 | help(); |
| 3945 | } |
| 3946 | shift; |
| 3947 | } |
| 3948 | |
| 3949 | # keep old default behaviour |
| 3950 | if (! ($xls_fh or $htm_fh or $txt_fh)) { |
| 3951 | $txt_fh = \*STDOUT; |
| 3952 | } |
| 3953 | |
| 3954 | # Check that all the charts options are specified. |
| 3955 | warn "-charts option not specified. Use -help for help.\n" if ($charts_option_specified && ! $charts); |
| 3956 | |
| 3957 | # Default to display tables by sending Host. |
| 3958 | $do_sender{Host} = 1 unless ($do_sender{Domain} || $do_sender{Email} || $do_sender{Edomain}); |
| 3959 | |
| 3960 | # prepare xls Excel Workbook |
| 3961 | if (defined $xls_fh) { |
| 3962 | |
| 3963 | # Create a new Excel workbook |
| 3964 | $workbook = Spreadsheet::WriteExcel->new($xls_fh); |
| 3965 | |
| 3966 | # Add worksheets |
| 3967 | $ws_global = $workbook->addworksheet('Exim Statistik'); |
| 3968 | # show $ws_global as initial sheet |
| 3969 | $ws_global->set_first_sheet(); |
| 3970 | $ws_global->activate(); |
| 3971 | |
| 3972 | if ($show_relay) { |
| 3973 | $ws_relayed = $workbook->addworksheet('Relayed Messages'); |
| 3974 | $ws_relayed->set_column(1, 2, 80); |
| 3975 | } |
| 3976 | if ($show_errors) { |
| 3977 | $ws_errors = $workbook->addworksheet('Errors'); |
| 3978 | } |
| 3979 | |
| 3980 | |
| 3981 | # set column widths |
| 3982 | $ws_global->set_column(0, 2, 20); # Columns B-D width set to 30 |
| 3983 | $ws_global->set_column(3, 3, 15); # Columns B-D width set to 30 |
| 3984 | $ws_global->set_column(4, 4, 25); # Columns B-D width set to 30 |
| 3985 | |
| 3986 | # Define Formats |
| 3987 | $f_default = $workbook->add_format(); |
| 3988 | |
| 3989 | $f_header1 = $workbook->add_format(); |
| 3990 | $f_header1->set_bold(); |
| 3991 | #$f_header1->set_color('red'); |
| 3992 | $f_header1->set_size('15'); |
| 3993 | $f_header1->set_valign(); |
| 3994 | # $f_header1->set_align('center'); |
| 3995 | # $ws_global->write($row++, 2, "Testing Headers 1", $f_header1); |
| 3996 | |
| 3997 | $f_header2 = $workbook->add_format(); |
| 3998 | $f_header2->set_bold(); |
| 3999 | $f_header2->set_size('12'); |
| 4000 | $f_header2->set_valign(); |
| 4001 | # $ws_global->write($row++, 2, "Testing Headers 2", $f_header2); |
| 4002 | |
| 4003 | # Create another header2 for use in merged cells. |
| 4004 | $f_header2_m = $workbook->add_format(); |
| 4005 | $f_header2_m->set_bold(); |
| 4006 | $f_header2_m->set_size('8'); |
| 4007 | $f_header2_m->set_valign(); |
| 4008 | $f_header2_m->set_align('center'); |
| 4009 | |
| 4010 | $f_percent = $workbook->add_format(); |
| 4011 | $f_percent->set_num_format('0.0%'); |
| 4012 | |
| 4013 | $f_headertab = $workbook->add_format(); |
| 4014 | $f_headertab->set_bold(); |
| 4015 | $f_headertab->set_valign(); |
| 4016 | # $ws_global->write($row++, 2, "Testing Headers tab", $f_headertab); |
| 4017 | |
| 4018 | } |
| 4019 | |
| 4020 | |
| 4021 | # Initialise the queue/delivery/rcpt time counters. |
| 4022 | for (my $i = 0; $i <= $#queue_times; $i++) { |
| 4023 | $qt_all_bin[$i] = 0; |
| 4024 | $qt_remote_bin[$i] = 0; |
| 4025 | } |
| 4026 | for (my $i = 0; $i <= $#delivery_times; $i++) { |
| 4027 | $dt_all_bin[$i] = 0; |
| 4028 | $dt_remote_bin[$i] = 0; |
| 4029 | } |
| 4030 | initialise_rcpt_times('all'); |
| 4031 | |
| 4032 | |
| 4033 | # Compute the number of slots for the histogram |
| 4034 | if ($hist_opt > 0) |
| 4035 | { |
| 4036 | if ($hist_opt > 60 || 60 % $hist_opt != 0) |
| 4037 | { |
| 4038 | print STDERR "Eximstats: -h must specify a factor of 60\n"; |
| 4039 | exit 1; |
| 4040 | } |
| 4041 | $hist_interval = 60/$hist_opt; #Interval in minutes. |
| 4042 | $hist_number = (24*60)/$hist_interval; #Number of intervals per day. |
| 4043 | @received_interval_count = (0) x $hist_number; |
