William Leibzon's Nagios Addons and Plugins

Conference Presentations

Nagios and Cloud Computing - Nagios 2011 Conference (September 2011)
This presentation is about use of nagios in cloud and virtualized environment and where number of servers and services is not fixed but can scale up and down. Several nagios cluster solutions, including DNX and Mod-Gearman are discussed as a way to deal with scaling nagios installations in the cloud
- Presentation slideshow is here: http://william.leibzon.org/nagios/presentations/nagios2011/presentation-nagiosconference2011.html
- Also available are slides on Nagios Exchange (official place for conference presentations posted there by Nagios team): http://exchange.nagios.org/directory/Multimedia/Presentations/2011-Nagios-World-Conference-Presentations/William-Leibzon--2D-Nagios-In-Cloud-Computing-Environments/details
- And PDF of this presentation is available here: http://william.leibzon.org/nagios/presentations/nagios2011/presentation-nagiosconference2011.pdf

Nagios Plugins

All my plugins can also be found on my GitHub repository https://github.com/willixix/WL-NagiosPlugins.

List of released plugins:

check_redis version 0.7 [size: 140k | last update: Aug 28, 2012] - This is Redis Check plugin. It gets statistics variables and allows to set thresholds on their value or their rate of change. It can measure response time, caclulate hitrate, memory utilization, replication delay (for slave) and other data. It can also check data of keys against specified thresholds. Plugin returns performance data for graphing and other processing for anything checked. You can download templates for PNP4Nagios here: check_redis.php
check_memcached version 0.8 [size: 124k | last update: Aug 28, 2012] - This is Memcached Check plugin. It gets stats variables and allows to set thresholds on their value or their rate of change. It can measure response time, caclulate hitrate, memory utilization and other data. It also returns status variables as perf data for graphing and processing, and you can download templates for PNP4Nagios here: check_memcached.php
check_by_snmp version 0.3 [size: 31k | last updated: Mar 28, 2012] - This plugin is used in conjunction with other plugins to retrieve data made available using snmpexec and provide this to other plugins for processing or to execute plugins remotely using snmpd and return the results to Nagios.
check_mysqld version 0.93 [size: 33k | last updated: Mar 01, 2012] - Replacement for check_mysql that allows to set thresholds for any variables in SHOW STATUS or for similar SQL queries. Data is also made available for graphing or other use (graph preview: queries | nagiosgrapher templates: full OR queries_only)
check_snmp_netint version 2.36 [size: 102k | last updated: June 09, 2012] - This plugin checks port status, STP, traffic and errors data of network interfaces. Based on check_snmp_int it is optimized for faster execution and better support of cisco switches. It does not require temporary files for bandwidth and utilization data.
check_snmp_temperature version 0.4 beta2 [size: 43k | last updated: Mar 31, 2012] - General temperature check plugin that directly supports dell, hp, cisco, juniper, alteon with other types easily specified if you know temperature SNMP OIDs. Data also returned as performance values for graphing or other processing.
check_netstat version 0.35 [size: 35k | last updated: Mar 01, 2012] - This plugin checks (directly on server or through snmp) number of open TCP or UDP connections on a list of ports and can send an alert if there are too many or too few. Data is also returned as performance values and can be used to graph TCP connections to/from the server.
check_files version 0.35 [size: 27k | last updated: Aug 21, 2012] - This plugin checks how many files of given type/name are in the directory and can return an error if there are too many. It can also check that given directory or file has not been un-updated for too long.
check_uptime version 0.51 [size: 24k | last updated: June 06, 2012] - This plugin can run locally or use SNMP and will attempt to auto-determine type of remote system supporting windows and several types or unix systems. It reports how long system has been up ('uptime') as well as system name and kernel. It can also issue WARNING and CRITICAL alerts for systems recently rebooted.
check_localtime version 0.11 [size: 4k | last updated: Nov 24, 2006] - This plugin returns localtime of the system and can do WARN and CRITICAL alerts if its within certain range. This is useful for regularly (daily, twice-daily) heartbeat alerts. Despite being one of my early plugins, it works well and does exactly what I want (so I never really needed to modify it) and has been used in every place I setup nagios on.
check_daytime version 0.2 [size: 10k | last updated: Nov 24, 2006] - This plugin checks that time on remote host is no more than specified number of seconds different then on nagios server. It supports checking time using TCP or UDP with daytime (port 13) and time (port 37) protocols. You should use it only if you can not use NTP to check time on remote system as this is not precise enough.
check_sasraid_megaraid version 1.92 [size: 44k | last updated: June 15, 2012] - This plugin uses SNMP to check status of LSI/Dell Megaraid card. It will work with SCSI Megaraid (Dell Perc3 & Perc4), SAS RAID (Dell Perc 5, Perc6) and MPTFusion cards. In addition to status of logical and physical drives it also checks for medium and other errors.
check_linux_procstat version 0.41 [size: 28k | last updated: Mar 25, 2012] - This plugin parses data from /proc/stat on linux system and outputs CPU utilization and load data. It outputs a lot more data (user, system, iowait, irq for each cpu and # of processes and context switches) as performance data for graphing and other use.
check_netsnmp_memory version 0.2 [size: 5k | last updated: Jan 9, 2012] together with
check_snmp_attributes version 0.32 [size: 46k | last updated: Dec 22, 2011] -
The check_netsnmp_memory plugin provides memory use information from SNMP including amount and percent of used real, system, buffer cache and swap memory. The plugin is an example of using generalized check_snmp_attributes.pl plugin library which allows to specify set of SNMP data values to be retrieved and variables as RPM expressions from these values which are the attributes users see and check set thresholds on.
check_snmp_table version 0.31 [size: 25k | last updated: Oct 6, 2006] - This is a general plugin for checking named values where names are in one SNMP table and values are in another SNMP table. You specify these names as attributes and specify threshold for each attribute. This plugin can also serve as a base for writing specific plugin for type of check you're doing.
check_snmp_dell_powersupply version 0.32 [size: 15k | last updated: Dec 22, 2007] - Plugin that checks that both of the redundant power supplies in a Dell server are functioning
check_snmp_baytech_pdu version 0.25 [size: 30k | last updated: Dec 19, 2011] - This plugin checks Baytech PDU retrieving various parameters (such as voltage, current, temperature, etc) for specific ports
check_jboss version 0.31 [size: 16k | last updated: May 2007] - This plugins checks one or more JBOSS attributes allows to set thresholds for when to send WARNING or CRITICAL alert. It uses twiddleutility and will therefore require local copy of JBOSS.
check_tftp version 0.11 [size: 3k | last updated: Nov 24, 2006] - This plugin tries to retrieve a specified file from tftp to verify its working
check_dnsname version 0.11 [size: 5k | last updated: Nov 24, 2006] - This plugin checks that specifies dns hostname exists and corresponds to expected ip address or has the same ip address as another "base" hostname.

