Version: 1.0.10.1 (2007-07-11) - Bugfix: PHP 4.3.2 math compensation Under certain circumstances, on PHP 4.3.2, the computation value was wrong when using the *= operator; apparantly this is fixed in later PHP versions, so the *= was removed and replaced with "var = var * .." - Bugfix: VI3 space in Virtual Machine name If there was a space in the Virtual Machine Name, then the RRD would fail to be created; this has been fixed now, spaces (as well as other characters) are now escaped; - Bugfix: RRD with space in it Fixed RRD code which broke if you had any RRD with space in it. This has been corrected and fixed. - added about unnoc page for display.php that will show all stats of the daemon/cron exec and show the current status of what unnoc is doing - updated disk SQL backend code - updated host checking code If a host is already down, then it does not go through the full process of making sure the host is actually down; it just does a quick check one time through (instead of two times through) to check the host; this will cut down on the time spent on already downed hosts Version: 1.0.10 (2007-06-27) - added more logging to system status plugin - bugfix: SNMP 2c fix If a server was configured for SNMP 2c, it would not read the version number properly, therefore it would still connect using version 1. This is fixed - added show_down_interfaces option to allow for graphing/displaying interfaces that are down - updated disk counter code to allow for manual configuration, as well as multiple devices lumped into one - bugfix: VI3 shows 0% when over 100% utilization If an ESX server was using over 100% CPU utilization, then CPU Used would display 0%; this is now corrected to view the proper CPU utilization - bugfix: VCMS managed ESX servers proper display Fixed a bug that would display all ESX servers if looking at a particular VI host, instead of just displaying only the ESX hosts that are associated with the VCMS server - separated Volumes and Aggregates for Netapps Added a 'disk-type' to the disks table which is only specified for netapp disks currently; lumps traditional volumes and flex volumes into the same box; aggregates are separate now - added disk alerts per unnoc.conf using the disk{} Now each device type (server, ms-server, netapp) shares the same code for checking disks that are configured in the unnoc.conf file; - added disk and net counters for VMware VI 3 added 2 new tables and 4 new PHP files - added ignore capability on a per disk basis - added last_success column, to determine when a status=0 host was last successful Version: 1.0.9.1 (2007-06-07) - bugfix: Updated links for VI pages, the wwwroot variable wasn't getting set or read properly - bugfix: updated model number for APC PDUs Canged from 1.3.6.1.4.1.318.1.1.4.1.4 to 1.3.6.1.4.1.318.1.1.12.1.5 - bugfix: All plugins (MySQL, OpenLDAP and PostgreSQL) were not getting updated properly in the database, so if one of them was down or not availble, it would send alerts until it was available; this has been fixed, it will only send alerts once Version: 1.0.9 (2007-06-01) - bugfix: updated Uptime plugin for non-US timezones - bugfix: fixed VMware VI3 check by IP address, also fixed ESX display - added perl-module-checker.pl, will find missing perl modules - bugfix: MS server process case-insensitive fix; all processes are now case insensitive, regardless of SNMP/config - added RRDTool pipes to Unnoc - added fedora, debian init.d scripts - added simple DNS checking - major code overhaul: added daemon mode, changed cron mode majorly; major code overhaul, moved all functions to central library locations (libexec) and unnoc-cron.pl and unnocd.pl check everything, using the libraries. - added 4 more interface counters: uni/mulicast, errors/discard packets - added APC bad battery alerts (failed battery) - added disk I/O gathering for certain UCD supported devices Version: 1.0.8 (2007-04-30) - added README.targets - moved all plugins to plugins/ directory, updated upgrade.sh - added documentation for changing from root to nonpriviledged user - added support for Net::Ping::External (root no longer required) - updated all plugin color graphs and created a standard display for plugins - added RRD PostgreSQL plugin grapher, plugins/check-postgresql.pl, added PHP graphs - updated mysql protocol code - updated snmp updown code to incorporate 'none' better - removed ignore_hosts file, consolidated all to unnoc.conf - added memory/CPU monitoring support on a per process basis for server types - added MS server process support; same as 'server' types, except its all done from the unnoc.conf file; CPU/memory monitoring are included also - bugfix: type ms-server and barracuda firewall was not reading alert/page settings properly; - added battery alert status for APC's, it will page only on power failure - added more details to memory usage graph page - added more details to display disk page, will display device now - added CPU usage for Cisco Devices, new data type 'cisco-generic' - bugfix: added "0 VMs found" on ESX page, also on VCMS page; also added server stats for VMWare, if the server is either not responding or is in maintenance mode - bugfix: non-zoom process displays had no lower limit set, so if a graph was always 1, for instance, then it would not display that - bugfix: Temp_units were not getting displayed on Barracuda zoom graphs, this has been fixed now Version: 1.0.7 (2007-01-25) - bufix: fixed Aironet and Apple