| 4044 | @delivered_interval_count = (0) x $hist_number; |
| 4045 | my $user_pattern_index = 0; |
| 4046 | for (my $user_pattern_index = 0; $user_pattern_index <= $#user_patterns; ++$user_pattern_index) { |
| 4047 | @{$user_pattern_interval_count[$user_pattern_index]} = (0) x $hist_number; |
| 4048 | } |
| 4049 | @dt_all_bin = (0) x $hist_number; |
| 4050 | @dt_remote_bin = (0) x $hist_number; |
| 4051 | } |
| 4052 | |
| 4053 | #$queue_unknown = 0; |
| 4054 | |
| 4055 | $total_received_data = 0; |
| 4056 | $total_received_data_gigs = 0; |
| 4057 | $total_received_count = 0; |
| 4058 | |
| 4059 | $total_delivered_data = 0; |
| 4060 | $total_delivered_data_gigs = 0; |
| 4061 | $total_delivered_messages = 0; |
| 4062 | $total_delivered_addresses = 0; |
| 4063 | |
| 4064 | $qt_all_overflow = 0; |
| 4065 | $qt_remote_overflow = 0; |
| 4066 | $dt_all_overflow = 0; |
| 4067 | $dt_remote_overflow = 0; |
| 4068 | $delayed_count = 0; |
| 4069 | $relayed_unshown = 0; |
| 4070 | $message_errors = 0; |
| 4071 | $begin = "9999-99-99 99:99:99"; |
| 4072 | $end = "0000-00-00 00:00:00"; |
| 4073 | my($section,$type); |
| 4074 | foreach $section ('Received','Delivered','Temp Rejects', 'Rejects','Ham','Spam') { |
| 4075 | foreach $type ('Volume','Messages','Delayed','Failed','Hosts','Domains','Emails','Edomains') { |
| 4076 | $report_totals{$section}{$type} = 0; |
| 4077 | } |
| 4078 | } |
| 4079 | |
| 4080 | # Generate our parser. |
| 4081 | my $parser = generate_parser(); |
| 4082 | |
| 4083 | |
| 4084 | |
| 4085 | if (@ARGV) { |
| 4086 | # Scan the input files and collect the data |
| 4087 | foreach my $file (@ARGV) { |
| 4088 | if ($file =~ /\.gz/) { |
| 4089 | unless (open(FILE,"gunzip -c $file |")) { |
| 4090 | print STDERR "Failed to gunzip -c $file: $!"; |
| 4091 | next; |
| 4092 | } |
| 4093 | } |
| 4094 | elsif ($file =~ /\.Z/) { |
| 4095 | unless (open(FILE,"uncompress -c $file |")) { |
| 4096 | print STDERR "Failed to uncompress -c $file: $!"; |
| 4097 | next; |
| 4098 | } |
| 4099 | } |
| 4100 | else { |
| 4101 | unless (open(FILE,$file)) { |
| 4102 | print STDERR "Failed to read $file: $!"; |
| 4103 | next; |
| 4104 | } |
| 4105 | } |
| 4106 | #Now parse the filehandle, updating the global variables. |
| 4107 | parse($parser,\*FILE); |
| 4108 | close FILE; |
| 4109 | } |
| 4110 | } |
| 4111 | else { |
| 4112 | #No files provided. Parse STDIN, updating the global variables. |
| 4113 | parse($parser,\*STDIN); |
| 4114 | } |
| 4115 | |
| 4116 | |
| 4117 | if ($begin eq "9999-99-99 99:99:99" && ! $emptyOK) { |
| 4118 | print STDERR "**** No valid log lines read\n"; |
| 4119 | exit 1; |
| 4120 | } |
| 4121 | |
| 4122 | # Output our results. |
| 4123 | print_header(); |
| 4124 | print_grandtotals(); |
| 4125 | |
| 4126 | # Print counts of user specified patterns if required. |
| 4127 | print_user_patterns() if @user_patterns; |
| 4128 | |
| 4129 | # Print rejection reasons. |
| 4130 | # print_rejects(); |
| 4131 | |
| 4132 | # Print totals by transport if required. |
| 4133 | print_transport() if $show_transport; |
| 4134 | |
| 4135 | # Print the deliveries per interval as a histogram, unless configured not to. |
| 4136 | # First find the maximum in one interval and scale accordingly. |
| 4137 | if ($hist_opt > 0) { |
| 4138 | print_histogram("Messages received", 'message', @received_interval_count); |
| 4139 | print_histogram("Deliveries", 'delivery', @delivered_interval_count); |
| 4140 | } |
| 4141 | |
| 4142 | # Print times on queue if required. |
| 4143 | if ($#queue_times >= 0) { |
| 4144 | print_duration_table("Time spent on the queue", "all messages", \@queue_times, \@qt_all_bin,$qt_all_overflow); |
