| 1 | 1 | simandl | HotSaNIC |
| 2 | | | HTML overview to System and Network Information Center |
| 3 | | | by Bernd Pissny ( hotsanic@bernisys.prima.de / http://www.bernisys.prima.de ) |
| 4 | | | |
| 5 | | | ---------------------------------------------------------------------- |
| 6 | | | |
| 7 | | | HotSaNIC version history |
| 8 | | | |
| 9 | | | 0.5.0 - (pre-released) o completely new concept! The directory structure |
| 10 | | | new features: has been changed a lot, all modules are kept in a |
| 11 | | | solaris support subdirectory now. The modules now run as single |
| 12 | | | more *BSD support daemons, the main script uses signalling |
| 13 | | | bugfixes techniques to control the modules. This saves a |
| 14 | | | lot of compilation time on every SAMPLE call. To |
| 15 | | | avoid too many modules running parallel, a random |
| 16 | | | scheduler has been added. |
| 17 | | | o modules now have a ".description" file which will |
| 18 | | | be printed when doing the main setup. |
| 19 | | | o all perl scripts now end with ".pl" |
| 20 | | | o the SYSTEM module is now configurable to enable or |
| 21 | | | disable some of its specific targets. New targets |
| 22 | | | "interrupts" and "uptime" have been added. Basic |
| 23 | | | OS X support (testing). MultiCPU systems fully |
| 24 | | | supported. swap stats are now stored in mem.rrd, |
| 25 | | | loadavg.rrd has been renamed to load.rrd. New |
| 26 | | | method gathering mem and swap info on NetBSD. |
| 27 | | | Comments have been added for later addition of |
| 28 | | | kernel >2.4 support. |
| 29 | | | o the protocol used by the PING module is now |
| 30 | | | configurable. You can also configure it to use the |
| 31 | | | ping shellcommand instead of Net::Ping. The |
| 32 | | | minimum number of pings to be sent is 5 now. |
| 33 | | | Timing will be done by Time::HiRes (if available) |
| 34 | | | and does a fallback to the previous sysdcall. If |
| 35 | | | no HiRes timers are available, you will get an |
| 36 | | | error message. |
| 37 | | | For better comparison of diagrams, the initial |
| 38 | | | upper border can be configured. The lower border |
| 39 | | | has been adapted for logarithmic scale. |
| 40 | | | Data will be stored in seconds now instead of |
| 41 | | | milliseconds. |
| 42 | | | The module should generate no more zombies now. |
| 43 | | | o SENSORS now has FreeBSD support |
| 44 | | | o TRAFFIC module now with OS X support. Devices are |
| 45 | | | no longer configured multiple times. Taken down |
| 46 | | | BSD interfaces handled correctly. Don't use DNS |
| 47 | | | lookups anymore when using netstat. BSD setup |
| 48 | | | patched. |
| 49 | | | o APPS now handles applications with "." in filename |
| 50 | | | correctly |
| 51 | | | o DISKIO now with basic kernel 2.4 device |
| 52 | | | autodetection. Some excludes applied in devicelist |
| 53 | | | (fd* md*). Module seems to worn on FreeBSD now. |
| 54 | | | Some comments have been added for later |
| 55 | | | implementation of new Kernel 2.4 |
| 56 | | | "/proc/partitions" support. |
| 57 | | | o PART module device detection corrected. Automatic |
| 58 | | | excludes list enhanced. Table sorted by |
| 59 | | | devicename. |
| 60 | | | o Within each module each Operating system will be |
| 61 | | | handled by an own library in the |
| 62 | | | "./modules/<modname>/platform/" directory. There's |
| 63 | | | also a lib "common.pm" that handles all common |
| 64 | | | functions and a fallback library "default.pm" |
| 65 | | | o HotSaNIC.pm is obsolete now, but is still there to |
| 66 | | | support modules written on older versions. To |
| 67 | | | adapt older modules to new concept, there's a |
| 68 | | | wrapper script available under "./tools/" |
| 69 | | | o all global libraries from HotSaNIC.pm were split |
| 70 | | | into different files, these are kept in the |
| 71 | | | "./lib/" directory |
| 72 | | | o all global functions in the read-data.pl scripts |
| 73 | | | have benn moved into the new HotSaNICmod.pm |
| 74 | | | library. read-data.pl just initializes the module |
| 75 | | | - the main loops, help texts etc are in the |
| 76 | | | library. |
| 77 | | | o all read-data.pl scripts support cmdline |
| 78 | | | parameters similar to sysV-style |
| 79 | | | (start/stop/status ...) |
| 80 | | | o modules' configuration can be reloaded while |
| 81 | | | running |
| 82 | | | o SOLARIS is now supported by some Modules: apcusv, |
| 83 | | | apps, ping, system (processes, CPU, load, users, |
| 84 | | | swap, mem). |