profile_nagios_executiontime version 0.21 [size 4k | last updated: Jan 15, 2012] - This is NOT a plugin but short script to profiler and find which checks take longer time to complete. Run it directly from unix shell of active nagios system.

You may also want to see my NagiosExchange Software Repository

Nagios Grapher Additions

ngraphs.cgi version 0.82 [size: 67k | last updated: Dec 23, 2011] (also you will need ngraphs.css stylesheet) - A replacement CGI for NagiosGrapher (copy it over graphs.cgi, should work with both 1.5 & 1.6) that allows to look in the specific service period in the past and has general user interface that has "look & feel" very similar to Nagios status.cgi making it appear that graphs is part of the actual nagios CGI base. Ngraphs CGI has several user views that are similar to what you have with status cgi - all service graphs on same host, all services in same servicegroup, specified (or all) service graphs on all hosts in same hostgroup, etc. All these views are accessible with point & click system making naviation easier; in addition regular nagios status CGI link box is also available just like with other nagios CGIs. In the future Ngraph CGI UI will be separated from NagiosGrapher related code to allow for it to be used for other graphing packages.
An screenshot of this CGI (actually slightly earlier version) can be downloaded here: printscreen_ngrapher5_nagioshost.pdf
/ngrapher_n5 [directory | last updated: July 20, 2006] - A modified version of NagiosGrapher 1.5 which includes new web interface (replacement for graphs.cgi - see above for its capabilities), modifications to core to add additional features [see changes.txt] and shell scripts to work with more complex graphing configs (as well as some examples of those configs)

PNP4Nagios Additions and Templates

(this is work in progress, give me few more weeks to post more here)

process_perfdata.pl [size: 60k | last updated: Dec 30, 2011] - This is modification for PNP4Nagios performance data collection daemon/script that aadds several new options. Most important is "DS" which you can specify multiple times in a config you add to pnp4naqios/etc/check_commands/plugin.cfg, such as:
DS = 1:eth0_in_prct:GAUGE
DS = 2:eth0_out_prct:GAUGE
DS = 3:eth0_in_octet:COUNTER
DS = 4:eth0_out_octet:COUNTER
Which I used for check_snmp_netint. Three values separated by ':' as in above are required, first is numeric DS, second is name from performance data to match this to, and 3rd is type of data. This basically lets you specify exactly which performance data is GAUGE and which is COUNTER as well as make sure certain performance names go into specific RRD index. Anything you don't specify will still go to RRD with next numeric index (i.e. 5) being used and default datatype GAUGE. In addition to DS two more options are added: PROCESS_UNSAFEDATA and UNSAFE_NAME_REGEX. If you set PROCESS_UNSAFEDATA to 1, PNP will now try to process data from plugins where some of the data may not be valid peformance data (some of my plugins use performance variable to cache non-performance data for next plugin run), this is unsafe since PNP will try to put everything into RRD, sometimes that is ok, sometimes not. So to be more precise what is not ok you use UNSAFE_NAME_REGEX to specify which performance variable it should ignore, I set this to '^[^i][^s][^o].+\.\d+' for check_snmp_netint. This is all in process_perfdata.pl already but beware that I use really non-standard path to pnp4nagios installation and you probably want to change it.