client displays; under some strange cirumstances dead wireless clients would still be active in the wireless table, and would show two lists of SSIDs for the same SSID; this is fixed now, in both Apple and Cisco wireless APs - bugfix: fixed client count for proper reading of number of clients (some older versions of IOS dont give the proper number of clients, we just could them manually instead). - VMWare VI SDK support: removed VMWare VI SNMP support, added VMWare VI SDK support, so that it reads each statistic through the SDK instead of through the SNMP interface - added external script support, when a host is available/unavailable - added Barracuda SPAM Firewall suport, with resource graphs and tables - added VMWare VI3/ESX support SNMP, with resource graphs and tables - split the stats table into separate processes, memory and load average tables - added more statistics and graphing for Netapps - added href links to check_apc_temp and check_em01_temp - added logfiles for debugging and loglevels - fixed MS Windows SNMP collector: disks, memory, cpu and interfaces are now monitored - added better robustness to host up/down checker - added an alert blackout configurable time period - added maillog to unnoc/log directory - added small error checking, errors usually associated with DNS - added per disk limits for Netapp - set the default page of generic ping only hosts to display latency graphs by default - added underlines for graph links - added host statuses to each display page - added alert groups, for grouping admins together (page groups and email groups) - added upgrade script for APC UPS's, added two DS's - added support for APC PDU power strips, including 3 phase as well - added support for 3 phase APC Smart-UPS - bugfix: devices weren't getting marked properly as down for snmp if an snmp device went down when multiple hosts were checked (via the cmd line), then the host wouldn't get marked properly as being down. this is fixed now - added graphs for response times for hosts - added response times for MySQL and OpenLDAP plugins - added response_time column to the 'plugins' table - added millisecond response times (latency) for check_host_status - bugfix: network throughput timestamp display - bugfix: wwwroot variable not working on main page links were not getting generated properly on the main page because the conf_array wasn't defined in the function. They are now displayed correctly Version: 1.0.6.2 (2006-02-24) - bugfix: devices weren't getting marked properly as down for snmp if an snmp device went down when multiple hosts were checked (via the cmd line), then the host wouldn't get marked properly as being down. this is fixed now Version: 1.0.6.1 (2005-11-28) - bugfix: page fixed pages were not getting sent out because a global variable was clobbering the page message. this is fixed. Version: 1.0.6 (2005-11-17) - add interfaces options - added print_href_ functions - renamed 'nopage' to 'page' - renamed 'environ' to 'em01' - updated all docs for new config styles, updated README.config - added a description to the unnoc.conf file, for the title page - changed unnoc.conf to per host basis, updated all source base - added upgrade-graph-colors.pl and integrated it into upgrade.sh - added uptimes to all plugins - added doc/KNOWN_ISSUES - added stripslashes and add_slashes to php/perl functions, to fix single quotes - removed aironets.cgi, airports.php - added throughput checking to standard interface functions - removed throughput table, consolidated to the interfaces table - consolidated all types of devices to 'host' in display.php - added a norrd array for hosts that aren't supposed to be graphed - added noalert feature - added environ, apc temp status to check_host() - added environ and APC to system status - bugfix: check-stats.pl wasn't always updating the DB table if a process was not normal; this is fixed, every run through the SNMP table it will update the table regardless of what the status of the process is - added experimental procfix support - added as standard X-axis graph that is more useful - updated airports to new SNMP algorithm, added IP address per client - removed airports and aironets sql table, combined to "wireless" - removed airports.php, consolidated to display.php - updated aironets algorithm to use native SNMP to read clients, no longer need Net::Telnet::Cisco - fixed check-stats.pl to no display "no response (try xx)" the entire time a server is down, it will not just display it once, and then will log again when it comes back online - added RRD support for Cisco Aironets - bugfix: sometimes an uptime wouldn't get displayed if it was only 0 days, it will get displayed now always - bugfix: if a ping host was down, then when it came back up it would not be set in the hosts table properly because it was searching for an uptime string; this is now fixed - changed the SNMP algorithm for check-stats.pl check-netapp.pl, check-apcs.pl, resulting in a major speed increase - added plugins RRD upgrade script, plugins-mrtg2rrd.pl - added RRD support for EM01 environment sensor - added RRD support for netapp's - added a 'disks' db table, added upgrade scripts, updated netapp and stats - added standard display function set, for std display of graphs - bugfix: fixed href pathnames for proper display if webroot != wwwroot - added RRD support for APC's - bugfix: check-stats.pl very weird bug where net-snmp 5.0.9 would display disks twice, the fix for