Skip to content

Monitoring 10+ cheap boxes without going crazy

VPS Hosting by MeritBudi 28 replies 1.4K views
8 #21

This is why I said one provider. If you diversify providers, you need monitoring on a third. But if you consolidate, you need monitoring outside the consolidation. There is no free lunch. I use Hetzner for compute and a tiny BuyVM slice for external monitoring. $2/month insurance policy.

virsh list --all | wc -l: 47
#22

BuyVM is out of stock for months. I tried. Ended up with a RamNode OpenVZ for $3/year just for external ping. Overkill for the task but available.

seedbox, NAS, tape, and three offsite
7 #23

hankels said:
RamNode OpenVZ
And we're back to virtualization tax. You run monitoring on OpenVZ? The irony.

virsh list --all | wc -l: 47
#24

It runs

curl
once per minute. I think the node will survive.

seedbox, NAS, tape, and three offsite
4 #25

Has anyone tried UptimeRobot instead of self-hosted Uptime Kuma? I have 50 monitors on the free plan, no server to maintain.

#26

SpaceChetah said:
Has anyone tried UptimeRobot instead of self-hosted Uptime Kuma?
5 minute intervals, no certificate expiry, no status page customization. It's fine for "is my blog up" not for "I run 14 production VPS." Also they had a 2-hour outage in 2022 where nobody got alerts. Self-hosted means you control the failure mode.

436 days. reboot is surrender.
6 #27

I didn't know about the 2022 outage. Source?

#28

I don't have a source, it was discussed on HN. Maybe I remember wrong. Regardless, external dependency for your dependency detection is philosophically questionable.

436 days. reboot is surrender.
#29

Thank you all. I think I will try Netdata parent first, simpler than Prometheus. And move Uptime Kuma to BuyVM if got stock. If no stock I use my friend VPS in Singapore as external ping.

Post a reply

You need an account to reply. Log in or register to join the conversation.

Post reply Preview Save draft