| 85 | | | o all diagram scripts are converted to PERL now |
| 86 | | | o main index now has multi-row support (mainly used |
| 87 | | | for MultiCPU machines and SNMP targets) |
| 88 | | | o instead of using the shell-magic cookie (#! ...) |
| 89 | | | directly, we use "#!/usr/bin/env ..." to avoid |
| 90 | | | conflicts on some systems. This makes the Automake |
| 91 | | | system obsolete (saves us work...). |
| 92 | | | o a lot more SNMP support had been addeed - all SNMP |
| 93 | | | related core-functions are in the |
| 94 | | | "./lib/HotSaNICsnmp.pm" file. These should be |
| 95 | | | quite compatible with quite many SNMP daemons. If |
| 96 | | | Net::SNMP isn't found, the system does a fallback |
| 97 | | | to the shell binaries. (snmpget, snmpwalk, |
| 98 | | | snmpbulkget, snmpbulkwalk). |
| 99 | | | o fixed bug which produces GIFs instead of PNGs |
| 100 | | | regardless of settings |
| 101 | | | o all logfiles moved to the new "./var/log/" |
| 102 | | | directory, PID files went in "./var/run/", all |
| 103 | | | settings are linked in "./var/settings/" by the |
| 104 | | | main setup script. Modules store their temp. data |
| 105 | | | in "./var/modules/<modname>/" |
| 106 | | | o main setup.pl script shows usage info when called |
| 107 | | | with parameter "?" |
| 108 | | | o setup will initially guess CONVERTMETHOD now |
| 109 | | | o some diagram-titles look a bit nicer now |
| 110 | | | o some files were moved and renamed, so when |
| 111 | | | upgrading it may be necessary to re-run the main |
| 112 | | | "setup.pl -g" to re-generate the rrdgraph script. |
| 113 | | | o some vital paths will be created automatically now |
| 114 | | | o all "use diagnostics" and "use warnings" were |
| 115 | | | moved to the libraries, since once loaded they act |
| 116 | | | globally in PERL. |
| 117 | | | o for better readability in system tools like "ps" |
| 118 | | | or "top", the module name will be passed as |
| 119 | | | cmdline parameter. |
| 120 | | | o new PID handling functions implemented |
| 121 | | | o some shell-calls have been converted to PERL |
| 122 | | | syntax to avoid backticks. |
| 123 | | | o the main diagrams script has been converted to |
| 124 | | | PERL. It now generates some console-output when |
| 125 | | | running( some people were confused by the previous |
| 126 | | | behaviour of it running silently in the |
| 127 | | | background). The logfile will now be handled by |
| 128 | | | rrdtimer. |
| 129 | | | o modules are automatically restarted every STIME if |
| 130 | | | not running |
| 131 | | | o rrdtimer.pl uses an anlternate daemonizing method |
| 132 | | | to avoid shells hanging while logging out (mostly |
| 133 | | | works!) |
| 134 | | | sending a HUP to rrdtimer, the modules will also |
| 135 | | | be terminated, sending USR2 only the modules will |
| 136 | | | be terminated (and will be restarted after STIME) |
| 137 | | | o main loop now sleeps more efficiently to use less |
| 138 | | | CPU time |
| 139 | | | o added more debugging verbosity to scheduler |
| 140 | | | o added MAILQ module to show stats for the Postfix |
| 141 | | | mailserver's queue items |
| 142 | | | o added a library to handle user interactions |
| 143 | | | o started to convert setup scripts to perl |
| 144 | | | o added NETSTAT module to show stats for open TCP |
| 145 | | | sockets |
| 146 | | | Version 0.5.0-pre3 available as .tgz package |
| 147 | | | Other versions are only available via anonymous CVS. |
| 148 | | | |
| 149 | | | -------------------------------------------------- |
| 150 | | | 0.4.1 - (unreleased) o data-diskio generates more associative filenames |
| 151 | | | new features (Thanx Michael Renner for the valuable hints!) |
| 152 | | | bugfixes o data-diskio now supports *BSD |
| 153 | | | o data-networks now has totals for input/output and |
| 154 | | | a summary over the graph's time |
| 155 | | | o data-traffic now has a summary over the graph's |
| 156 | | | time |
| 157 | | | o HotSaNIC.pm fixed, the module name was not |
| 158 | | | detected correctly if HotSaNIC resided in subdirs |
| 159 | | | containing "-" chars! |
| 160 | | | There are also some comments added to (briefly) |
| 161 | | | explain each function's usage. |
| 162 | | | o data-traffic trouble with alias interfaces like |
| 163 | | | eth0:0 fixed |
| 164 | | | o data-ping now has a configurable maximum of |
| 165 | | | parallel processes (settings: PARALLEL="4") |
| 166 | | | o data-shoutcast is a new module to monitor user |
| 167 | | | activities on shoutcast servers |