this was to just check if we've seen the disk before, if we had then we skip it - updated check-mysql.pl to work properly, added vars to unnoc-cron.pl - combined rrd/devices and rrd/servers to just rrd/, included script to do this - added build_path functions for both perl and php - changed lib/ structure: dispersed rrd_graphs.php, rrd_functions.php and functions.php to 30 different php files in the lib/ directory, MUCH better - added a smaller installation index.php example file (index-5.php) - bugfix: single quotes for interface description in check-stats.pl: added single quotes around interface name for rrd creation because in some instances, the interface description could be two words - changed check-stats.pl to write all interface info to the interfaces table - removed interface info from the stats table - added interfaces db table - changed plugin variables to cron_plugin_*, from cron_rrd_stats_*, updated unnoc.conf - removed 10 variables out of the unnoc.conf, and put them in a global var section in config-parse.pl and functions.php - added subs check_host_down, check_no_response to functions.pl, updated bin/*.pl - updated snmpd.conf example files with disk percentages - added disk minimum percentage support Version: 1.0.5 (2005-10-20) - removed the variables: mrtg_servers and www_mrtg_servers - added MRTG 2 RRD conversion script for servers only, stats-mrtg2rrd.pl - added RRD tuning scripts: stats-rrdtune.pl, rrd-fix.sh - bugfix: functions.php now properly displays a host's color if it had no_response was greater than 1 - updated RRA's in all RRDTool create commands, defaults to 86400:0, and 600/700 for AVERAGE - added neat_number function to check-stats.pl disk free error message - added lib/rrd_graphs.php - moved all RRD functions to separate php file, lib/rrd_functions.php - added RRD MySQL plugin grapher, plugins/rrd-stats-mysql.pl - added RRD LDAP plugin grapher, plugins/rrd-stats-openldap.pl - added new variable rrd_graph_colors to unnoc.conf - added etc/graph_colors.conf, for customizing RRD graph colors - added lib/display_int_graph.php - added lib/display_misc_graph.php - changed graph display from AVERAGE values to MAX values on all RRD graphs - added variable table placement of mem stats, depending on the procs and disks - added 4 new columns to mysql db, for interfaces - added 4 new columns to mysql db, for memory, updated mysql_table and upgrade SQL file - added RRD support more RRD's: for for CPU usage, Memory usage and TCP connections, and network interface monitoring with auto display function from display_server_stats() - added memory statistic collecting, added memory usage chart for dispay_server_stats - added proper stylesheet integration to display_server_stats function - fixed upgrade-config.pl, will print all arrays every time - changed all != 'NULL' occurences to IS NOT NULL, added an upgrade SQL file to take care of the db - updated upgrade-config.pl to re-write throughput array properly - updated unnoc.conf with more documentation and examples - added session->close to: aironets.cgi, check-aironets.pl check-airports.pl, check-apcs.pl, throughput.pl - updated partial documentation for alternate SNMP version/port - fixed README.config and README.config-upgrade (they were clobbered for 1.0.4) - bugfix: fixed division by zero in functions.php for throughput plugin - bugfix: check-environ.pl hang bug every now and then check-environ.pl would simply hang, for no reason. I added the alarm feature back into it, and this fixed the problem. If it fails to update, it will silently exit out - added display_rrd_*_zoom functions for zoomed graphs - added support for SNMP version 2 and 2c, and alternate SNMP port numbers - added MRTG templates for aironets, gateways and airports in unnoc/mrtg/et/ - added add_aironet, add_gateway, and add_airport in the unnoc/mrtg/bin/ Version: 1.0.4 (2005-10-09) - added upgrade-config.pl script to make upgrading config files easier - added README.config-upgrade - added upgrade.sh shell script - updated README.config - added RRDTool documentation (README.rrdtool) - removed variable throughput_time from unnoc.conf - added disk usage graph link for all disk usage stats (avail, used, etc) - added host_down checking feature to check-aironets.pl, check-airports.pl and check-netapp.pl - added a feature to check-stats.pl where it will check first if a host is down in the hosts table, if it is then it won't bother checking against it (this is so that the admin doesn't receive two alerts if a server goes down) - added display_netapp_disk.php - added correct step times and start/end times for graph generation in both functions.php and rrd-graph.pl - added on-the-fly generation of graphs in functions.php - added display_proc.php, display_load.php, display_disk.php in lib/ dir - added rrd_display_* functions to functions.php - moved neat_number() function to functions.pl, removed from check-stats.pl and check-netapp.pl - added neat_number() to functions.php - added week, month, year graph support - added rrd-graph.pl, however support is limited: disabled week, month, year graph creation because I need to add code for it to run "every now and then", not everytime it's called (or else the CPU usage skyrockets) - separated the actual RRD graph creation with RRD updates (check-stats.pl) - bugfix: added environ MRTG checking from unnoc-cron.pl previousy it wasn't checking the @environ array, only hosts/stats/etc. Now it's checking the environ array as well - added initial RRDTool support, highly experimental Version: 1.0.3 (2005-09-24) - added cron_* variables to the unnoc.conf file - added unnoc-cron.pl, a perl wrapper for calling all unnoc scripts from crontab - updated the add_server / add_device files to relfect new unnoc-cron.pl - added upgrade SQL file - added examples/mrtg/environ1/index.php example php file - added airports.php, for calling all airports - added a README.config file, explaining the unnoc.conf file - fixed more lib/functions.php paths in the examples/ dir - added apple airport mrtg config file in examples/ dir - bufix in lib/functions.php: changed the throughput functions to powers of 2 instead of base 10. 