NSClient++ Help (#1) - Migration from 0.3 to 0.4 (#960) - Message List
I'm trying to migrate to 0.3 to 0.4. I have a problem with hostname in nsca.
Version of NSClient : NSClient++ 0,4,0,163 2012-04-14 x64
My nsca hostname is defined like this :
[/settings/NSCA/client] ; HOSTNAME - The host name of this host if set to blank (default) the windows name of the computer will be used. hostname = vwbWnyFG7cUdsZ138CeBtKuAo
On my nsca server, I received this hostname : vwbwnyfg7cudsz138cebtkuao
The hostname defined and received are different (upper case is not respected). So my nagios server does'nt recognize the host :
 Warning: Passive check result was received for service '6kFiUYHLfPjZoWBaTcnsQ9umN' on host 'vwbwnyfg7cudsz138cebtkuao', but the host could not be found!
Is there a configuration option to set or it is a bug in NSClient ?
Sounds like a bug... darnit... And here I was about to release tonight :)
Just out of curiosity why is "vwbWnyFG7cUdsZ138CeBtKuAo" then eric_the_windows_server?
Anyways, Ill look into it tonight and get back during the weekend...
Michael Medinmickem04/20/12 12:45:10 (13 months ago)
Our hostname and service name are unique token which are generate by our software.
Thank you for your feedback.jimmy04/20/12 14:27:53 (13 months ago)
I have two another problem.
Here is my configuration :
[/modules] CheckSystem = 1 CheckExternalScripts = 1 NSCAClient = 1 Scheduler = 1 [/settings/default] allowed hosts = 127.0.0.1 [/settings/NSCA/client] hostname = vwbWnyFG7cUdsZ138CeBtKuAo [/settings/NSCA/client/targets/default] encryption = 3 host = my_nsca_server port = 5667 password = xxxxx [/settings/scheduler/schedules/default] channel = NSCA interval = 1m report = all [/settings/external scripts] allow arguments = 1 allow nasty characters = 1 command timeout = 60 [/settings/external scripts/alias] a-check-win-cpu = checkCPU warn=$ARG1$ crit=$ARG2$ time=20m time=5m time=5s a-check-win-ram = CheckMEM MaxWarn=$ARG1$% MaxCrit=$ARG2$% ShowAll type=physical [/settings/log] date mask = %Y-%m-%d %H:%M:%S file name = nsclient.log [/settings/scheduler/default] interval = 60s [/settings/scheduler/schedules] 1Uz6uNPTXwDM27atSyh45kvq3 = a-check-win-ram "85" "95" [/settings/scheduler/schedules/6kFiUYHLfPjZoWBaTcnsQ9umN] alias = 6kFiUYHLfPjZoWBaTcnsQ9umN command = a-check-win-cpu "85" "95" interval=2m
1. Perf data
I don't have any perfdata for my two services. I don't find any option to force performance data to be send to my nsca server
I want to monitor my CPU load at 5s, 5m and 20m. The "time=5s" option generate an output error :
ERROR: Could not get data for 5s perhaps we don't collect data this far back?
and an error in nsclient.log 2012-04-20 15:24:29:
e:..\..\..\..\trunk\modules\CheckSystem\PDHCollector.cpp:232: Failed to get CPU value: \238(_total)\6: No timeframe given on command line (ie. time=0)
I see that there is an CPUBufferSize option but i don't know if it will resolve my problem and I understand how too configure it.
Thanks for your helpjimmy04/20/12 15:30:03 (13 months ago)
- lower case host names: Fixed in 164.
- perf data: Fixed in 165
- CheckCPU might be fixed in 166 but I did not have time to fix this this weekend... Seems I have botched the default values for frequency and buffer size (frequency seems to be missing all together even). So anything below 10s is currently not supported :(
Michael Medinmickem04/23/12 09:08:32 (13 months ago)
Thanks to have fix 1 and 2.
Oki for 3.
Another point : I think that performance data of checkMem script should be rounded. Exemple of performance data :
'physical memory %'=53%;85;95 'physical memory'=2.1000000000000001G;3.34;3.73;0;3.93
Value of 'physical memory' (2.1000000000000001G). is not rounded as its treshold (3.34;3.73)jimmy04/23/12 11:21:07 (13 months ago)
Is it possible to download version 165 / 166 somewhere?mopp04/23/12 21:01:48 (13 months ago)