First Impression/Questions on Nagios XI
Posted: Thu May 14, 2020 12:46 pm
Hello I'm a Linux system administrator tasked to investigate a monitoring solution for my company. I'm currently experimenting with a Nagios XI installation on a freshly updated CentOS7-64 installation.
Concerns:
- There are many subtle differences between Nagios Core and Nagios XI in terms of the file systems and configuration. If I were to pickup a book or watch videos on Nagios Core 4, there are pretty significant differences in the filesystem that lead to a lot of confusion between Nagios Core 4, and Nagios XI 5. Nagios XI 5 does not seem to have as much documentation or training material yet.
- As a system admin, I'm pretty comfortable using the Linux CLI. Majority of the configuration files I've stumbled across warn me to not modify them directly, but instead use the web UI to make modifications, or create a custom configuration file and upload it. Sometimes its much easier to find a specific configuration using the filesystem rather then having to click around a web UI.
- Configurations seem to need to exist on both the client and server. The server needs to pass down the arguements to the plugins of what to check, and the client needs to be configured to accept those arguements - This seems pretty tedious and isn't a common practice in most monitoring tools.
Questions:
- Which monitoring agent to install? Half of the official Nagios videos I watch infer to install NRPE, while the other half infers to install NCPA. The agent comparison PDF on the official site makes it seem like NCPA is much more feature fledged, and seems to be the new standard - So why does all the documentation/videos all want to install NRPE? This was especially confusing when I used the Nagios XI wizard to add a host and it provided me with a NRPE client instead of a NCPA.
- We use Ansible in our environment, so I was happy to find out there are prepackaged ansible playbooks for installing the agent to all my network hosts, the playbook installs the NCPA agent which again is sending mixed messages.
- I watched an official Nagios video on how to use Ansible to automatically add systems to the Nagios inventory, but it doesn't actually begin checking any services. If all of the basic check services use the NRPE agent, but the ansible playbook installs NCPA how would that work?
- Once systems are in the hosts file, how do I begin adding checks to them? Are the checks specific to the agent? Do I need to add the checks on the client, and also configure the server to check for them?
- After I used the wizard to add a Linux host, the wizard gave me the option to start checking for a few standard things - CPU, Disk, etc.. But I could not find anywhere to modify these checks/commands in the Bulk modification tool, if I wanted to change the definition of warning/critical levels for disk usage across hundreds of systems how would I do it?
I'm sure a lot of these questions are just due to lack of playing with the tool/environment, I won't be able to spend much more time playing with it as we need to find a monitoring solution quickly. I'm hoping someone can take a moment and help address my comments/concerns. Thank you!
Concerns:
- There are many subtle differences between Nagios Core and Nagios XI in terms of the file systems and configuration. If I were to pickup a book or watch videos on Nagios Core 4, there are pretty significant differences in the filesystem that lead to a lot of confusion between Nagios Core 4, and Nagios XI 5. Nagios XI 5 does not seem to have as much documentation or training material yet.
- As a system admin, I'm pretty comfortable using the Linux CLI. Majority of the configuration files I've stumbled across warn me to not modify them directly, but instead use the web UI to make modifications, or create a custom configuration file and upload it. Sometimes its much easier to find a specific configuration using the filesystem rather then having to click around a web UI.
- Configurations seem to need to exist on both the client and server. The server needs to pass down the arguements to the plugins of what to check, and the client needs to be configured to accept those arguements - This seems pretty tedious and isn't a common practice in most monitoring tools.
Questions:
- Which monitoring agent to install? Half of the official Nagios videos I watch infer to install NRPE, while the other half infers to install NCPA. The agent comparison PDF on the official site makes it seem like NCPA is much more feature fledged, and seems to be the new standard - So why does all the documentation/videos all want to install NRPE? This was especially confusing when I used the Nagios XI wizard to add a host and it provided me with a NRPE client instead of a NCPA.
- We use Ansible in our environment, so I was happy to find out there are prepackaged ansible playbooks for installing the agent to all my network hosts, the playbook installs the NCPA agent which again is sending mixed messages.
- I watched an official Nagios video on how to use Ansible to automatically add systems to the Nagios inventory, but it doesn't actually begin checking any services. If all of the basic check services use the NRPE agent, but the ansible playbook installs NCPA how would that work?
- Once systems are in the hosts file, how do I begin adding checks to them? Are the checks specific to the agent? Do I need to add the checks on the client, and also configure the server to check for them?
- After I used the wizard to add a Linux host, the wizard gave me the option to start checking for a few standard things - CPU, Disk, etc.. But I could not find anywhere to modify these checks/commands in the Bulk modification tool, if I wanted to change the definition of warning/critical levels for disk usage across hundreds of systems how would I do it?
I'm sure a lot of these questions are just due to lack of playing with the tool/environment, I won't be able to spend much more time playing with it as we need to find a monitoring solution quickly. I'm hoping someone can take a moment and help address my comments/concerns. Thank you!