Homelab monitoring tools

[TechMinerUK]

Hey all,

I've got a reasonably sized home lab at this point (3x Hyper-V hosts, 2x physical Ubuntu boxes, 5x Windows VMs, 7x Ubuntu VMs and Home Assistant) however I'm struggling to find a simple solution for monitoring them.

In terms of monitoring I'm not looking for anything spectacular but just to keep an eye on the following:

• Free disk space
• Uptime
• RAM spikes
• CPU spikes

So far I have had a play with Zabbix, Grafana/InfluxDB/Telegraf, Check_MK, PRTG and a few other off the shelf tools but I've not found something which is straight forward in terms of setup and alerts. Grafana/InfluxDB looked promissing but setting up the dashboards and monitoring was extremely faffy (Likely due to my lack of experience with it), PRTG seemed cumbersome and bloated for the use, Check_MK was great however the manual agent updates concerned me for security and Zabbix seemed very powerful but very complex to setup.

For uptime Uptime Kuma does the basics in terms of letting me know when something is fudged before it causes too much chaos however it would be nice to get disk space and the other pieces managed.

I was wondering what everyone else uses, if you have any suggestions etc for something that may fit the bill

Any help and advice would be super

 

[TechMinerUK]

I run PRTG in my lab on the free license, but I ran it at my previous job with a paid license so know it pretty well.

Centreon was mentioned in a meeting the other day but other than looking at the website, I've not dived into it.

Centreon Open Source | IT Infrastructure Monitoring Software

Centreon's open source edition is a powerful and flexible IT infrastructure monitoring solution for businesses. Learn more and try it for free.

www.centreon.com

I've not heard of them before but they look decent so I will have a try of them

 

[TechMinerUK]

I wrote a bespoke software to do this, monitor various servers running Windows, ESXi servers, Hyper V etc

Like yourself, I need to monitor the disk storage and RAID status, CPU RAM usage, services running and if they failed, restart that also check port listening ports for applications (service may be running but ports not listening)

To make it easier to monitor, all the data that was collected was written to a back end database and a simple front end web page to view the status of the various servers.

This is something I would love to do however my skills in coding is "limited" to say the least

I run a Zabbix server on a VPS to monitor my kit and the family/friends kit. I also use the same VPS for a Unifi controller - not expensive and can be setup for email notifications etc. I know it looks daunting to start but you soon get used to it. I use a 1080p 6 x 4 graph dashboard monitoring via SNMP (linux/windows boxes/VM's) for most things and Zabbix Agent for pfsense firewalls I look after.

I tried Zabbix previously however it seemed quite resource heavy for what I wanted it to do granted that may just be my lack of knowledge on it.

How much time did you spend getting it setup the way you have it now, is it something that you regularly maintain or has it been set and forget once you got it configured.

LibreNMS? I think that will do near to what you want.

I'll have a look into LibreNMS as it's been years since I looked into it and tested it properly

NetData? It's easy to setup and looks pretty.

Grafana Cloud is good - you get 50K metric series for free.

Prometheus is probably the most prolific OSS tool available - it's got integrations to most things

I had a go with NetData however the tiers for the cloud dashboard put me off which was a shame as the actual agent looked quite good and seemed to be relatively light weight

Prometheus looked good but seemed to require a considerable amount of config to get everything up and running how I wanted, granted that may just be me being impatient