🔥Prometheus to Levitate

We make use of the Prometheus remote-write feature to allow Prometheus to write to an external remote endpoint i.e. Levitate's remote-write endpoint.

Prerequisites

Create a Levitate cluster by following Getting Started.

Keep the following information handy after creating the cluster:

• $levitate_remote_write_url - Levitate's Remote write endpoint
• $levitate_remote_write_username - Cluster ID
• $levitate_remote_write_password - Write token created for the cluster

Writing data to Levitate

The information flow is as follows:

Just update the Prometheus configuration with the following changes for theremote_write:

remote_write:
  - url: "$remote_write_url"
    basic_auth:
      username: "$remote_write_username"
      password: "$remote_write_password"
    remote_timeout: 60s

After that, restart Prometheus, and data will be written to Levitate.

Using Prometheus Agent mode

With Levitate as Long term storage, you don't really need to store time series in your Prometheus. You can run Prometheus only in agent mode and stream data directly to Levitate.

The agent mode is limited to discovery, scrape and remote write which fits our needs nicely.

You don't need to have the overhead of running full-fledged Prometheus anymore.

Here is a sample docker-compose.yaml that runs Prometheus in agent mode.

services:
  prometheus:
    container_name: prometheus-01
    image: prom/prometheus:v2.39.1
    ports:
      - 9091:9090/tcp
    volumes:
      - ./prometheus.yaml:/etc/prometheus/prometheus.yaml
    command:
      - '--config.file=/etc/prometheus/prometheus.yaml'
      - '--enable-feature=agent'
    network_mode: host
    restart: always

The prometheus.yaml will need to have remote_writesection as follows:

global:
  scrape_interval: 1m
  scrape_timeout: 30s

scrape_configs:
  - job_name: prometheus-self
    static_configs:
    - targets: ["127.0.0.1:9090"]

remote_write:
  - url: "$remote_write_url"
    basic_auth:
      username: "$remote_write_username"
      password: "$remote_write_password"
    remote_timeout: 60s

Verification

The Levitate Cluster will start showing a summary of ingested samples.

Next steps

Create a read token for your Levitate Cluster and follow our guide to Configure Grafana to visualize the time series data getting sent to Levitate.

Scaling Prometheus Remote Write

Read https://last9.io/blog/how-to-scale-prometheus-remote-write/ where we talk about how to scale Prometheus Remote write for Federated clusters considering High Availability requirements.

Troubleshooting

Please get in touch with us on Discord or Email if you have any questions.