How much RAM does OpenObserve need?

OpenObserve is extremely lightweight: under 10 MB RAM at idle. For real workloads (ingesting logs from multiple services), 512 MB to 1 GB is comfortable. A 2 GB VPS runs OpenObserve alongside your other applications without issues.

Is OpenObserve compatible with my existing log agents?

Yes. OpenObserve supports OpenTelemetry (OTLP), Elasticsearch-compatible bulk API, Prometheus remote_write, syslog, Fluent Bit, Vector, and Filebeat — making migration from existing setups straightforward.

How does it compare to Grafana + Loki + Prometheus?

The Grafana/Loki/Prometheus stack runs three separate services requiring complex configuration and higher RAM. OpenObserve does the same job in a single binary: no cluster setup, built-in dashboards, and a shared columnar storage backend — significantly lower operational overhead.

Yes. The self-hosted Community Edition (AGPL-3.0) is completely free. A cloud-hosted SaaS edition is also available (openobserve.ai) with a free tier. On ServOrbit, you only pay for the VPS — OpenObserve itself costs nothing.

Monitor your infrastructure.

OpenObserve

All your observability data in one place — logs, metrics, traces, dashboards — self-hosted on your VPS.

256 MB (1 GB recommended for production workloads) RAM 1 vCPU Port 5080 Available

Source code

Tech stack

DockerRustOpenObserve

Minimum RAM256 MB (1 GB recommended for production workloads)

Minimum CPU1 vCPU

Default port5080

Compatible OSChoice of Linux distributions

OpenObserve is a modern observability platform that unifies logs, metrics, traces, and dashboards in a single lightweight binary written in Rust. It replaces complex Elasticsearch + Kibana stacks at a fraction of the cost: 140× lower storage costs, sub-10 MB RAM at idle, and a familiar SQL/PromQL query interface.

Deployed on your VPS in under two minutes (single Docker command), OpenObserve provides everything you need to monitor your applications and infrastructure: OpenTelemetry-compatible ingestion, built-in alerting, live dashboards, and even LLM trace monitoring. No JVM to tune, no cluster to manage — just run and observe.

Key features

Unified observability: logs, metrics, traces and dashboards in one platform

140× lower storage cost vs Elasticsearch (compressed columnar storage, written in Rust)

OpenTelemetry-native: compatible with any OTLP agent (Fluent Bit, Vector, Filebeat...)

PromQL and SQL query interfaces — no new query language to learn

Built-in alerting: Slack, PagerDuty, email and webhook notifications

LLM/AI trace monitoring: trace OpenAI, LangChain and LiteLLM calls natively

Sub-10 MB RAM at idle, single binary — minimal VPS footprint

When to use this solution?

1

Application log management

Centralize logs from all your services (Nginx, PHP, Node.js, Docker) in one searchable interface. Ingest via syslog, Fluent Bit, Vector or OTLP — sub-second search over gigabytes.

2

Infrastructure metrics monitoring

Collect PromQL-compatible metrics from Node Exporter, cAdvisor or Prometheus and visualize them in OpenObserve dashboards — without running a separate Prometheus server.

3

LLM / AI application tracing

Trace every OpenAI, LiteLLM or LangChain call: latency, token count, model used, error rate. Pair with the LiteLLM gateway already in your catalogue for complete AI observability.

Deploy OpenObserve on your VPS

Guide optimized for ServOrbit Cloud VPS.

01

Order a ServOrbit VPS

Any plan with 1 GB+ RAM is sufficient. OpenObserve uses under 10 MB at idle; choose Ubuntu 24.04.

02

One-click install

Select OpenObserve in the ServOrbit marketplace and click Deploy. Docker is installed and the container starts automatically — no manual configuration needed.

03

Open the dashboard

Navigate to https://<your-domain> in your browser — the nginx vhost installed with the app proxies to OpenObserve and terminates TLS, so the URL carries no port number. Without a domain there is no public entry point: open an SSH tunnel first, ssh -L 5080:127.0.0.1:5080 root@<your-vps-ip>, then browse to http://localhost:5080 (5080 is the default port; use the one shown for your app in your VPS control panel). Log in with the credentials shown in your VPS control panel.

04

Send your first log

From any other server, run: curl -u [email protected]:<password> -XPOST https://<your-domain>/api/default/default/_json -H 'Content-Type: application/json' -d '[{"message":"hello openobserve"}]'. From the VPS itself, target http://127.0.0.1:5080 instead. The log appears instantly in the Logs explorer.

05

Connect an agent (optional)

Install Fluent Bit or Vector on your application servers and point their output at OpenObserve's HTTP API, on the app's own address: https://<your-domain> if a domain is attached, or http://127.0.0.1:5080 for an agent running on the VPS itself. The native OTLP port 5081 is not published by the recipe on any interface — not even locally — so it is not a usable target. Logs and metrics flow in within seconds.

Frequently asked questions

OpenObserve is an open-source observability platform (AGPL-3.0) written in Rust that combines log management, metrics monitoring, distributed tracing, and dashboards in a single binary. It is designed as a lightweight self-hosted alternative to the Elasticsearch/Kibana (ELK) stack and Grafana+Loki+Prometheus setups.

Embed the deploy button

Maintaining a project that uses OpenObserve? This button lets your readers deploy it on a VPS in one click, without reading Docker documentation.

Markdown

[![Deploy OpenObserve on ServOrbit](https://servorbit.com/brand/deploy/button.svg)](https://servorbit.com/vps-cloud?template=openobserve&utm_source=deploy-badge&utm_medium=referral&utm_campaign=openobserve)

HTML

<a href="https://servorbit.com/vps-cloud?template=openobserve&utm_source=deploy-badge&utm_medium=referral&utm_campaign=openobserve"><img src="https://servorbit.com/brand/deploy/button.svg" alt="Deploy OpenObserve on ServOrbit" height="40"></a>

The button points to a VPS order with the template preselected. The image is served from servorbit.com — nothing to host on your side.

Complementary solutions

Monitoring & Observability

Activate OpenObserve on your infrastructure.

Dedicated Cloud VPS — IPv4 included, European datacenter, support included. Your data never leaves your server.

Recommended configuration: 256 MB (1 GB recommended for production workloads) RAM · 1 vCPU