| 168 | | | o data-diskio diagrams script converted to PERL |
| 169 | | | o rrdtimer now logs errors to the main logfile |
| 170 | | | (makes debugging a bit easier) |
| 171 | | | o data-part samples configured drives only (more |
| 172 | | | convenient on systems using autofs) |
| 173 | | | It also checks if NFS partitions are configured |
| 174 | | | and scans them on demand. This reduces deadlocks. |
| 175 | | | o data-worms now skips .gz and .bz2 files while |
| 176 | | | scanning logs |
| 177 | | | o data-traffic has been adapted to fix the "ghost |
| 178 | | | graphs" issue when graphing SNMP sources (hope it |
| 179 | | | works now) |
| 180 | | | o data-traffic graphs interfaces which are "down" as |
| 181 | | | "unknown" instead of a zero line. |
| 182 | | | All modules now contain dupe-control to check if a |
| 183 | | | process is already running or hangs. (processes |
| 184 | | | running >1h will be killed) |
| 185 | | | |
| 186 | | | IMPORTANT NOTICE FOR UPGRADERS: |
| 187 | | | The database in the diskio module has changed! - You |
| 188 | | | will need to run the update script in the module's |
| 189 | | | subdir in order to adapt the databases and the |
| 190 | | | webpage. Otherwhise you will get some missing images, |
| 191 | | | dead links and strangely high I/O-values or maybe in |
| 192 | | | some cases a zero-line. |
| 193 | | | This version is only available via anonymous CVS. |
| 194 | | | |
| 195 | | | -------------------------------------------------- |
| 196 | | | 0.4.0 - 20020108 All features of the work version 0.3.6 are included in |
| 197 | | | new features 0.4.0 plus: |
| 198 | | | extensions o data-traffic has been speed-up a bit |
| 199 | | | bugfixes o dupe-control checks for PIDs now |
| 200 | | | o project's contact address has changed to: |
| 201 | | | hotsanic@bernisys.prima.de |
| 202 | | | IMPORTANT NOTICE FOR UPGRADERS: |
| 203 | | | The database in the traffic module has changed! - You |
| 204 | | | will need to run the update script in the module's |
| 205 | | | subdir in order to adapt the databases. Otherwhise you |
| 206 | | | will get strangely high traffic values or maybe in |
| 207 | | | some cases a zero-line. |
| 208 | | | |
| 209 | | | -------------------------------------------------- |
| 210 | | | 0.3.6 - 20011217 o The main settings now contain a switch to select |
| 211 | | | new features the format of the generated images (.gif or .png). |
| 212 | | | extensions If you already have built own modules, please take |
| 213 | | | bugfixes care that they use the new feature ;) |
| 214 | | | o All graphs now have create-timestamps in their |
| 215 | | | titles |
| 216 | | | o Bugfix in dnet module: OGR stats will no longer |
| 217 | | | accident'ly be logged as DES. |
| 218 | | | o Bugfix in ping module: makeindex.pl does no longer |
| 219 | | | generate dead links. |
| 220 | | | o The ping module now does a statistic analysis of |
| 221 | | | its results to suppress annoying peaks. |
| 222 | | | o New module: data-networks which uses |
| 223 | | | iptables-accounting to let you know where your |
| 224 | | | traffic goes to/comes from |
| 225 | | | o data-traffic now supports Open/Free BSD ( Thanks |
| 226 | | | to Matt Burke! ) |
| 227 | | | o data-traffic now supports SNMP (both GetBulk and |
| 228 | | | SNMPwalk method) |
| 229 | | | o data-traffic's initial scales are set down to 1000 |
| 230 | | | bytes/s for any kind of device (better readable |
| 231 | | | graph at low traffic) |
| 232 | | | o data-traffic's diagram script has been upgraded to |
| 233 | | | evaluate settings instead of executimg the |
| 234 | | | settings-file |
| 235 | | | o data-traffic now can show a bits/s legend instead |
| 236 | | | of bytes/s only |
| 237 | | | o data-network now can show a bits/s legend instead |
| 238 | | | of bytes/s only |
| 239 | | | o data-part now supports Open/Free BSD ( Thanks to |
| 240 | | | Matt Burke! ) - the resolution is now 1024 bytes. |
| 241 | | | o data-dnet now supports configurable logfile names |
| 242 | | | ;) |
| 243 | | | o HotSaNIC.pm (a perl library containing common |
| 244 | | | functions) now exists to eliminate multiple |
| 245 | | | definitions and redundant functions in different |
| 246 | | | modules |
| 247 | | | o all initializing stuff has been moved from |
| 248 | | | "rrdgraph" to "rrdtimer" |
| 249 | | | o rrdtimer now has a signal-handler for the TERM |
| 250 | | | signal to properly shut down the daemon |
| 251 | | | o the whole tool now should be CVS-proof ;) |
