New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Best "set and forget" monitoring for 50+ servers with public status page?
Hi everyone,
I'm looking for recommendations on how to monitor a fleet of 50+ servers using an all-in-one tool.
My requirements are:
- Real-time & History: I need to see Disk, RAM, CPU, and Network usage (both current and historical graphs).
Alerts: Notifications for downtime or resource thresholds (e.g., Disk > 90%).
Public Status Page: I want a single public page showing the stats of all servers.
I've already tried Beszel, but I'm looking for other options.
Does anyone have a solid setup for this many nodes? I’d prefer self-hosted but am open to SaaS if the free/cheap tier covers 50 nodes.
Thanked by 1mans_xd
Comments
i am following this Thread, interested
Don't know anything free simpler than beszel
I prefer to monitor using node_exporter / prometheus / grafana
But it requires more effort to configure it properly, especially for alerts
Zabbix is only real choice in this scenario, but massive learning curve. We use site24x7, not free for that many but its outside of our infrastructure and just works. We use it for servers and main network. Internally we use zabbix to monitor all our switches and network nodes, has templates for switches that do most everything.
Zabbix or netdata with custom status page based on the API with the help of tailwind react and vite should be enough i guess.
ltstats by @lukast__ .
It's rather simplistic but it works very well.
I use it for like 5 months now and haven't had any issues.
komari also looks promising, but I haven't personally use it. ( @sh97 does )
There is a guy here that has like +350 idle servers, and I saw one time he has a public status page with this kind of feature, I just forget his username and the tool he was using
@zGato ? He's big time nezha shill.
Yes that is the one, Thanks!
I am wondering what his experience with it? like its lightweight when its monitoring that much of servers, does it work well.
145 down !
So, did the detox cure actually work?
You already got the best... Beszel
I found his public status page, also looks good and feels like Nezha.
partially + i broke a bunch of servers
with 50 I'm sure you're fine with Nezha, but with like +200 the UI starts to get laggy. However, it does seem to be handling it like a charm, lol
I vote for ltstats too, simple and works well, also the only one(?) that show %steal as Komari doesn't have that.
Still waiting for ltstats to be able to monitor my Windows VPS....anyone has a workaround maybe ? lol
may KOMARI will help you
Might not be exactly cheap, but it's been very reliable so far for us: Hetrixtools, if you just want to pay, and completely forget. Just FYI: will cost 19.95$ per month for 60 servers.
Checkmate.
https://github.com/bluewave-labs/Checkmate
Thanks for the suggestion, Do you have any experience with it ? are you using it now ?
checkmk. they have a community version and it does real alerting as well (like custom processes and stuff) that many lightweight "monitors" do not offer.
I am also using beszel but do not understand how it tracks certain things. For example, when there is a downtime I try to understand at what time it took place and how long it was but beszel shows different time when I switch between time periods like 1 min, 1h, 12h etc. Not sure whether this is beszel or me
https://github.com/lukastautz/ltstats
It says sh97's Chicken Farm Komari Monitor
Zabbix CE for enterprise grade monitoring (used at work)
I personally use Prometheus + Alert manager + grafana setup.
Easiest would probably something like Hetrixtools. If you want to spend just a little bit of time, a self-hosted prometheus (with node exporter) with Grafana as dashboard works like magic.
That is a lot of chickens
Maybe uyuni for suselinux, ltstas, beszel
It is easy task setup prometheus (with node exporter) and grafena, but it takes some time.
+1 for hetrix. its cool and simple.