| 4145 | print_duration_table("Time spent on the queue", "messages with at least one remote delivery", \@queue_times, \@qt_remote_bin,$qt_remote_overflow); |
| 4146 | } |
| 4147 | |
| 4148 | # Print delivery times if required. |
| 4149 | if ($#delivery_times >= 0) { |
| 4150 | print_duration_table("Delivery times", "all messages", \@delivery_times, \@dt_all_bin,$dt_all_overflow); |
| 4151 | print_duration_table("Delivery times", "messages with at least one remote delivery", \@delivery_times, \@dt_remote_bin,$dt_remote_overflow); |
| 4152 | } |
| 4153 | |
| 4154 | # Print rcpt times if required. |
| 4155 | if ($#rcpt_times >= 0) { |
| 4156 | foreach my $protocol ('all', grep(!/^all$/, sort keys %rcpt_times_bin)) { |
| 4157 | print_duration_table("Receipt times", "$protocol messages", \@rcpt_times, $rcpt_times_bin{$protocol}, $rcpt_times_overflow{$protocol}); |
| 4158 | } |
| 4159 | } |
| 4160 | |
| 4161 | # Print relay information if required. |
| 4162 | print_relay() if $show_relay; |
| 4163 | |
| 4164 | # Print the league tables, if topcount isn't zero. |
| 4165 | if ($topcount > 0) { |
| 4166 | my($ws_rej, $ws_top50, $ws_rej_row, $ws_top50_row, $ws_temp_rej, $ws_temp_rej_row); |
| 4167 | $ws_rej_row = $ws_temp_rej_row = $ws_top50_row = 0; |
| 4168 | if ($xls_fh) { |
| 4169 | $ws_top50 = $workbook->addworksheet('Deliveries'); |
| 4170 | $ws_rej = $workbook->addworksheet('Rejections') if (%rejected_count_by_reason || %rejected_count_by_ip || %spam_count_by_ip); |
| 4171 | $ws_temp_rej = $workbook->addworksheet('Temporary Rejections') if (%temporarily_rejected_count_by_reason || %temporarily_rejected_count_by_ip); |
| 4172 | } |
| 4173 | |
| 4174 | print_league_table("mail rejection reason", \%rejected_count_by_reason, undef, undef, undef, $ws_rej, \$ws_rej_row) if %rejected_count_by_reason; |
| 4175 | 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; |
| 4176 | |
| 4177 | foreach ('Host','Domain','Email','Edomain') { |
| 4178 | next unless $do_sender{$_}; |
| 4179 | print_league_table("sending \l$_", $received_count{$_}, undef, $received_data{$_},$received_data_gigs{$_}, $ws_top50, \$ws_top50_row); |
| 4180 | } |
| 4181 | |
| 4182 | print_league_table("local sender", \%received_count_user, undef, |
| 4183 | \%received_data_user,\%received_data_gigs_user, $ws_top50, \$ws_top50_row) if (($local_league_table || $include_remote_users) && %received_count_user); |
| 4184 | foreach ('Host','Domain','Email','Edomain') { |
| 4185 | next unless $do_sender{$_}; |
| 4186 | print_league_table("\l$_ destination", $delivered_messages{$_}, $delivered_addresses{$_}, $delivered_data{$_},$delivered_data_gigs{$_}, $ws_top50, \$ws_top50_row); |
| 4187 | } |
| 4188 | 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); |
| 4189 | 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); |
| 4190 | |
| 4191 | print_league_table("rejected ip", \%rejected_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %rejected_count_by_ip; |
| 4192 | 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; |
| 4193 | print_league_table("non-rejected spamming ip", \%spam_count_by_ip, undef, undef, undef, $ws_rej, \$ws_rej_row) if %spam_count_by_ip; |
| 4194 | |
| 4195 | } |
| 4196 | |
| 4197 | # Print the error statistics if required. |
| 4198 | print_errors() if $show_errors; |
| 4199 | |
| 4200 | print $htm_fh "</body>\n</html>\n" if $htm_fh; |
| 4201 | |
| 4202 | |
| 4203 | $txt_fh->close if $txt_fh && ref $txt_fh; |
| 4204 | $htm_fh->close if $htm_fh; |
| 4205 | |
| 4206 | if ($xls_fh) { |
| 4207 | # close Excel Workbook |
| 4208 | $ws_global->set_first_sheet(); |
| 4209 | # FIXME: whyever - activate does not work :-/ |
| 4210 | $ws_global->activate(); |
| 4211 | $workbook->close(); |
| 4212 | } |
| 4213 | |
| 4214 | |
| 4215 | # End of eximstats |