| 252 | | | o some "setup" scripts now exist to create initial |
| 253 | | | configurations easier. They are called |
| 254 | | | automatically by the main "setup.pl" script. |
| 255 | | | o "compile" now supports option "r" for recompiling |
| 256 | | | (might be necessary after upgrading) |
| 257 | | | o a new module "data-diskio" exisits which reports |
| 258 | | | I/O stats of the pyhsically installed harddrives |
| 259 | | | o data-worms has been rewritten completely because |
| 260 | | | it wasn't sampling correctly at all. Should be ok |
| 261 | | | by now. |
| 262 | | | o data-system now has BSD-support ( Thanks to Matt |
| 263 | | | Burke! ) |
| 264 | | | o data-system now supports sampling CPU stats for |
| 265 | | | multi-CPU machines (images are being generated, |
| 266 | | | but the index.html will not yet be created - But |
| 267 | | | i'm working on it...) - Thanks Aide Florent for |
| 268 | | | the ideas and some support! |
| 269 | | | o data-system main CPU stats may be generated in |
| 270 | | | percentage instead of absolute values |
| 271 | | | o all modules can generate a browser auto-refresh |
| 272 | | | tag now. |
| 273 | | | o all thumbnail images are moved to the |
| 274 | | | corrosponding subdirs on the webpage to re-gain |
| 275 | | | more overview. |
| 276 | | | o HotSaNIC.pm now contains a parser function for the |
| 277 | | | settings. All modules have been adapted to use |
| 278 | | | this function instead of parsing the lines on |
| 279 | | | their own (read-data.pl and makeindex.pl are |
| 280 | | | affected). |
| 281 | | | o added a directory "tools" where a few helpers |
| 282 | | | could be located. |
| 283 | | | o tools: makeperlheaders (call if ping module |
| 284 | | | complains about missing syscall.ph) |
| 285 | | | o stop function in "rrdgraph" script is now |
| 286 | | | BSD-proof :) ( Thanx to Peter Reich! ) |
| 287 | | | o tools: warnings.pl generates warnings if a defined |
| 288 | | | threshold matches (for example 20KB/s on eth0 over |
| 289 | | | the last 5 minutes) |
| 290 | | | o data-traffic: diagram script converted to perl now |
| 291 | | | o data-system: shows user-stats now |
| 292 | | | o all modules: main options are being transfered to |
| 293 | | | the read-data scripts by rrdtimer as |
| 294 | | | commandline-arguments (saves parsing of mein |
| 295 | | | settings over and over for some scripts) |
| 296 | | | o setup.pl: option "y" tells the script that it |
| 297 | | | shall use all available modules, settings are |
| 298 | | | being backed-up instead of creating a settings.new |
| 299 | | | file and leaving the original untouched |
| 300 | | | o rrdtimer: debugging mode also available in |
| 301 | | | daemon-mode (configured in main settings) |
| 302 | | | o all modules: dupe control added. If a module is |
| 303 | | | already running, don't start it again until the |
| 304 | | | old instance exits (control-file: "running.dat"). |
| 305 | | | o data-system: now with NetBSD support |
| 306 | | | o data-system: showing processes on FreeBSD |
| 307 | | | plus: many many little bugfixes i forgot to mention |
| 308 | | | here ;) |
| 309 | | | |
| 310 | | | -------------------------------------------------- |
| 311 | | | 0.3.5 - 20010923 Bugfixes: |
| 312 | | | bugfixes o Missing directories in the module-subdirs are now |
| 313 | | | new module being created as they should be. Sorry for the |
| 314 | | | inconvenience! |
| 315 | | | o I corrected some misbehavior with directories |
| 316 | | | containing spaces. (correction may not affect |
| 317 | | | every module yet - remember that the project is |
| 318 | | | still in ALPHA state) |
| 319 | | | enhanced logging - timestamps are now present in some |
| 320 | | | modules and logfiles get rotated when they reach the |
| 321 | | | configured size. (at the moment only one backup is |
| 322 | | | kept - cyclic backups will be implemented soon.) |
| 323 | | | A new module "data-worms"exists which counts the |
| 324 | | | accesses caused by worm-activities in your |
| 325 | | | apache-logfile. |
| 326 | | | |
| 327 | | | -------------------------------------------------- |
| 328 | | | 0.3.4 - 20010918 I have extended "rrdtimer" a lot! |
| 329 | | | rrdtimer o The script now has many sub-routines which will |
| 330 | | | make debugging much easier - well, at least i hope |
| 331 | | | so ;). |
| 332 | | | o The main "makeindex.pl" now has been integrated |
| 333 | | | into "rrdtimer" to get the whole thing more |
| 334 | | | compact. To generate the html index-files just |
