Which Companies Offer Grafana Support?

AceMQ offers Grafana support: 24/7 cover with a 15-minute emergency response, for self-managed Grafana on any version. The other sources are Grafana Labs and the open source community, compared below.

The open source support window is nine months per version, so an unattended Grafana falls out of support quickly.

How long Grafana supports a version

Grafana Labs documents its release and support cadence. A minor release ships every other month, a major release once a year around April or May, and patching releases at least every month. Each minor release is supported for nine months after its release date. The last minor release of a major version gets extended support for 15 months. Every supported minor version receives patch releases with bug fixes and security patches until its end of life, and nothing after that.

In practice, a Grafana that was installed, connected to its data sources and left alone for a year is unsupported, and Grafana has had enough security advisories that this matters.

What Grafana Labs supports

Technical support from Grafana Labs comes with a paid Grafana Cloud plan or a Grafana Enterprise contract. It covers the Grafana Labs products: Grafana itself, and the wider platform of Loki for logs, Mimir for metrics, Tempo for traces, Pyroscope for profiles, k6 for performance testing, and incident response tooling. Users of open source Grafana get the documentation and the Grafana Labs community forums, where answers are good and nobody owes you one.

That scope is wide. Grafana Cloud is a full observability platform: metrics, logs and traces from OpenTelemetry and Prometheus, application and frontend observability, synthetic monitoring, continuous profiling, incident response management with OnCall, Enterprise plugins for commercial data sources, and newer AI features for investigating an incident or monitoring AI applications. Grafana Labs technical support covers how those products behave. It is tied to your subscription and contract, and it answers questions about the tool, not about whether your team has instrumented the right signals or built dashboards that give an on-call engineer actionable insight at 3am.

Where independent Grafana support fits

• Self-managed open source Grafana with no contract behind it, often on a version past its nine months.
• Dashboards and queries: dashboards that take a minute to load, a query that overloads the Prometheus or database data source behind it, plugins that broke on upgrade.
• Alerting: alert rules that stopped firing after the move to unified alerting, notification routing, on-call teams drowning in noise.
• The estate: hundreds of dashboards nobody owns, provisioning and configuration as code, authentication and organization design.
• The data underneath. Grafana is the window. When a monitor shows no data, the fault is usually in Prometheus, Loki or the application instrumentation, which a Grafana-only support contract does not reach.

AceMQ provides 24/7 Grafana support with a 15-minute emergency SLA and named senior engineers, together with Prometheus and the rest of the monitoring stack. Examples: a dashboard outage remediation and Grafana alerting support.

The Grafana incidents that actually page people

• Dashboards that take a minute to load. A dashboard with forty panels fires forty queries at once, each over a long time range with no recording rules behind them. The data source, often Prometheus, slows for everyone. The fix is query design, variables that do not explode into hundreds of series, and caching where it exists.
• Alerts that stopped firing. After the move from legacy alerting to Grafana unified alerting, rules were migrated automatically and not always faithfully. No-data and error states behave differently, and notification policies replaced per-alert channels. Teams find out when an incident arrives without its alert.
• The Grafana database. The default SQLite database locks under concurrent use, and a Grafana that grew from a pilot into a shared service needs PostgreSQL or MySQL behind it, migrated without losing dashboards, users and alert state.
• Plugins after an upgrade. The move away from AngularJS panels broke older plugins and dashboards built on them. Unsigned or abandoned plugins stop loading. Each needs a replacement panel or a migration.
• Authentication and permissions. OAuth, SAML and LDAP integration, team sync, and folder permissions that have drifted so far that nobody knows who can edit the production dashboards.

Staying inside a nine-month window

With a minor release every other month and nine months of support for each, a self-managed Grafana needs an upgrade at least twice a year to stay supported. That is realistic only if upgrades are boring. The practices that make them boring are the same ones a support engagement puts in place: dashboards and data sources provisioned as code and kept in version control, a staging instance that mirrors production plugins, a database backup before every upgrade, and a read of the breaking changes in the upgrade guide for every version you skip. Teams that cannot sustain that cadence are usually better served by Grafana Cloud for the front end, with support focused on the data sources that stay in-house.

Choosing

On Grafana Cloud, use the support in your plan. On Grafana Enterprise and current, Grafana Labs is the first call for a product defect. If you run open source Grafana, or your incidents are about data, dashboards and alerts rather than the product, you need support that covers the whole observability path.