Nagios-lite-like node wanted


This is what search returns and this is not what I'm looking for.

My smart home UI is built with Node-RED.
Some devices in my smart home solution send LTW over MQTT, others don't because they don't use MQTT. Some LTW are designed for normal operation, others are really "a scream from the tomb". That's not perfect design of the smart home but it works few years already.

Rarely, very rarely, random device stops functioning. I discover that few days later when I need something from this or that device.

Typical solution is running Nagios. Nagios can execute list of "checks", keep and track the status of the checks, and alert/execute_some_code if the status passed defined threshold. Nagios is the beast :slight_smile:

However, for my small network it's a mammoth in a crystal shop.
I'm going to build Node-RED variant of Nagios which:

  • Runs defined commands per every unit in the smart home and compares results with expected criteria.
  • Puts all the results in InfluxDB (or other DB).
  • In case of failure of a certain unit it sends email and/or IM (pushbullet, telegram, etc.)

Having "there is nothing new under the sun" I wonder if someone already wrote such node.
I'd prefer to use even buggy software and debug/improve it rather than writing from scratch.

Any advice/idea welcome.

Depending on what the "Runs defined commands" entails, I created a network map flow (map is shown using node-red-dashboard), that pings all my devices every x seconds and shows its status on a map. This could be expanded to take certain actions to be taken when a device is unreachable. I mainly use this for clicking the ip's, as I cannot remember all the ip's/devices in my network, so it becomes a glorified bookmark page for me :slight_smile:

1 Like

Thank you @bakman2
This looks cool and sophisticated. Looking into it...

Note it is called "dynamic" but this only applies to the icon colors. The network is setup as a static topology, located in the inject node in the inject node in the "Set Topology" section

Well, after 1 hour of attempts to understand and run it I gave up. It's too complex for stupid one like i am.
Therefore, i spent 3 hours more and wrote "fail-safe" (polite idiom for "fool-proof") solution:

Maybe I'll improve it later to be even simpler.

Well, after 1 hour of attempts to understand and run it I gave up.

Ofcourse you could have asked.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.