I have tried Nagios, zabbix, netdata, grafana and all the possible combination of Monitoring system… but still cannot find anything like PRTG… am i the only one suffering by this? i mean the functionality of PRTG and the installation process are unbetable.

No conf file, no database or apache or rules, is just simple as it should be.

All the rest seems to be hard to maintain, condfigure and make it work.

1 point

Yeah, I can feel your pain, went through this too. I found a good alternative, called LibreNMS. You no need to setup all the staff like db and bla bla, just download ova image from the web site and roll it on you server. PRTG good one but sometimes very dumb

permalink
report
reply
1 point

I’m completely OTel with Prometheus, Jaeger, and Loki now. No reason to fart around with dated collection/analysis tools with data silos – OTel is good enough for work, it’s good enough for the lab. I say this as someone who used Nagios and MRTG for many years, and have contributed to Nagios Core.

I’m pretty comfortable working with Ansible and Chef, though everything except the network equipment in my lab runs on k8s and Talos now. I’ve managed fairly complex Prometheus and Nagios configs with Chef and Ansible in the past – not too tricky once you have a config management tool on your belt IMO.

permalink
report
reply
1 point

Is the Talos you mentioned Talos Linux? Seems so from the context but just wanted to be sure.

permalink
report
parent
reply
1 point

Yup. Just my $0.02 – I wouldn’t say there’s significant benefit over running a mainstream Linux distro and dropping something like RKE2, k3s, Kind, etc on the hardware. If/when I rebuild this cluster, I’ll likely switch back to my existing Ansible roles for RKE2 on Ubuntu. I think the intersection of “people who need a very thin distro for just k8s” and “people running baremetal self-managed k8s” is pretty narrow. All the major cloud vendors already produce a thin distro of their own for “just k8s”. Choosing Talos likely adds up fast when you’re talking hundreds of clusters and thousands of nodes in colos or on-prem.

I have an ARM based cluster I use for network/CNI labbing which runs k3s on Armbian.

permalink
report
parent
reply
1 point

Yeah, Zabbix has a learning curve. But it’s the one that you may actually want to use at work. It’s also not rocket science. I’m an idiot and I figured it out.

permalink
report
reply
1 point

webmin is godd for me, is simple, lightweight, no database, no historical data (they are stored only on client in browser localStorage). I’m using webmin in my raspberries. Bonus: it has a ton of features, it’s very useful for managing linux. BTW it has support for multiple instances and 2FA for example.

permalink
report
reply
1 point

that’s the difference between the paid software, where the user-friendliness is the priority, and the open-source software, where the functionality is the priority.

Also check netdata which has an exceptional baseline config, you can instantly use it without any need to fiddle with the settings. Of course it will need some config later if you want to to go fancy, but it is instantly usable.

permalink
report
reply

Homelab

!homelab@selfhosted.forum

Create post

Rules

  • Be Civil.
  • Post about your homelab, discussion of your homelab, questions you may have, or general discussion about transition your skill from the homelab to the workplace.
  • No memes or potato images.
  • We love detailed homelab builds, especially network diagrams!
  • Report any posts that you feel should be brought to our attention.
  • Please no shitposting or blogspam.
  • No Referral Linking.
  • Keep piracy discussion off of this community

Community stats

  • 9

    Monthly active users

  • 1.4K

    Posts

  • 6K

    Comments