I've been tasked with finishing the setup of a Nagios-based monitoring system. I've gotten the basics down and the system was already set up when I got there (just with very few servers added).
Right now I have at least one linux and windows server up on there and working properly, so I don't need help with basic setup, but I do have a number of questions...
Is there any glaring thing that should be monitored outright of almost every server that many people overlook? Is monitoring HDD health important? Right now, every server that can be pinged is getting pinged as a check...is that redundant (is that how nagios checks to see if a host is up unless otherwise stated)?
Is there any way to confirm proper functionality of the ubuntu-based iptables firewall machines? We've had some issues in the past that seemed to randomly occur where all the network interfaces on ONE firewall just stop going out to the internet, which is super weird because they're all puppet'd (2 run xubuntu-desktop on top of the regular firewall config to provide terminals to look stuff up, download stuff, and update tickets). The firewall script is just a simple init.d script that drops the proper iptables in (the init.d file is puppet'd), but no real status command or process that runs.
Is there any EASY way to confirm that puppet is properly running and updating the puppets like it should?
We have some home grown projects, many web based. Is it worth it to convince the developers to create NRPE scripts to confirm proper functionality of their apps rather than just using check_http or check_https?
I know that with windows machines, you can use nsclient++ and easily check to see if a service is running (such as w3svs or termservices), however I have been unable to find such a thing for linux. Is there any way to check that a LINUX service is running? Does "status <service>" return an error code with the proper int? Can I just use
omegle shagle voojio check_by_ssh?