Nagios Support Forum

I had the same problem. I rebooted my machine and the ghost machine went away as well. /var/log/messages reported a lot of errors saying that the host was checked, but that it wasn't in the configuration.

Still no idea. I wouldn't ever recommend rebooting as a trouble shooting step, how can you be sure that waiting 3min didn't fix the problem?

Here's what I had happening to me. I was using 1.1G and I deleted one of my hosts. Then I applied the configuration. I would click on "delete" host then log out. I cleared the cache, closed the browser and all that. I even went through the /usr/local/nagios/etc/ filesystem and deleted everything that referenced the host. I still got notifications throughout the entire weekend about the host being down.

There's two things I did to solve the problem. I downloaded 1.1H and ran "upgrade", then I rebooted my system.
Now, I don't recommend rebooting the system, but I think that restarting nagios and the other daemons would probably solve the problem. I didn't have much time so I decided to take the quick approach of rebooting the system. The ghost problems all went away and the system doesn't keep putting weird messages in the /var/log/messages file.

I'm new to the Nagios product and still evaluating it BTW.

You might want to correct an uncorrected bug(scheduled for the next release). The nagiosadmin Contact has the wrong template applied in the Nagios Core configuration. Should be prefixed with "xi_", though this normally causes notifications not to be sent.

http://go.nagios.com/forum/256/1113

I am still evaluating nagios xi. It looks great and seems to be everything I need plus more and I will be purchasing it.

I am, however, having an issue with just one of my hosts.
It is a windows server, I have all the default monitors configured, plus a few custom services. The problems is that the only performance graphs I see are ping (there are 2 of them). My 13 other windows servers have a full set of perf graphs.

The one difference in this host is that I did configure it with a host name in all caps, then I deleted it and re-added in all lower case letters (I even deleted the config backups). In Nagios core the host and its services are lower case, but on my Nagios XI dashboard and views it is still in caps.
Another symptom is that if I click the "CPU usage" service from the "Service Details" view I do see the load %, but if i click then the "Perf graph" tab I get "No performance graphs were found for this service.", and if i click the "Configure" tab and "Re-Configure this service" link all the fields are blanked out (not the case with my other servers).

I am wondering if there is some confusion there. I did open the mysql databases with phpMyAdmin, and the host tables have the server list in lowercase letters.

Is there somewhere else I should look? Anyone see this before?

Thanks in advance for any input.

The configuration database is in Postgress, not MySQL.
This could be the result of some ndoutils/ndodb foo.

hrm, ok the VM notes page say that phpPGadmin is installed, but it does not appear to be, because i get "The requested URL /phppgadmin was not found on this server." when i go to http://myserver/phpPgAdmin/. Any ideas? I should warn you that I am not a big linux guy.

Ok, I managed to get phpPgAdmin installed and I can see the db. I found a duplicate server and removed it, though still no perf graphs.

Was this also an issue with the case changing?
PNP, the graphing application, keeps it's own file record of hosts/services it's likely this is having problems.

Any information you can provide might be helpful.

I'll work with this on Monday.

yep, as per my first post, I changed the case for this particular server.