I did:
apt-get install nagios nagios-plugins nagios-nrpe-plugin nagiosgrapher
and adjusted nagios cfg to:
process_performance_data=1
service_perfdata_command=ngraph-process-service-perfdata-pipe
I have one server with Ubuntu 11.04 (Desktop Edition) and one with 10.04 (Server Edition).
Now the thing is that the 10.04 server recognised almost all plugin for graphing (Ping, load, users, http response time,...)
The 11.04 server only recognised http response time. I am especially interested in load monitoring though.
The flag for logging performance data is in every service and also in my generic-service template , so this is not the problem.
In addition I have wrote a lot of specific monitoring plugins for our servers myself. They return the proper status code and some information Text.
The documentation ab out nagiosgrapher seems to be very rare.
Now I have the following questions:
How do I configure a new plugin to work with nagios grapher correctly? Where/how do I define the regex and logging/displaying properties?
How do I maintain what services are graphed, besides the flag in nagios service definition? - Do I have to reconfigure / update nagiosgrapher some how? How to find out why a service is not graphed that should be?
edit:
Well I already found out that under /etc/nagiosgrapher/ngraph.d there are 3 folders. Two of them containing service specific files:
Ubuntu 11.04 Desktop
joe@flimmit-joe:/etc/nagiosgrapher/ngraph.d$ find .
.
./extra
./extra/check_snmp.ncfg
./extra/check_bacula.ncfg
./extra/check_spamassassin_effectiveness.ncfg
./extra/check_gsm.ncfg
./extra/check_request_tracker.ncfg
./extra/check_windows.ncfg
./extra/check_hardware.ncfg
./extra/check_weather.ncfg
./nmgraph.ncfg
./templates
./standard
./standard/check_imap.ncfg
./standard/check_nagios.ncfg
./standard/check_ntp.ncfg
./standard/check_mailq.ncfg
./standard/check_ftp.ncfg
./standard/check_dns.ncfg
./standard/check_mysql.ncfg
./standard/check_pop.ncfg
./standard/check_smtp.ncfg
./standard/check_http.ncfg
./standard/check_ldap.ncfg
the other server
Ubuntu 10.04 server
root@ct39140:/etc/nagiosgrapher/ngraph.d# find .
.
./extra
./extra/check_hardware.ncfg
./extra/check_gsm.ncfg
./extra/check_request_tracker.ncfg
./extra/check_weather.ncfg
./extra/check_spamassassin_effectiveness.ncfg
./extra/check_snmp.ncfg
./extra/check_bacula.ncfg
./extra/check_windows.ncfg
./nmgraph.ncfg
./templates
./standard
./standard/check_ping.ncfg
./standard/check_users.ncfg
./standard/check_mysql.ncfg
./standard/check_ldap.ncfg
./standard/check_smtp.ncfg
./standard/check_http.ncfg
./standard/check_imap.ncfg
./standard/check_procs.ncfg
./standard/check_ftp.ncfg
./standard/check_pop.ncfg
./standard/check_mailq.ncfg
./standard/check_load.ncfg
./standard/check_nagios.ncfg
./standard/check_ntp.ncfg
./standard/check_dns.ncfg
./standard/check_swap.ncfg
./standard/check_disk.ncfg
Now I see that the 10.04 Server indeed contains more plugisn (why I dont know), which are exactly the services that the other one does not graph!
Heureka!
So I copied the plugin files, made sure they are accessible and restarted nagios, but the services are still not monitored.
Any ideas?