| 335 | | | type "rrdtimer i" - that's all. I left the |
| 336 | | | "makeindex.pl" in this directory for |
| 337 | | | compatibility, but this is just a shellscript that |
| 338 | | | calls "rrdtimer i". ;) |
| 339 | | | o The way you call "rrdtimer" has changed a bit - |
| 340 | | | just type "rrdtimer" on the console to get some |
| 341 | | | help. |
| 342 | | | Soon "rrdtimer" will be extended to replace "diagrams" |
| 343 | | | and "convert.sh" as well. |
| 344 | | | |
| 345 | | | -------------------------------------------------- |
| 346 | | | 0.3.3 - 20010916 I found a proper name for my tool: From now on it |
| 347 | | | bugfixes: shall be called "HotSaNIC" which stands for "HTML |
| 348 | | | rrdtimer overview to System and Network Information Center". |
| 349 | | | On execution "rrdtimer" checked for existing daemons |
| 350 | | | using the wrong program-name (rrdgraph instead of |
| 351 | | | rrdtimer). |
| 352 | | | "rrdtimer" doesn't any longer just execute |
| 353 | | | "read-data", but also checks for uncompiled |
| 354 | | | "read-data.pl" scripts, too - while "read-data" has |
| 355 | | | the highest priority. |
| 356 | | | |
| 357 | | | -------------------------------------------------- |
| 358 | | | 0.3.2 - 20010915 I accidently left a debug-line "rrdtimer" script which |
| 359 | | | bugfixes: produces unnecessary output to logfiles. |
| 360 | | | rrdtimer The "makeindex.pl" in the sensors-module had to be |
| 361 | | | sensors:makeindex fixed to generate html indexes for each sensor as |
| 362 | | | well. |
| 363 | | | I checked every script for mix-ups of integer and |
| 364 | | | string comparison. There shouldn't be any errors by |
| 365 | | | now. |
| 366 | | | |
| 367 | | | -------------------------------------------------- |
| 368 | | | 0.3.1 - 20010911 ATTENTION! |
| 369 | | | bugfixes: There is a rather serious BUG in versions < 0.3.1 ! |
| 370 | | | rrdtimer Please upgrade to version 0.3.1. |
| 371 | | | ping:makeindex |
| 372 | | | Symptom: If you restart the daemon it may happen |
| 373 | | | that no modules are queried and thus your |
| 374 | | | databases will not be updated. |
| 375 | | | |
| 376 | | | Cause: This is because i mistakenly used "le" to |
| 377 | | | compare two integer values instead of |
| 378 | | | using "<=" under PERL. Sinc e i initially |
| 379 | | | compare "120" with a timer-value which |
| 380 | | | begins with "1000", PERL states that "120" |
| 381 | | | is GREATER than "1000.*" - which is |
| 382 | | | absolutely correct in a string context. |
| 383 | | | Thanks to Bodo, Peter and Robert for the |
| 384 | | | hint that made up my mind. :) |
| 385 | | | |
| 386 | | | Cure: You may prevent this "overflow" by editing |
| 387 | | | the "rrdtimer" script and simply |
| 388 | | | exchanging all occurences of " le " by |
| 389 | | | " <= ". |
| 390 | | | |
| 391 | | | Fix: I have fixed this in version 0.3.1 and to |
| 392 | | | be on the safe side i initialise "$last" |
| 393 | | | with "$now-100" and "$lastscan" with |
| 394 | | | "$now-1000" instead of zero. That should |
| 395 | | | do it for a really really long time now |
| 396 | | | ;). |
| 397 | | | |
| 398 | | | The "makeindex.pl" in the ping-module had to be fixed |
| 399 | | | a bit. Now it generates index-files properly again. |
| 400 | | | Some bugs may occure since i will change the script's |
| 401 | | | syntaxes and variables a bit to make it possible in |
| 402 | | | future to create a library with global functions which |
| 403 | | | are commonly used in the modules' scripts. |
| 404 | | | |
| 405 | | | -------------------------------------------------- |
| 406 | | | 0.3.0 - 20010909 Added a module that pings a list of hosts, stores the |
| 407 | | | almost all ;) results in databases and creates nice graphs out of it |
| 408 | | | ;) |
| 409 | | | The sensors module has been (almost ...) completely |
| 410 | | | rewritten to give you full support to all |
| 411 | | | sensor-types. BUT! You have to know how these sensors |
| 412 | | | are represented in the /proc filesystem and how they |
| 413 | | | have to be treated properly. The module just helps to |
| 414 | | | gather the data. WHICH it shall gather ... well, |
| 415 | | | that's about up to you. But if you have some idea what |
| 416 | | | /proc is and where sensors are located it should be no |
| 417 | | | trouble at all to configure the module. Hint: some |
| 418 | | | sensors need proper scaling! This is (poorly) |
| 419 | | | described in the settings file. You should read the |
