Your OnCall still runs. It just cannot phone you.
When Grafana archived OnCall OSS it also switched off Cloud Connection — the hosted layer that delivered SMS, phone calls and push for self-hosted installs. The escalation chains still fire. There is just nowhere for them to go.
No credit card. No per-seat pricing at any tier.
Read this before the comparison table
Monitorion is not a drop-in replacement for Grafana OnCall, and every page telling you otherwise is selling you something.
OnCall is an alert router. It takes alerts your Prometheus or Grafana Alerting stack already produces and works out who to wake. Monitorion is an uptime monitor with on-call built in — it produces the alerts itself and then escalates them.
We are a good fit if…
You mostly used OnCall to get paged when your sites, APIs, certificates or servers went down. Monitorion does detection and paging in one product, so there is no Prometheus to run and no Twilio account to maintain.
We are the wrong tool if…
You route alerts out of a Prometheus stack you intend to keep. You need a router, not a monitor — fork OnCall, move to Grafana Cloud IRM, or look at PagerDuty or incident.io. We would rather say so than waste your evening.
Grafana OnCall vs Monitorion
| Grafana OnCall OSS | Monitorion | |
|---|---|---|
| Status | Archived 24 March 2026 | Active, independently run |
| What it does | Routes alerts you already generate | Detects outages and pages you |
| Needs Prometheus / Grafana | Yes, in practice | No |
| SMS & phone calls | Bring your own Twilio since March | Built in — 50/mo on Pro |
| Push notifications | Mobile app dropped OSS stacks | Email, Slack, Telegram, Discord, Teams + 6 more |
| Escalation policies | Yes | Yes — multi-step with on-call rotation |
| Hosted option | Grafana Cloud IRM — $19/mo + $20 per user over 3 | $29/mo flat, unlimited team members |
| Free tier | Self-host only, assemble your own delivery | 25 monitors, email alerts, no card |
| Open source | AGPLv3, archived | Private worker agent is open source |
Grafana Cloud IRM pricing from grafana.com/pricing, Pro tier: $19/month platform fee including 3 active IRM users, then $20 per additional active user.
What happened, in order
- 11 March 2025 — OnCall OSS enters maintenance mode. No new features; security fixes limited to CVSS 7.0 and above.
- 21 April 2025 — OnCall Cloud and Grafana Incident are folded into the unified Grafana Cloud IRM app.
- 24 March 2026 — the repository is archived and moved to cold storage. Cloud Connection is disabled, ending SMS, phone and push delivery for self-hosted installs, and the IRM mobile app stops supporting OSS stacks.
Credit where it is due: Grafana was straightforward about this. It published the dates a full year ahead, kept the licence open, said plainly that it was not trying to force anyone onto its cloud, and offered to support a community fork. That is a better exit than most products manage. It still leaves you owning a notification pipeline you did not previously own.
Questions
What exactly happened to Grafana OnCall?
Grafana OnCall OSS entered maintenance mode on 11 March 2025 and the repository was archived on 24 March 2026. On that same date Grafana disabled Cloud Connection, which was what delivered SMS, phone calls and push notifications for self-hosted installations. The Grafana IRM mobile app also stopped supporting OSS stacks.
Does my self-hosted OnCall still work?
Yes, and this is worth being clear about: Grafana did not break your deployment. It still runs, it is still AGPLv3, and Grafana explicitly encouraged community forks rather than forcing anyone onto its cloud. What stopped is the hosted delivery layer. Your escalation chains still fire — they just have nowhere to send a phone call unless you wire up your own provider.
So what do I actually have to replace?
Notification delivery. Grafana now points self-hosters at Twilio, Exotel or Zvonok for SMS and voice, and Pushover, Gotify or ntfy for push, all via outgoing webhooks. That is a Twilio account, a billing relationship, credentials in your config, and a delivery path you now own and must keep working.
Is Monitorion a drop-in replacement for OnCall?
No, and you should be suspicious of anyone claiming otherwise. OnCall is an alert router: it receives alerts from Prometheus, Grafana Alerting or Alertmanager and decides who gets woken up. Monitorion is an uptime monitor with on-call built in — it generates the alerts itself. If you used OnCall mainly to get paged when your sites and services went down, Monitorion replaces the whole chain. If you route alerts out of a Prometheus stack you intend to keep, you want a router: fork OnCall, move to Grafana Cloud IRM, or look at PagerDuty or incident.io.
What does Monitorion cost?
Free for 25 monitors with email alerts and no card. Pro is $29/month for 100 monitors, 1-minute checks, three checking regions, 50 SMS a month and phone-call alerts. There is no per-seat charge at any tier, so adding people to the rota does not change the bill.
Why would this not be archived in two years as well?
A fair question to ask after OnCall and Freshping. Three concrete answers rather than a promise: monitoring is the entire company rather than one app inside a larger platform, CSV export is on every plan including free so your configuration is never hostage, and the private worker agent is open source. Ask the same three questions of every alternative on your list.
Detection and paging, in one place
25 monitors free forever with email alerts. Phone calls, SMS and multi-region consensus on Pro at $29/month — no per-seat charge, however many people are on the rota.