I’ve not managed to make any meetings as yet, but am glad to see that a local homelabs group has appeared in Brisbane!
Do any members of the group have suggestions for a good network monitoring system for home/homelab use? I have played a little with LibreNMS, OpenNMS, and Icinga2 but they’re way more capable (and complex) than what I’m looking for. I suspect that Nagios and Zabbix fall into that category too but admittedly haven’t installed them to try myself.
Ideally I’m looking for something that can do basic traffic/uptime monitoring via SNMP and ping, and keep logs for 1-7 days only. I don’t really want to install agents on individual machines/VMs and don’t have a need to track CPU/RAM type metrics from them. I used Cacti for some basic network monitoring a long time ago (15+ years) but am curious if there are any stand alone distributions, docker containers, or other FOSS projects that provide basic network monitoring before I explore Cacti again.
Are any of the Homelab Brisbane members doing network monitoring, and if so, what’s your preferred solution?
Welcome to homelabs and I am sorry you have not made it to one of the meetings. We have moved venues three times in the past and are happy to try somewhere else if the venue and time are a problem.
I played with Nagios many years ago but these days most of my timeis spent in docker. For resource monitoring I use a combination of docker stats, portainer and on a remote server glances. Uptime-kuma is a good application for monitoring your remote sites and notifications can be called for your favourite notification system. (They list over 50 options.)
@jdownie may have more experience that I, so I’ll check with him.
Uptime Kuma looks very slick and perhaps a combination of that and Cacti for traffic monitoring via SNMP might be the way to go. I set up some basic graphs in Cacti showing traffic to/from switch ports after my earlier post and that’s about all I wanted for the time being! Uptime Kuma’s interface and alerts might be the missing piece for my purposes.
Thanks for the tip about Portainer as well. It sounded familiar but wasn’t something I’ve explored yet. I’m new to the world of Docker and Portainer looks like it’ll be a useful tool when I go a bit further down that rabbit hole.
As for the meetings, my schedule is probably more of an issue than venue or timing but I’m hoping that things line up for me to join in soon.
Thanks for the info on glances! I often use btop when trying to monitor something, and having a web interface that performs the same function would be a great complement to David’s Uptime Kuma suggestion and Cacti.
I also use Home Assistant and it never occurred to me to integrate computer/network monitoring into that interface too. I’ll check out to see if something on the Add-on store or HACS would work for me. Hopefully I can make one of the online meetings soon and check out how you’ve got your monitoring set up.
Yes, another ancient thread resurrection, sorry! Have gradually been catching up on my HLB Homework over the past month. Thought I’d open this up again to discuss, and because I know that a lot of participants here love a Sankey chart thanks to the various projects we’ve all been engaged in on energy management!
I toyed with a few different options for this task, and it admittedly took me a while to pin down what problem I actually wanted to solve. Uptime Kuma looks brilliant for homelab use and I ran it here for a while, but for my use case “it works” = up, “it doesn’t work” = down, and I wasn’t getting much value add from Uptime Kuma that I wouldn’t notice myself or get from a family member saying x doesn’t work . I will likely come back to Uptime Kuma as a final step once the homelab rebuild is done, as I do like the idea of a simple and accessible dashboard for some services.
I ultimately went with Akvorado for network monitoring. I’d written off doing NetFlow at home as it seemed a bit much and I didn’t really want to spin up new hardware just for NetFlow stuff. Akvorado specifically was initially set aside as I didn’t want to run a dedicated box as a flow collector and analyser, and it’s targeted towards ISPs or IXes rather than small networks.
However, after coming back to it and giving it a go in a [grossly undersized] VM, it’s working well and solves the main problem I wanted to solve - where is traffic going to/from? I use port mirroring and Wireshark when I need to look at specifics, but was always missing a bird’s-eye view. Akrovado fills that gap, even with pretty resource constrained RAM/storage. I only want a few days of NetFlow data to supplement Wiresharking and answer those “why is my vacuum cleaner uploading stuff” type questions. No doubt plenty of other projects (e.g., ntopng or OPNsense) would do what I wanted as well, but I’m impressed with Akvorado so far given that it was a single docker compose file and a few YAML edits. SNMP and NetFlow export was already set up on my homelab hardware, so everything just magically worked once I plugged the various credentials and addresses in. I’ll still get Wireshark out when I need to have a closer look at something, but Akvorado has helped in knowing what to actually look at in the first place.
Probably a bit out of the homelab space, but I thought it was worth adding to the discussion. Happy to demo or discuss at the next in person/Jitsi for anyone curious.
. I will likely come back to Uptime Kuma as a final step once the homelab rebuild is done, as I do like the idea of a simple and accessible dashboard for some services.