Grafana OnCall OSS is Sunsetting: Your Complete Migration Checklist

Grafana OnCall OSS is Sunsetting: Your Complete Migration Checklist

On March 11, 2025, the DevOps community was informed of yet another decommissioning. Grafana Labs officially moved Grafana OnCall Open Source (OSS) into maintenance mode. We are now in the final countdown. On March 24, 2026, the project will be fully archived.

For teams running mission-critical infrastructure on the OSS version, this isn't just a "read-only" status on GitHub. It represents a fundamental shift in reliability. Specifically, the Cloud Connection, the engine that handles your SMS, phone calls, and mobile push notifications, will be deactivated on that same day.

At All Quiet, we’ve built an incident management platform that agrees with the philosophy that made Grafana OnCall OSS so popular. As you evaluate your path forward, your primary concern shouldn’t be "convenience" ; it should rather be a technical step forward for an even better Total Cost of Ownership. This guide provides a definitive checklist to ensure your migration doesn't break your incident response process, but rather helps you move it to the next level.

1. The Core On-Call Engine

The heartbeat of any incident response system is the rotation logic. If a tool can't handle your specific shift patterns, it’s a non-starter.

• Rotations & Handovers: All Quiet supports the complex rolling-user patterns SREs rely on in Grafana, including weekly handovers and custom daily shifts.
• iCal Integration: Personal calendar visibility is a must-have. All Quiet provides native iCal feeds so your team can see their on-call shifts alongside their personal lives.
• Override Management: Incident management isn't static. Whether it's a last-minute shift swap or a scheduled maintenance window, All Quiet allows for instant overrides without requiring a full Terraform apply.

2. Advanced Routing & Label-Based Logic

SREs might have been forced to route "alerts"; but nowadays need to route actual "signals” to protect focus. If you currently use attribute-based routing from your payloads in Grafana to separate "Database Lag" from "API 500s," you need that same precision.

• Payload Parsing: All Quiet natively understands the Grafana Alert Manager payload. You can route alerts based on any label, severity, environment, service, or cluster.
• Regex Support: We support complex pattern matching, allowing you to direct "Critical" incidents to the right squad member, while snoozing late-night"Minor" incidents until they either resolve themselves or only pop-up again during working hours.

3. Escalation: Ensuring 100% Coverage

What happens if your primary responder is in a dead zone or doesn't acknowledge the alert?

• Multi-Tier Chains: Just like Grafana OnCall, All Quiet allows you to define what happens if an alert isn't acknowledged. You can escalate from a push notification to a phone call, and eventually to a secondary responder or a team lead.
• Round-Robin Distribution: For larger teams, All Quiet can distribute alerts across a pool of engineers to ensure that no single person is overwhelmed during a high-traffic incident.

4. The "Pay-as-you-Grow" Trap

As you look at alternatives, you'll find that Grafana Cloud IRM is the recommended path. However, for those of you who aren’t fully bought into Grafana, the cost is the primary blocker. With a base fee of $19/month and an additional $20 per active user, a team of 20 SREs is looking at $419/month before even accounting for metrics and log ingestion fees.

For organizations that care about efficiency and hate complexity in their vendor pricing, this "scalability tax" is often based on features you don't need. With All Quiet, teams can roughly halve their monthly spend^* on incident management tooling compared to the example above and find a more predictable, cost-effective home that focuses on the core mission: getting the right alert to the right person at the right time.

The Migration Window is Closing

Choosing a successor to your OSS stack shouldn't feel like a forced upgrade of your Incident Response Process combined with a full move to a new observability tool. It should feel like a plug-out and plug-in of your favourite console.

In our next post, we will move beyond high-level technical parity and look at the operational dividend of migrating to All Quiet, specifically how you can facilitate a migration via our Terraform provider that cleans up your technical debt by consolidating your Infrastructure as Code (IaC).