| 420 | | | lm_sensors howto to get the idea of what's happening |
| 421 | | | and put the correct scaling-values in the settigs. |
| 422 | | | Remember: Just use plain floating point numbers, not |
| 423 | | | fractions like in lm_sensors! |
| 424 | | | The partition-plugin has been altered to support NFS |
| 425 | | | volumes (and other) as well. To achieve this the |
| 426 | | | "settings" file had to be adapted and the names of the |
| 427 | | | databases have changed. Simply rename the old |
| 428 | | | databases to keep your old recordings. |
| 429 | | | I also changed the whole behaviour a bit: |
| 430 | | | ALL CONFIGURATION OPTIONS for MODULES have been moved |
| 431 | | | into the module-directories !!! |
| 432 | | | This makes it a bit easier to install or remove |
| 433 | | | modules without fiddling with the main config, WHICH |
| 434 | | | STILL RESIDES IN THE MAIN DIRECTORY! |
| 435 | | | The main "makeindex" script has been converted to perl |
| 436 | | | to let it run faster (and it sure is MUCH faster now!) |
| 437 | | | The "compile" script now can uncompile all scripts |
| 438 | | | using the "u" option |
| 439 | | | I began to convert some of the modules' "makeindex" |
| 440 | | | scripts to perl to make them run faster |
| 441 | | | |
| 442 | | | -------------------------------------------------- |
| 443 | | | 0.2.6 - 20010805 Added a module that gives you very primitive |
| 444 | | | (obsolete!) lm_sensors support (just the gl518sm sensors for now) |
| 445 | | | modules It will sure take some time to get these to work |
| 446 | | | properly... |
| 447 | | | The swap-space readings in the system-module are |
| 448 | | | extended by a "max usage" line which plots over the |
| 449 | | | whole thing to give you a hint about the "real" usage. |
| 450 | | | This maybe helps a bit in structuring your system. |
| 451 | | | |
| 452 | | | -------------------------------------------------- |
| 453 | | | 0.2.5 - 20010710 I changed the index-generation scripts so they |
| 454 | | | modules generate valid html 3.2 ... |
| 455 | | | |
| 456 | | | -------------------------------------------------- |
| 457 | | | 0.2.4 - 20010607 The "clearall" script has now an option to remove |
| 458 | | | diagrams precompiled binaries created by "compile" |
| 459 | | | clearall |
| 460 | | | "diagrams" now checks if it has been called before and |
| 461 | | | if maybe it's stalled somehow and kills it's evil |
| 462 | | | brother ;) |
| 463 | | | |
| 464 | | | -------------------------------------------------- |
| 465 | | | 0.2.3 - 20010605 Previously "rrdtimer" called all existing modules |
| 466 | | | rrdtimer parallel, but this behaviour may cause problems on |
| 467 | | | modules some systems, so i changed the default behaviour to |
| 468 | | | call all modules in a sequence. The parallel mode is |
| 469 | | | still available and can be activated by calling |
| 470 | | | "rrdtimer p" |
| 471 | | | |
| 472 | | | An additional module called "networks" has been |
| 473 | | | created. It uses ipchains to measure the traffic to |
| 474 | | | dedicated networks and evaluates these traffic samples |
| 475 | | | pretty much like the "traffic" module - except min/max |
| 476 | | | plots which would be quite a thing to implement |
| 477 | | | properly. Maybe this will come some day as an option, |
| 478 | | | but i don't think so right now. |
| 479 | | | A speedup has been added: |
| 480 | | | |
| 481 | | | Try to run the "compile" script to compile all modules |
| 482 | | | via "perlcc". This may give you a major speedup ( |
| 483 | | | about 2-3x ) because the perl modules don't have to be |
| 484 | | | compiled over and over again before they are executed |
| 485 | | | ( this happens each 10 seconds ) - but be aware that |
| 486 | | | errors may occur if some important perl-libraries are |
| 487 | | | missing! In that case you have got to live with it and |
| 488 | | | reconstruct the "read-data" scripts in each module |
| 489 | | | directory. |
| 490 | | | |
| 491 | | | I changed the global behaviour a bit. "rrdtimer" now |
| 492 | | | calls the diagram scripts itself - so cron-entries |
| 493 | | | aren't needed anymore. |
| 494 | | | |
| 495 | | | The modules now generate timestamps in the |
| 496 | | | "rrdtimer.err" logfile. |
| 497 | | | |
| 498 | | | The name of the PID-file is now free configurable in |
| 499 | | | the settings file. |
| 500 | | | |
| 501 | | | -------------------------------------------------- |
| 502 | | | 0.2.2 - 20010514 Debugging mode in "rrdtimer" now only shows 4 digits |