Nagios Core Patches

- for nagios 3.3 : nagios-3.3.1-support_disabled_status.patch [patch size: 18k | last updated: Nov 15, 2011]
- for nagios 3.0 & 3.2: nagios-3.2.0-support_disabled_status.patch [patch size: 18k | last updated: Oct 10, 2009]
- for nagios 2.8 & 2.9: nagios28-support_disabledstatus.patch [patch size: 19k | last updated: Mar 30, 2007]
- for nagios 2.5: nagios25-support_disabledstatus.patch [patch size: 19k | last updated: Nov 1, 2006]
- for nagios 2.2 & 2.3: nagios22-support_disabledstatus.patch [patch size: 19k | last updated: May 2006]

This is nagios patch that adds "Disabled" status for detail/overview/summary CGI view which is for all services/hosts that have active checks disabled, such services no longer count under "all problems" with this patch. There are no nagios configuration options added by this patch.

- for nagios 2.8 : nagios28-support_samehostdependency.patch [patch size: 22k | last updated: Mar 30, 2007]
- for nagios 2.5 : nagios25-support_samehostdependency.patch [patch size: 22k | last updated: Nov 1, 2006]

Nagios patch that adds 'dependent_same_host 1' directive for servicedependency object which is used in place of dependent_host_name or dependent_hostgroups. This allows to have config where hosts are specified with "hostgroup_name" directive and services by either "servicegroup_name" or directly with "services. Examples of using this are:

define servicedependency{
       host_name                       *
       servicegroup_name               master_service
       dependent_same_host             1
       dependent_service_description   *
       execution_failure_criteria      n
       notification_failure_criteria   u,c
}
define servicedependency{
       hostgroup_name                  net_snmp,cisco
       servicegroup_name               snmp_master
       dependent_same_host	       1
       dependent_servicegroup_name     snmp
       execution_failure_criteria      n
       notification_failure_criteria   u,c
}

- for nagios 2.8 : nagios28-support_groupextinfo.patch [patch size: 225k | last updated: Mar 30, 2007]
- for nagios 2.5 : nagios25-support_groupextinfo.patch [patch size: 225k | last updated: Nov 1, 2006]
- for nagios 2.3 : nagios25-support_groupextinfo.patch [patch size: 221k | last updated: July, 2006]

Nagios "GroupExt" patch - adds supports for hostgroupext and servicegroupextinfo objects and allows to have multiple extinfo configs for same host or service ('allow_duplicate_extinfo 1' nagios.cfg option). It also adds the following new config options for all extinfo objects: 'target_frame', 'notes_textinfo', 'action_textinfo', 'notes_icon', 'notes_icon_alt', 'action_icon', 'action_icon_alt'. New for 2.5 patch is general config option (goes in nagios.cfg) 'allow_extinfo_missingconfig' that downgrades to WARNING when host/service is not defined. Example of using this is:
```
define hostgroupextinfo{
        hostgroup_name          nagios
        action_url              /nagios/cgi-bin/ngrapher/ngraphs.cgi?action=HGROUPGRAPH&hostgroup=nagios
        action_icon             ngrapher/graph.png
        action_icon_alt         View Graphs For Hostgroup nagios
        action_textinfo         View Graphs For Hostgroup nagios
        target_frame            _self
}
```
- for nagios 2.8 : nagios28-largerbuffers.patch [patch size: 6k | last updated: Mar 30, 2007]
- for nagios 2.5 : nagios25-largerbuffers.patch [patch size: 6k | last updated: Nov 01, 2006]

This is a patch to increase size of nagios strings & buffers that deal with results from plugins allowing to process performance data of up to 3k in size.

Other Patches Related to Nagios

nagios-plugins-1.4.15-certexpire_warncrit.patch[patch size: 12k | last updated: Jan 10, 2012]
Patch to Nagios Plugins 1.4.15 that extends certificate expiration threshold to allow specifying both warning and critical values. Before this patch only one value could be specified which is days certificate expires when WARNING alerts start to be sent, and CRITICAL alerts are sent when certificate actually expired. With this patch two values can optionally be specified together (as in -C 15,5) with CRITICAL alerts being sent 5 days before certificate expires.If 2nd value is not specified it is taken as 0 which is the old behavior. This change applies to check_http, check_tcp, check_smtp.
p1_new.pl[size: 32k | last updated: Jan 10, 2012]
After backing up original p1.pl you can copy this to nagios/bin/p1.pl. This allows plugins running under nagios embedded perl to redirect STDOUT (which is normally captured in special variable in p1.pl). This is rarely need as you usually want output going to Nagios but couple libraries do fail to run in embedded perl if this is not allowed. Please note that this DOES NOT yet allow to redirect output for external programs executed from perl script with exec() and system(), you need to do dup2() in your perl script if you want that.
Long.pm[size: 76k | last updated: Jan 10, 2012]
Long.pm.patch[size: 1k | last updated: Jan 10, 2012]
This is a patch and/or replacement for perl GetOpt::Long package. You can copy the file over /usr/lib/perl5/5.8.8/Getopt/Long.pm (or patch it). This fixes issues for some plugins that don't run in nagios embedded perl (result UNKNOWN status and errors), none of those plugins are mine but there are quite a few such as fairly common check_snmp_storage.pl.