/usr/share/check_mk/checks-man/services is in check-mk-server 1.2.6p12-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 | title: Windows Services
agents: windows
catalog: os/ps
license: GPL
distribution: check_mk
description:
Check the current state of a Windows service. Each service is either {running} or {stopped}
or in a transitional state like {starting} or {stopping}. This check becomes critical if the
service in question is not in the state {running}. It is getting UNKNOWN if the service is missing
on the system at all.
If you have just a small number of Windows hosts to monitor,
manual check configuration with {checks +=} will probably do.
With a larger number of hosts we propose using {inventory_services}.
This allows auto-detection of services according to their names,
current states and start types. The auto-detection can also make use
of host tags in order to use different rules on different sets of
hosts (for example for monitoring certain services only on production
servers). Inventory and manual checks can be used in parallel.
Since version 1.2.1i2 there is a new option {inventory_services_rules}
available, which is used to configure same options as the {inventory_services}
option but in a standardized format which is configurable via WATO.
item:
The name of the service as string. Please note, that the agent replaces
spaces in the service names with underscores. If you are unsure
about the correct spelling of the name then please look at the
output of the agent (cmk -d HOSTNAME). The service names are in the first column
of the section {<<<services>>>}. Please do not mix up the service
names with the display names of the services. Those are just being displayed
as a further information.
inventory:
The inventory function of this check helps you to auto-detect then services
that should be monitored on your hosts. Because Check_MK cannot know, which
services are important for you, you have to configure some rules.
As of version 1.1.10i2 it is very flexible, while still being compatible
with the old simple-style format.
In the variable {inventory_services} you specify a list of inventory
declarations. In the easiest form, such a declaration is simply
a string: the name of a service. On every host that service is found
{running}, a check for that service will be created during inventory.
If you prefix the name with a tilde {~}, then the string is interpreted
as a regular expression matching the {beginning} of the service name.
So the declaration {"~Audio"} will mach {AudioSrv}, and {".*Mobile"}
will match all services containing the work {Mobile}. All patterns
are case sensitive. Please note: the check being created will
{not} contain the regular expression but the precise spelling of the
service. One rule with a regular expression can create several checks.
It is also possible to depend on the current state and/or the start
type of the service when deciding whether to create a check. States
and start types are appended and separated by a space. For example
{"LanmanServer auto"} will match all services with the name {LanmanServer}
which have the start type {auto}. Other possible start types are
{boot}, {demand}, {disabled} and {system}. Please look at the agent output
if you want to know, which kind of start types exist in your
environment. The declaration {"Browser running"} means the same as
just ommitting { running}, since that is the default. If you want
to combine a state and start type, do this like the agent does,
by using a slash: {"LmHosts running/auto"} will create a check for
the service {LmHosts}, if that service if found running {and} has
the start type auto.
If you need your inventory rules to depend on properties of the host
or just apply on some selected hosts, you can use "tuple"-entries
instead of strings. Such an entries consist of an optional list of
host tags, the keyword {ALL_HOSTS} or an explicit list of hosts and
a {list} of service declarations like the one discussed above. Please
consult the examples for details.
Since version 1.2.1i2 there is a new option {inventory_services_rules}
available, which is used to configure same options as the {inventory_services}
option but in a standardized format. The option holds a list of rules
where each rule contains one entry matching a list of service matching
strings (regexes) state/start mode.
examples:
# inventory will add checks for the following services
# if found running on the target host:
inventory_services = [
# Monitor "MySQL_Server" if found running on a host
"MySQL_Server",
# Monitor "TSMMgR" if start type is auto (regardless if running)
"TSMMgR auto",
# Monitor "LmHosts", if found running and start type is auto
"LmHosts running/auto",
# Monitor all services containing "OraLsnr", if start type auto
"~.*OraLsnr auto",
# Pick *all* services with start type auto (a lot!)
"~ auto",
# Rules based on host tags (service declarations are in a list!)
# On all hosts with the tag prod look for "Browser" and "CiSvc":
( [ "prod" ], ALL_HOSTS, [
"Browser running",
"CiSvc running/auto",
]
),
# On super mega important hosts, monitor all services
# with start type auto as separate checks!
( [ "super", "mega" ], ALL_HOSTS, [ "~ auto" ] ),
]
# Inventory Services Rules (like configured when using WATO)
inventory_services_rules = [
# Monitor the "Browser" service on all hosts
((['Browser'], None, None), [], ALL_HOSTS),
# Add monitoring of all services starting with "MSSQL " on all hosts
# which have the "db" tag
((['MSSQL .*'], None, None), [ 'db' ], ALL_HOSTS),
# Add monitoring of all automatically started services on all hosts
# which have the "important" tag
(([], None, 'auto'), [ 'important' ], ALL_HOSTS),
]
# Manual checks without inventory. Those checks will be created
# without calling cmk -I - regardless of whether those services
# exist and that hosts.
checks += [
# make sure, that service "cimlistener" is running on all hosts
# with the tag "win"
( [ "win" ], ALL_HOSTS, "services", "cimlistener", {} ),
# The service "Apache" should is monitored only on two specific hosts
( ["websrv01", "websrv02"], "services", "Apache", {} ),
# Warn if the service Messenger is running:
( ["somehost"], "services", "Messenger", {
"states" : [ ( "running", None, 1 ) ],
"else": 0,
}),
]
[configuration]
inventory_services (list of entries): List of service declarations.
For each matching service found on the target host, one check will
be added to your monitoring.
inventory_services_rules (list of rules): List of service inventory rules.
For each matching entry in each rule found on the target host one check
will be added to your monitoring.
[parameters]
parameters(dict): A dictionary with the following optional keys:
{"states"}: A list of triples. Each triple specifies the monitoring
state for a service that is in a certain state. The first element
is {"running"}, {"stopped"} or {None} (if the state is not relevant).
The second element is the start type and can be {"auto"}, {"demand"},
{"disabled"}, {"unknown"} or {None} (if its not relevant). The third
element of the tuple is {0} for OK, {1} for WARN, {2} for CRIT and
{3} for UNKNOWN.
{"else"}: The monitoring state that should be applied if none of
the entries in {"states"} matches.
Per default, the check will get critical if the state is not
{"running"}.
|