| 503 | | | rrdtimer after decimal point for better readability. |
| 504 | | | databases |
| 505 | | | system All databases have been moved to a subdirectory "rrd/" |
| 506 | | | in each module-subdirectory. So please move the .rrd |
| 507 | | | files into each module's subdir to the corrosponding |
| 508 | | | <module-dir>/rrd/ subdir. |
| 509 | | | |
| 510 | | | Additional logging-functionality for system-module. |
| 511 | | | Now the loadavg values for the 5min and 15min average |
| 512 | | | are logged as well. The process-stats show an |
| 513 | | | additional value "disc" representing processes being |
| 514 | | | swapped to disc at this time. |
| 515 | | | |
| 516 | | | IMPORTANT NOTICE FOR UPGRADERS: |
| 517 | | | Be sure to run "update.pl" after you installed the new |
| 518 | | | data-system module! The system.rrd database has to be |
| 519 | | | split to add the new logging-channels! Another |
| 520 | | | possibility is to just let the new scripts run and |
| 521 | | | remove system.rrd, but all previously sampled data |
| 522 | | | will be lost by doing so! |
| 523 | | | |
| 524 | | | -------------------------------------------------- |
| 525 | | | 0.2.1 - 20010510 Some bugs were fixed in the traffic, dnet and part |
| 526 | | | traffic, dnet, part modules: |
| 527 | | | convert o the "diagrams" script in the traffic-module |
| 528 | | | settings produced some errors due to a wrong |
| 529 | | | variable-assignment. |
| 530 | | | o the "diagrams" script in the part module produced |
| 531 | | | errors because "rrdtool" was called with some |
| 532 | | | parameters mixed up. |
| 533 | | | o the convert.sh script only converted the graphics |
| 534 | | | from system and dnet. Now it handles all graphics. |
| 535 | | | I played a bit with the settings-file and some of the |
| 536 | | | scripts. Now there is a configuration-entry to set |
| 537 | | | where logfiles will be stored. |
| 538 | | | Then i altered the behaviour of the DNET-module. Now |
| 539 | | | you can configure networks and their names to receive |
| 540 | | | better readable statistics for dynamic IPs. This is |
| 541 | | | done in the "data-dnet" subdir using a file called |
| 542 | | | "networks.cf" |
| 543 | | | Some speed-ups were achieved by using hashes to store |
| 544 | | | multiline values and by changing file-access |
| 545 | | | behaviours a bit. Saved about 10% execution-time :) |
| 546 | | | |
| 547 | | | -------------------------------------------------- |
| 548 | | | 0.2.0 - 20010509 Well, this is something revolutionary ;) |
| 549 | | | rrdtimer I had the time to read the PERL-FAQs and found some |
| 550 | | | dnet interesting aspects like daemonizing (i.e. fork into |
| 551 | | | background) and other stuff. So I decided to |
| 552 | | | completely rewrite the main script. This is no longer |
| 553 | | | bash, but plain PERL! Included are some features like |
| 554 | | | a debug-mode with different debug-levels to get your |
| 555 | | | modules straight. Just experiment with it if you like. |
| 556 | | | :) |
| 557 | | | Some changes were made to the dnet module as well. |
| 558 | | | I've rewritten the routines into PERL now and i hope |
| 559 | | | they will do what they are supposed to. I don't quite |
| 560 | | | know, so please inform me if anything goes wrong |
| 561 | | | there. |
| 562 | | | |
| 563 | | | -------------------------------------------------- |
| 564 | | | 0.1.9 - 20010506 I had some spare-time to spend, so while I was |
| 565 | | | dnet chatting on ICQ, I decided to give myself a little |
| 566 | | | push and have almost completely re-written the core of |
| 567 | | | the DNET module in PERL. Speed-up seems to be a factor |
| 568 | | | of about 10, though this was not really needed since |
| 569 | | | the module gets called only once a day anyway. But |
| 570 | | | major speed-up has been achieved in building the |
| 571 | | | initial databases which will reach back a couple of |
| 572 | | | months ... these are about 400 (for each host !) |
| 573 | | | database updates sequentially and this took quite a |
| 574 | | | while back in BASH ;) |
| 575 | | | I also discovered and removed a little bug in this |
| 576 | | | module. When a host has not sent any block the last |
| 577 | | | day, then the database would have taken the values of |
| 578 | | | the day before. This looks rather impressive in the |
| 579 | | | statistics, but is of course dramatically wrong :) |
| 580 | | | Ok, these stats are just looking fine right now. |
| 581 | | | |