2^10 = 1024 and 2^20 = 1048576. This makes the bandwidth usage much more realistic - initial add of Apple Airport plugin, with documentation - added 4 new variables relating to mrtg in etc/unnoc.conf - bugfix in check-stats.pl: previously if you had a really high Load average and then it came back from that load average to a warning, it would be set to 'fine' but 'with a warning' in the database, and you would never receive a notice that it was okay. once it was brought down to green, it was considered coming from a warning and therefore not alert worthy, so it would basically slip under the radar. fixed now - display_server_stats now displays the uptime of the server, added two new columns to mysql db - update check-aironets.pl and aironets.cgi it now pulls description from SNMP instead of from the unnoc.conf; also changed config file option "aironets" from an array to a hash - added a default value $no_response_tries set to 1 in check-stats.pl if it doesn't exist - added status_() functions to functions.php, to separate colored dots on index.php - changed check-stats.pl, loads are logged always regardless of load_alert - bugfix in update-uptimes.pl, avoid illegal division by zero - centralized some functions, particularly send_email() and unnoc_log() - bugfix in check-stats.pl, msyql removal code for no_response works now - updated read_config() function to not require a var passed to it - added alerts for check-apcs.pl, also added humidity support - added check-environ.pl and environ-mrtg.pl for support for esensors em01 - added an option uptime_type to unnoc.conf and added min/max values to uptime plugin Version: 1.0.2 (2005-09-12) - added update-uptimes.pl - added check_ignore() function, instead of manual foreach @ignore_hosts loop - brought system-status.pl up to date with new check-stats.pl sql schema - added check-aironets.pl, does the same thing that aironets.cgi did except writes all output to a SQL data ase - added no_response_tries to unnoc.conf - added $wwwroot to unnoc.conf for relational paths - updated check-netapp.pl to write to SQL instead of directly to index.html files - wrote a migration tool re-write-stats.pl for migrating from 1.0.1 to 1.0.2 - added 1 new function to functions.pl (check_uptime) - added 6 new functions to functions.php for reading various parts of the db - major change for check-stats.pl, rewritten, it now writes/reads directly from the database - added unnoc logos to images/ dir - added community field to mysql of check-apcs.pl Version: 1.0.1 (2005-08-31) - minor bugfix, check-apcs.pl - minor bugfix, check-stats.pl - removed reset_update_throughput.pl, integrated into throughput.pl - updated throughput.pl to update mysql db directly, added php functions as well - updated README, formated much nicer, much more information rich - added README.apc, README.netapp, README.aironet, README.throughput - updated aironets.cgi code, added variables to unnoc.conf - added check-apcs.pl - conglomerated some functions - updated mysql_table, crontab example file - removed update_all.sh - changed paths around, set a plugin architecture - updated system-status.pl, it works now - added 6 new variables to unnoc.conf, 5 for system status, one for unnoc_logfile - updated check-hosts.pl and check-stats.pl considerably - updated throughput.pl and reset_update_throughput.pl - moved throughput.pl output from updating index.php directly to separate file - created the Changelog file (this file) - set $load_alert variable, configurable to what load averages cause alerts - updated doc/INSTALL file: - check-apcs.pl docs updated - fixed mysql path in the mysqll install - updated the nopage section to reflect variable, instead of zero byte txt file - changed nopage variable from a zero byte txt file to a variable in the unnoc.conf file - edited ignore_hosts, removed hosts and added a comment - removed some screenshots that wern're supposed to be in there - pulled the initial 1.0.0 release because of a few passwords in some files Version: 1.0.0 (2005-08-26) - first release - first import into CVS on sf.net - fixed some mysql code and hash functions in both check-stats.pl and check-hosts.pl (removed 'id'). Version: 1.0.0-pre2 (2005-08-25) - made only one definition of $webroot in the config file - separated all index.php functions into a functions.php library - separated javascript stuff from the index.php file into lib/*.js - updated mysql code to reflect grey dots (added a mysql 'valid' table) - fix a bug in the color dot code on the index.php page Version: 1.0.0-pre1 (2005-08-24) - removed meta txt files, have it pull all status from the mysql db - make unnoc.conf THE ONLY config file - set nested vars in noc.conf - added a config-parse.pl file - wrote the php config parser