Grafana Agent

Grafana Agent collects and forwards telemetry data to open source deployments of the LGTM (Loki, Grafana, Tempo, Mimir) Stack, Grafana Cloud, or Grafana Enterprise, where your data can then be analyzed. You can install Grafana Agent on Kubernetes and Docker, or as a system process for Linux, macOS, and Windows machines.

The Grafana Agent is open source and its source code is available on GitHub at

Grafana Agent is for engineers, operators, or administrators who want to collect and forward telemetry data for analysis and on-call alerting. Those operating Grafana Agent must install and configure Grafana Agent to properly collect telemetry data and monitor the health of running agents.


There are other ways of sending metrics, logs and traces to the LGTM Stack, Grafana Cloud or Grafana Enterprise, but there are a few advantages of using the Grafana Agent. These features are outlined below.

  • Provides a one-stop solution for collecting metrics, logs, and traces.
  • Collects out-of-the-box telemetry from popular projects like MySQL through integrations
  • Works seamlessly with the LGTM Stack. Alternatively, metrics can be sent to any Prometheus-compatible endpoint, and traces can be sent to any OTLP-compatible endpoint.
  • Offers new solutions to help scale metrics collection like host_filtering and sharding
  • Provides the Grafana Agent Operator, which enables individual teams to manage their configurations through PodMonitors, ServiceMonitors, and Probes.


Grafana Agent focuses metrics support around Prometheus' remote_write protocol, so some Prometheus features, such as querying, local storage, recording rules, and alerts are not present. remote_write, service discovery, and relabeling rules are included.

The Grafana Agent has a concept of an “instance” each of which acts as its own mini Prometheus agent with its own scrape_configs section and remote_write rules. More than one instance is useful when you want to have separate configurations that write to two different locations without needing to consider advanced metric relabeling rules. Multiple instances also come into play for the Scraping Service Mode.

The Grafana Agent for collecting metrics can be deployed in three modes:

Prometheus remote_write drop-in

The default deployment mode of the Grafana Agent is a drop-in replacement for Prometheus remote_write. The Agent acts similarly to a single-process Prometheus, doing service discovery, scraping, and remote writing.

Host filtering

Host filtering configures Agents to scrape targets that are running on the same machine as the Grafana Agent process.

  1. Gets the hostname of the agent by the HOSTNAME environment variable or through the default.
  2. Checks if the hostname of the agent matches the label value for __address__ service-discovery-specific node labels against the discovered target.

If the filter passes, the target is scraped. Otherwise, the target is ignored and not scraped.

To use Host Filtering mode, you set a host_filter flag on a specific instance inside the Agent’s configuration file. When you set this flag, the instance only scrapes metrics from targets that are running on the same machine. This is useful for migrating to sharded Prometheus instances in a Kubernetes cluster, where the Agent can be deployed as a DaemonSet and distribute memory requirements across multiple nodes.

Note that Host Filtering mode and sharding your instances means that if an Agent’s metrics are being sent to an alerting system, alerts for that Agent might not be able to be generated if the entire node has problems. This changes the semantics of failure detection, and alerts would have to be configured to catch agents not reporting in.

For more information on the host filtering mode, refer to the operation guide.

Scraping Service

Scraping Service Mode clusters a subset of agents. It acts as a go-between for the drop-in mode (which does no automatic sharding) and host_filter mode (which forces sharding by node). The Scraping Service Mode clusters a set of agents with a set of shared configurations and distributes the scrape load automatically between them. For more information on Scraping Service, see Scraping Service.


Grafana Agent supports collecting logs and sending them to Loki using its loki subsystem. This is done using the upstream Promtail client, which is the official first-party log collection client created by the Loki developer team.


Grafana Agent collects traces and forwards them to Tempo using its traces subsystem. This is done using the upstream OpenTelemetry Collector. Agent can ingest OpenTelemetry, OpenCensus, Jaeger, Zipkin, or Kafka spans. For more information on how to configure, refer to receivers. The agent is capable of exporting to any OpenTelemetry GRPC compatible system.