From: <Stu...@am...> - 2004-11-17 18:40:48
|
-----Original Message----- From: Einecker, Stuart M. Sent: Wednesday, November 17, 2004 11:38 AM To: 'Ben Clewett' Subject: RE: [Perfparse-users] method 4 Pipe; data storage and creating graphs ? on Solaris 8 Thanks It looks like it is a plugin/performance data issue. I had tackled several nagios server and plugin challenges over the past week which confused the issue. Thanks for pointing me in the right direction. -----Original Message----- From: Ben Clewett [mailto:BCl...@pe...] Sent: Wednesday, November 17, 2004 9:20 AM To: Stu...@am... Cc: per...@li... Subject: Re: [Perfparse-users] method 4 Pipe; data storage and creating graphs ? on Solaris 8 Hi Stuart, You seem to have some interesting problems. My guess is that the plugins do not produce performance data. For instance shown after the '|': OK - load average: 0.38, 0.50, 0.50|load1=0.380;1.500;2.000;0.000 load5=0.500;1.000;2.000;0.000 load15=0.500;1.000;2.000;0.000 Can you run your plugins from the command line using the same parameters as Nagios and ensure that they all produce performance data similar to that above? Regards, Ben Stu...@am... wrote: > Hi all > > > > I've been working with perfparse for about a week or so and > > I was wondering if I could get some input on problem I'm working > > through. > > > > Currently I'm only able to graph 2 out of 7 servers monitored; > > and only 4 out of 5 services on those 2 servers. 3 of the services > > are on the local server where nagios and perfparse daemons are running. > > I have 9 active services total. > > The services are PING DNS DISK LOAD NTP for the 9 servers > > On the successful servers > > Server1 is graphing PING LOAD DISK > > Server2 is graphing DNS and ignoring PING. > > The OS is Solaris 8. Version 103.1 of perfparse. > > > > I'm using method 4 to pipe data from nagios into the mysql DB. > > The data stored for the 2 servers and services are in the > > "perfdata_service_bin" and metric tables. All the data is going raw > > > > I have removed the mysql data directory and recreated the db several times > > with the same affect. A couple of times I pulled the 2 servers out of > > the nagios config and then recreating the data. No results until I put > the 2 servers > > back in. So I'm wondering how does perfparsed > > know about only these 2 servers and how do I get data in the bin and > metric tables > > for the other servers. > > > > > > In the perfparse.cfg I have the following set: > > Service_Log = "|/usr/local/nagios/var/rw/serviceperf.log" > > Service_Log_Save_Position = "no" > > Daemon_Lock = "/usr/local/nagios/var/perfparsed.lock" > > Daemonize = "no" > > No_Raw_Data = "no" > > No_Bin_Data = "no" > > > > The nagios_perfparse.cfg services command is set to: > > > > command_line /usr/local/nagios/bin/perfparse_nagios_pipe_command.pl / > > usr/local/nagios/var/rw/serviceperf.log "$TIMET$" "$HOSTNAME$" > "$SERVICEDESC$" " > > $OUTPUT$" "$SERVICESTATE$" "$PERFDATA$" > > > > The "process-host-perfdata" command doesn't seem to be used. > > There is no variables to set or used for host info in the perfparse.cfg > > and I didn't see any variables for the host setup when I do a > > ./perfparsed -show_config > > > > > > +-------------------------+ > > | Tables_in_nagios | > > +-------------------------+ > > | perfdata_delete_policy | Empty > > | perfdata_graphs | Empty > > | perfdata_host | server1 server2 server3 server4 > server5 server6 server7 > > | perfdata_host_group | Empty > > | perfdata_prefs | Empty > > | perfdata_registry | Server Registration data > > | perfdata_service | Service info for all servers > > | perfdata_service_bin | Data for only 2 servers > > | perfdata_service_metric | 4 services for server1 and server2 > > | perfdata_service_raw | Raw data for all servers > > | perfdata_state | Empty > > | perfdata_summary | Empty > > | perfdata_summary_data | Empty > > +-------------------------+ > > 13 rows in set (0.01 sec) > |