FOR SELF-HOSTED INFRASTRUCTURE
Monitoring for people who self-host and want it to keep working.
You set up the server. You don’t want to think about it again. Glassmkr is the layer that tells you when something needs your attention.
3 nodes free. No card required.
curl -fsSL https://app.glassmkr.com/install.sh | bash Crucible v0.13.0 on npm
The problem
You self-host because you want control. Plex on your own hardware, Nextcloud for your family, game servers for a small community, maybe some hobby ML inference. You’re not running a Fortune 500 datacenter. You also don’t want your services to silently break and not notice for a week.
The big monitoring tools weren’t built for you. Datadog wants to charge you per host with a minimum spend that makes no sense for three home servers. The DIY route (Prometheus + Grafana + Alertmanager) is a project you’d have to spend a weekend on, and then maintain.
You want monitoring that catches the real failures, doesn’t bug you with noise, and doesn’t add another layer of infrastructure to babysit.
How Glassmkr fits
You don’t have to figure out what to monitor.
We’ve already done that work. 60 alert rules ship enabled by default. They cover the things that genuinely break self-hosted infrastructure: disks failing, filesystems filling up, memory exhaustion, network interfaces going down, security patches piling up.
Installation is one bash command.
No Docker compose stanzas to write, no Helm charts to maintain, no Prometheus targets to declare. One line. Two minutes. Running.
Notifications go to where you already live.
Telegram and Discord work great for self-hosters. You’ll see your alerts in the same app you check ten times a day. No separate “monitoring inbox” to remember to look at.
The agent is readable.
Crucible is MIT licensed. If you self-host because you don’t want black-box software running on your servers, you’ll appreciate that you can read every line of the agent before you install it.
Pricing scales with what you can actually afford.
3 nodes free covers most home setups. If you have a real homelab with 10 servers, that’s $21 a month, less than one streaming service.
Specific alert rules that matter
Self-hosted infrastructure breaks in specific ways. The rules most relevant:
- Disk capacity: your media server’s storage filling up before the next Plex scan
- Disk health: SMART degradation on the drives in your NAS or homelab
- Filesystem state: read-only remounts (silent killer of self-hosted databases)
- Memory pressure: OOM kills (your services dying silently)
- Network state: interface down (your tunnels and exposed services going dark)
- Service health: systemd units failing (Nextcloud, Plex, custom services that crashed and didn’t get restarted)
- Unexpected reboots: your server rebooted at 3am and you didn’t know
- Security state: pending kernel updates, exposed SSH password auth, firewall disabled
The agent you can read
Self-hosters take security seriously. We built the agent to match.
Crucible is MIT licensed, on npm, source on GitHub. ~150-line bash installer. Runs as the glassmkr user, not root. HTTPS-only communication. Ships only metrics and alert state, never logs or command output or anything secret. Audit it before you install it. We’d want you to.
Pricing reminder
$3 per node per month. First 3 nodes free. No card required.
- For a typical home setup of 3 servers: free
- For a serious homelab of 10 servers: $21/month
- For a small ISP/hosting business with 50 nodes: $141/month
Try it on one server.
curl -fsSL https://app.glassmkr.com/install.sh | bashQuestions? Email [email protected]. The whole company is a small team; you’ll get a real reply.