| 582 | | | -------------------------------------------------- |
| 583 | | | 0.1.8 - 20010421 some changes were made in the "diagrams" scripts. Now |
| 584 | | | diagrams the diagrams are created in "interlace" gif mode, |
| 585 | | | traffic which will improve the load-behaviour of the webpage. |
| 586 | | | I also tuned in some additional "rrdtool"-features - |
| 587 | | | now the byte-oriented graphs have a legend on the |
| 588 | | | Y-axis and 1K is calculated by 1024 and not 1000 as |
| 589 | | | before. |
| 590 | | | The traffic module now can cope with asymmetric |
| 591 | | | interfaces as well, the MAX option in the |
| 592 | | | settings-file has a new syntax to set in and out speed |
| 593 | | | independant (of course as before in bits/s) |
| 594 | | | |
| 595 | | | -------------------------------------------------- |
| 596 | | | 0.1.7 - 20010408 further little speed-up was archieved by not using so |
| 597 | | | traffic many temp-files in the traffic-plugin. |
| 598 | | | |
| 599 | | | -------------------------------------------------- |
| 600 | | | 0.1.6 - 20010405 A little speed-up was archieved by doing some changes |
| 601 | | | read-data in the "read-data" scripts for the system and the |
| 602 | | | settings partition plug-in. Some awful shell-calls were |
| 603 | | | replaced by PERL calls which do not write in |
| 604 | | | sample-files any more, but directly into the |
| 605 | | | PERL-array :). |
| 606 | | | Also some new setting-file entries exist now to tell |
| 607 | | | the program which interfaces/devices are to sample and |
| 608 | | | which stat-graphs are to be bulit. |
| 609 | | | However, this applies only for the sample-part of the |
| 610 | | | traffic and the partition section for now... but i'm |
| 611 | | | working on it ... ;) |
| 612 | | | |
| 613 | | | -------------------------------------------------- |
| 614 | | | 0.1.5 - 20001219 Corrected a little bug in the "makeindex" files. |
| 615 | | | makeindex Weekly stats indexpage will now be built correctly. |
| 616 | | | The "makeindex" script now removes all temporary files |
| 617 | | | it creates. It also doesn't nag about existing |
| 618 | | | directories, it just silently creates them if needed. |
| 619 | | | |
| 620 | | | -------------------------------------------------- |
| 621 | | | 0.1.4 - 20001215 Corrected a missing settings-call in the "makerrd" |
| 622 | | | makerrd scripts which caused the tool to create no databases |
| 623 | | | at all. |
| 624 | | | Thanks to Markus Bajohr! |
| 625 | | | |
| 626 | | | -------------------------------------------------- |
| 627 | | | 0.1.3 - 20001209 I just corrected a bug in the part-plugin which caused |
| 628 | | | part the script to create numerous databases with no |
| 629 | | | further use, but a lot of load while building the |
| 630 | | | graphs ... |
| 631 | | | Thanks to Michael Neumann! |
| 632 | | | |
| 633 | | | -------------------------------------------------- |
| 634 | | | pre 0.1.3 I report changes and bugs since version 0.1.2 - all |
| 635 | | | earlier versions were just like the beginning of all |
| 636 | | | life ;) |
| 637 | | | |
| 638 | | | o.K. let's see. What can "rrdgraph" do right now ? |
| 639 | | | |
| 640 | | | The system-plugin ( This was my first plugin. Actually |
| 641 | | | it hasn't been a plugin up to 0.0.1 ) will collect |
| 642 | | | data about: |
| 643 | | | o processes (sleeping, running, stopped, zombie) |
| 644 | | | o CPU time (idle, nice, user, system) |
| 645 | | | o loadaverage |
| 646 | | | o memory (free, buffered, shared, cached) |
| 647 | | | o swapfile (free, used) |
| 648 | | | The traffic-plugin will collect data about all your |
| 649 | | | local network interfaces except loopback. |
| 650 | | | |
| 651 | | | The partitions-plugin collects data about your |
| 652 | | | partition-usage. |
| 653 | | | Some bugs have to be fixed here! |
| 654 | | | For example network-mounted directories will be saved |
| 655 | | | under just a part of their name. This may cause |
| 656 | | | serious problems if there are two similar paths... I |
| 657 | | | haven't checked yet, but I guess it will. |
| 658 | | | For now I will override this bug by filtering out all |
| 659 | | | network-mounted filesystems. Later i'll try to convert |
| 660 | | | the "/" in the device-pathnames to "_" and hopefully |
| 661 | | | all will be fine in future... |
| 662 | | | |
| 663 | | | -------------------------------------------------- |
| 664 | | | |
| 665 | | | ---------------------------------------------------------------------- |
