Qdrant exposes its metrics in Prometheus/OpenMetrics format, so you can integrate them easily with the compatible tools and monitor Qdrant with your own monitoring system. You can use the /metrics endpoint and configure it as a scrape target.

Metrics endpoint: http://localhost:6333/metrics

The integration with Qdrant is easy to configure with Prometheus and Grafana.

Monitoring multi-node clusters

When scraping metrics from multi-node Qdrant clusters, it is important to scrape from each node individually instead of using a load-balanced URL. Otherwise, your metrics will appear inconsistent after each scrape.

Monitoring in Qdrant Cloud

To scrape metrics from a Qdrant cluster running in Qdrant Cloud, note that an API key is required to access /metrics. Qdrant Cloud also supports supplying the API key as a Bearer token, which may be required by some providers.

Exposed metrics

Each Qdrant server will expose the following metrics.

app_infocounterInformation about Qdrant server
app_status_recovery_modecounterIf Qdrant is currently started in recovery mode
collections_totalgaugeNumber of collections
collections_vector_totalgaugeTotal number of vectors in all collections
collections_full_totalgaugeNumber of full collections
collections_aggregated_totalgaugeNumber of aggregated collections
rest_responses_totalcounterTotal number of responses through REST API
rest_responses_fail_totalcounterTotal number of failed responses through REST API
rest_responses_avg_duration_secondsgaugeAverage response duration in REST API
rest_responses_min_duration_secondsgaugeMinimum response duration in REST API
rest_responses_max_duration_secondsgaugeMaximum response duration in REST API
grpc_responses_totalcounterTotal number of responses through gRPC API
grpc_responses_fail_totalcounterTotal number of failed responses through REST API
grpc_responses_avg_duration_secondsgaugeAverage response duration in gRPC API
grpc_responses_min_duration_secondsgaugeMinimum response duration in gRPC API
grpc_responses_max_duration_secondsgaugeMaximum response duration in gRPC API
cluster_enabledgaugeWhether the cluster support is enabled

There are also some metrics which are exposed in distributed mode only.

cluster_peers_totalgaugeTotal number of cluster peers
cluster_termcounterCurrent cluster term
cluster_commitcounterIndex of last committed (finalized) operation cluster peer is aware of
cluster_pending_operations_totalgaugeTotal number of pending operations for cluster peer
cluster_votergaugeWhether the cluster peer is a voter or learner

Kubernetes health endpoints

Available as of v1.5.0

Qdrant exposes three endpoints, namely /healthz, /livez and /readyz, to indicate the current status of the Qdrant server.

These currently provide the most basic status response, returning HTTP 200 if Qdrant is started and ready to be used.

Regardless of whether an API key is configured, the endpoints are always accessible.

You can read more about Kubernetes health endpoints here.