icli - Icinga Command Line Interface

icli [-v|-vv|-vvv] [-z filter] [-h hosts] [-g hostgroups] [-s services] [-c config] [-C] [-f status-file] [-F rw-file] [-lh|-ls|-lq|-ld] [-a action[:args]] [host/service ...]

version 0.48

icli is a command line interface to Icinga. By default it lists all services and their states.

Note that when supplying custom config and status file paths, icli also works with Nagios. 100% compatibility is not guaranteed, however.

icli only works when executed on the host running the Icinga daemon. To use it on another host, shell aliases (like alias icli='ssh $icingahost icli') or similar are recommended.

You can narrow down the list of services you want displayed either using filters (like icli -z!o), the -h/-s arguments (icli -h aneurysm -s Libraries,Websites) or commandline args (icli aneurysm/{Libraries,Websites} with shell expansion).

-a|--action action[:args]

Run action on all matching hosts or services. args is a comma-separated list of action arguments and depends on the action in question. action may also be a one or two letter shortcut.

The following actions are supported:

a|acknowledge comment

Acknowledge service problems with string comment. This creates a sticky acknwoledgment with notification and no expire time. The comment will not be persistent.

Note: Acknowledgement of host problems is not yet supported.

d|downtime start,stop,duration,comment[,opts]

Schedule a non-triggered host or service (depending on the filter arguments) downtime. start and stop are timestamps and must be formatted as YYYY-MM-DDTHH:MM:SS, where the "T" is literal. The timestamp is assumed to be in the same time zone as the system running icli.

If duration is 0 (zero), a fixed downtime between start and stop is scheduled. Otherwise, a flexible downtime which will start between start and stop and last duration is scheduled. In this case, duration must be a real number appended with an optional unit (s for seconds, m for minutes, h for hours, d for days, w for weeks). If no unit is specified, seconds are used.

If a host is selected and opts contains "children", a downtime for all of its children will be scheduled with the same parameters as the host's. Likewise, if opts contains "trigger_children", a triggered downtime for all of the host's children will be scheduled.

comment refers to the downtime's comment field and must not contain the "," (comma) character.

r|recheck

Schedule an immediate recheck

R|force_recheck

Schedule a forced, immediate recheck

--api1 apiroot

Set Icinga 1.x (Classic UI) JSON API root to apiroot. apiroot must be the root of Icinga's CGI directory. For instance, if your tactical overview has the URL https://monitoring.finalrewind.org/cgi-bin/icinga/tac.cgi, you need to set apiroot to https://monitoring.finalrewind.org/cgi-bin/icinga.

Please refer to the "JSON API" section in this manual for more information about this API.

-c|--config config

Read config from file

-C|--no-colours

Disable colours in output

-f|--status-file file

Read the status from file

-F|--rw-file file

Use file as external commands file.

-g|--hostgroup hostgroup

Limit selection to hosts in hostgroup (comma separated list)

-h|--host hosts

Limit selection to hosts (comma separated list)

-l|--list downtimes|hosts|services|queue

List either services (the default) or hosts. Note that only the first character of the argument is checked, so icli -lh, icli -ls etc. are also fine.

-m|--match regex

Limit selection to hosts/services whose plugin output matches regex (perl regular expression, case insensitive. see perlre).

-o|--overview

Display "tactical overview"-style overview. By default (or when used with -ls) the number of all hosts and services (both total and divided by their state) is shown.

When used with -lh, lists all hosts with the number of ok / warning / ... checks on each host.

--realm realm

Set HTTP Basic Auth realm to realm. Defaults to Icinga Access. Only useful when combined with a JSON or XML API.

-U|--as-contact name

Only operate on service visible to name.

NOTE: This is meant to help find out which services a user has access to. It is NOT intended as a way to restrict access and should never be used that way.

-s|--service services

Limit selection to services (comma separated lists). Can be combined with -h/-g to further narrow down the selection, but may also be used stand-alone.

-v|--verbose

Increase output verbosity. Can be combined up to -vvv

-V|--version

Show version information

-x|--cut-mode mode

What to do with lines which are too long for the terminal: nothing, cut off, line break (with proper indentation). The default is line breaks

-z|--filter expression

Limit selection to hosts/services passing the filter. expression is a comma separated list of filters, only hosts/services to which all filters apply are selected. See also "FILTER EXPRESSIONS"

This is the standard output method. It contains the following:

* Service description

* -v: Service Flags (Acknowledged, Flapping, Passive, !no checks)

* Service state (ok / warning / critical / unknown)

* -v: Current attempt / Max attempts

* Plugin output

Enabled with -ld

* Host name

* Host state (ok / down / unreachable)

* -v: Current attempt / Max attempts

* Plugin output

Enabled with -lq

* Host name

* Service name

* Last check

* Next check

Each expression can be negated with an exclamation mark, e.g. "!A" for all non-acknowledged services.

A

Check state has been acknowledged

D

The host this service belongs to is Down or Unreachable

F

Service is flapping between states

N

Notifications for this service are disabled

P

Only passive checks are enabled. Note that !P simply means that active checks are enabled, no matter the status of passive checks

S

Check state is soft. For instance, it used to be OK and is now critical, but has not reached its maximum number and caused a notification yet. Good to find (or ignore) service problems which might just be temporary, non-critical glitches.

o

Host/Service state is OK

w

Service state is Warning

c

Service state is Critical

u

Service state is Unknown

p

Host or service state is Pending

d

Host state is Down

x

Host state is Unreachable

Zero, unless errors occured.

None.

* autodie (included with perl >= 5.10.1)

* DateTime

* DateTime::Format::Strptime

* DateTime::TimeZone

* Term::Size

It is probably not clear from the documentation when an action will operate on hosts and when on services.

Note that this software is not yet stable. Command line options may be changed / removed and thus break backwards compatibility at any time.

Either via mail to <derf@finalrewind.org> or on <http://github.com/derf/icinga-cli/issues>.

icli -r -s 'APT Updates'

Schedule a check of the "APT Updates" service on all hosts having it

icli -lq -h aneurysm -g chaosdorf-hosts

List check queue for all hosts in the hostgroup "chaosdorf-hosts", plus the host aneurysm

icli -z!o,!A,!S,!D

Show all service problems which are already hard states and have not yet been acknowledged. Also weed out problem services on hosts which are down anyways

Copyright (C) 2010 by Daniel Friesel <derf@finalrewind.org>

  0. You just DO WHAT THE FUCK YOU WANT TO.