I swear you have installed an RPM build at one time or another.
what is the output of
ps -ef|grep nagios
stop nagios and repeat above.
if you get nothing finally, then start nagios like this.
/usr/local/nagios/bin/nagios /usr/local/nagios/etc/nagios.cfg
Oh, and when you recompile, you should start it by doing it clean like this.
make clean
then you can run ./configure
Still did not work.
I am now cleaning every file named nagios off of my system and will rebuild again from scratch. I saved all of the .cfgs because those took awhile to setup.
you may want to start over on those .cfg files too. of course you can keep the hosts.cfg and services.cfg, but definitely make a new nagios.cfg and cgi.cfg
How exactly do you start nagios anyway? It sounds to me, that is a big part of the problem. Also, did you do as I instructed with the ps -ef|grep nagios? etc??? I was curios as to what you saw?
Success.
I don’t know what it was - honestly. But I used ‘locate’ and cleaned out every7 nagios file I could. BTW… when I did the ps -ef I only saw one process.
Your knowledge has been immensly helpful to me. If you ever need a hosting account you let me know.
Thank you. Thank you and Thank you again. Now that I have this up this is the coolest damn web app I think I’ve ever played with.
Patrick
OK, now that you have it working with text based storage of status, etc, you might wanna have some fun by recompiling nagios to use mysql. If you think nagios is cool, it’s even cooler with a database to store your long term data. I went even further and made another machine a replication database(master/slave). Currently I have one machine making 1074 active checks and 509 passive checks. Anyway, have fun, you have lots to do now and a ton more gizmo’s to add, like performance stats, rrd graphs,…endless list of stuff…
Oh great. Passive and active? Now I have to figure out what those mean.
The first thing I am going to do is tracert everything between me and my servers and get that all added. Then start working on services. I’d like to get the whole path laid out to where I am happy.
[quote=“patrick24601”]Oh great. Passive and active? Now I have to figure out what those mean.
The first thing I am going to do is tracert everything between me and my servers and get that all added. Then start working on services. I’d like to get the whole path laid out to where I am happy.
[/quote]
Don’t stop with just tracert, since that will only show you the routers between you and the device. Most times, your network trouble will be with switches, cables, port’s on a switch. Map your network out by hand on paper, and then use snmp checks on the switche ports, device port’s, to see if the port is operational. If a cable breaks, the port will go down, and now Nagios will tell you exactly what is broken. From my experience in a huge network environment, 99.9% of network trouble has been a cable needing to be replaced or reseated. Try to find this kind of trouble without Nagios, and it will take you 45 minutes to ping all the switches(that is if you even know what switches to ping). Now that you have Nagios all mapped out, you could unplug every cable in your network, and you would be able to recreate the connectivity just from your new Nagios “Status Map” link.
Example:
Nagios PC plugs into port 3 on switch1. Switch1 is trunked to switch2 via port 24. Router1 is connected to switch2 port 5. Your new nagios would monitor all IMPORTANT PORTS. You don’t really want to only check if router1 will ping. You really want to know that all cables/ports are working. See the power in monitoring/fixing your network this way? People will praise the ground you walk on, when you are done, I guarantee it.
Since this is not my network - it goes from office to the DC in Texas, how can I do anything but ping checks? For example - when the first stop on my network is cox how do I know what port and how to check that one port?
I’ll do some searching on this via google also.
Ahh, yea, if you don’t have access to the ATT or local Ma Bell telco offices, then you wouldn’t be able to find out all the switch ports, etc. I thought you where working on a local network.
So yea, checking router up/down is about all you could do. YOu may be able to use snmp checks though(if it’s enabled), on the equipment that you do know of, discovered by your tracert run.