ICMP Blocked - Windows Host

[Anthony]

Hi,

> I am using NSClient++ to monitor several Windows boxes. It is not
> possible to allow ICMP echo on these boxes.
> 
> For this reason, I can monitor processes and services but the host
> itself shows as up.
> 
> Is there a way around this?
> Is it possible to do a check other than ping to show a host status of
> up?

Have you considered passive monitoring instead?

NSClient can be configured to push results instead of being queried on
demand. You can use this in conjunction with setting the host as
passively checked only and setting a freshness threshold such that if
the freshness threshold is exceeded, then the active check command
specified is run, which can do something like trigger a WARNING or
CRITICAL state. I've used "check_dummy" as the active check command
to do this.

I rolled out two standard and one custom config file (altered from a
template) for each nsclient install, so after installing NSClient++
with defaults, I'd just drop in my standard files and restart the
service.

Of course, this means that every windows host has a copy of the
password to submit results to NSCA. You may or may not like this
approach.

Ie. Nagios:
-----------
define host{
        use ...
        name                    windows-server
        check_command           host-pasv-stale
        active_checks_enabled   0
        passive_checks_enabled  1
        freshness_threshold     600
        check_freshness         1
        }
...
define command{
        command_name    host-pasv-stale
        command_line    $USER1$/check_dummy 3 "Computer or monitoring is switched off"
        }

(This is unknown state, you might prefer another state number and message)

NSClient++ NSC.ini:
-------------------
[modules]
NSCAAgent.dll
CheckWMI.dll
CheckSystem.dll
CheckDisk.dll
CheckEventLog.dll
CheckHelpers.dll
CheckExternalScripts.dll

[Settings]
use_file=1

[includes]
;# The order when used is "reversed" thus the last included file will be "first"
;# Included files can include other files (be carefull only do basic recursive checking)

custom.ini
common.ini

NSClient++ common.ini:
----------------------

[External Script]
command_timeout=60
allow_nasty_meta_chars=0

[External Scripts]
; Can be called as "Example_Command" later on
; Stripped out some inhouse commands here
Example_Command=C:\EXAMPLE\EXAMPLE.EXE ....

[External Alias]
; Can be called as "FaxRX" later on
FaxRX=CheckProcState MinCritCount=0 MaxCritCount=2 faxrx.exe=started

[NSCA Agent]
; Have the windows box phone home every 60 seconds
interval=60
; Must match NSCA server config
encryption_method=....
password=....
nsca_host=....
nsca_port=....

[NSCA Commands]
; Common commands all hosts will run
; The next line - host_check - is what sends the passive "I'm alive" message...
host_check=CheckOK "NSClient++ running"
CPU_Check=checkCPU warn=100 crit=100 time=1 warn=95 crit=99 time=5 warn=90 crit=95 time=15
DiskFree=CheckDriveSize MinWarnFree=1g MinCrit=500m CheckAllOthers FilterType=FIXED Drive=A Drive=B
WinServices=checkServiceState CheckAll "exclude=SysmonLog" "exclude=ATI Smart" "exclude=Ati HotKey Poller" "exclude=ATI HotKey Poller" exclude=Fax exclude=NSClientpp exclude=vmount2 exclude=RoxLiveShare10 exclude=RoxWatch10 "exclude=Roxio Upnp Server 10" exclude=SessionLauncher exclude=WebClient

Example NSClient++ custom.ini:
------------------------------
[NSCA Agent]
; If you want to override the windows hostname with something else used in Nagios
; hostname=...

[NSCA Commands]
; Format:
; WhatNagiosCallsIt=WhatItsDefinedAsHere
FaxRX=FaxRX
Explorer=CheckProcState MinCritCount=0 MaxCritCount=2 explorer.exe=started

------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null