DIY Uptime monitor - show & tell

havoc

Trying to troubleshoot some intermittent network instability & didn't feel like configuring smokeping. So instead I decided to see if I can hack something together that just continuously scans the 255 IPs in the subnet.

20 lines of code, thrown into a LXC is now streaming data to a InfluxDB for the ~25 devices/VMs on my net

https://pastebin.com/XKUY1J3M

...pretty chuffed with the outcome. Hopefully by tomorrow morning it'll have accumulated enough data to see what's going on

ehab

well done havoc , i know how you feel.

.. by tomorrow morning ..

you will be exploring/debugging something else ... so enjoy what is remaining of the 1st day and have a moment of

chrisfy

Good job, awesome feeling to complete things like this. What are you using for the graph?

dahartigan

I really want a self hosted hetrixtools server monitor, but there's always hetrixtools.

havoc

@chrisfy said:
Good job, awesome feeling to complete things like this. What are you using for the graph?

No separate tool - InfluxDB 2 has charting built in. So much easier than a prometheus+grafana stack.

@dahartigan said:
I really want a self hosted hetrixtools server monitor, but there's always hetrixtools.

There is always netdata. They've got a live demo on their site & install is like a 2 minute job. Though not technically an uptime tool I guess. Or zabbix if you like complicated.

SGraf

@dahartigan said:
I really want a self hosted hetrixtools server monitor, but there's always hetrixtools.

Yesterday i finished the back-end code for a "blacklist monitor" project i am working on. Currently working on the front-end.

One of the follow steps is to evaulate uptime monitoring from multiple locations.