Introduction   •   Docs   •   Features   •   Benchmarks     

Parseable is an open source, columnar data lake platform - purpose built for observability. Send logs, metrics & traces to Parseable via popular logging agents, OpenTelemetry, Kafka, eBPF or other integrations. Use native observability features like alerting, dashboards, anomaly detection, APM, PromQL & SQL - all within a single binary.

Why Parseable?

Purpose built for observability and designed around proven data engineering patterns, Parseable gives you everything you need to make sense of your telemetry data, right out of the box.

Some of the key highlights include:

• Data lake architecture: Stateless compute over object storage as the backing store. Storage and compute scale independently, so you're not paying for one to grow the other.

• Fully featured: Parseable is feature complete with alerting, dashboards, anomaly detection, APM, and more, all from a single binary without stitching together multiple tools.

• Agent ready: Whether you need to observe your AI agents or use LLMs to analyze your telemetry data, Parseable supports both natively.

• OpenTelemetry native: With native OTel support, you can send telemetry data to Parseable without any custom modifications or plugins. Parseable can be used as a drop-in replacement for your existing OpenTelemetry Collector setup.

Quickstart

Download the latest published binary from this fork (VibeEngineeringAI/parseable). The installer requires a published release containing the asset for your platform; it will fail if one is unavailable.

curl -fsSL https://raw.githubusercontent.com/VibeEngineeringAI/parseable/main/scripts/download.sh | sh

For Windows

powershell -c "irm https://raw.githubusercontent.com/VibeEngineeringAI/parseable/main/scripts/download.ps1 | iex"

Once you have Parseable running, ingest data with the command below. This will send logs to the demo stream. You can see the logs in the dashboard.

curl --location --request POST 'http://localhost:8000/api/v1/ingest' \
--header 'X-P-Stream: demo' \
--header 'Authorization: Basic YWRtaW46YWRtaW4=' \
--header 'Content-Type: application/json' \
--data-raw '[
    {
        "id": "434a5f5e-2f5f-11ed-a261-0242ac120002",
        "datetime": "24/Jun/2022:14:12:15 +0000",
        "host": "153.10.110.81"
    }
]'

Access the UI at http://localhost:8000. Log in with the default credentials admin / admin.

Source builds of this fork include a community PromQL query API for OTLP metrics, with instant and range queries and no Enterprise dependency. See the guide for the supported language subset and limits.

For production deployments, refer to the installation guide ↗︎ for best practices and hardening tips.

Tip

Try out the Parseable cloud — 14 days free trial, no credit card required.

Contributing

Pull requests use the CI workflow on GitHub-hosted Ubuntu 24.04 runners. It selects frontend, Rust, and integration jobs from the changed paths and reports one aggregate CI passed check. Frontend changes still exercise the embedded UI through the integration image; Rust-only changes skip the browser suite. A newer PR commit cancels its superseded CI run.

Browser tests run in four shards alongside production and Storybook checks. Clippy runs alongside Rust tests. Integration builds one cached Docker image and shares it between standalone and distributed Quest jobs. PRs keep Quest's functional tests but exclude its six sustained k6 tests. Pushes to main and the nightly run (06:17 UTC) run the full Quest load suite and upload Rust coverage. Manual CI runs can select full to exercise those checks before merging.

Release builds cache Rust dependencies per target, feature set, and profile. The manual release workflow defaults to the existing fat-LTO profile. Its release-thin-lto option is an experiment that only uploads workflow artifacts; it accepts a tag, branch, or commit containing that Cargo profile and never publishes release assets or images. Compare build times and runtime benchmarks on the same source revision before changing the production profile.

Contribution guide ↗︎

Verify images

Parseable builds are attested for build provenance and integrity using the attest-build-provenance action. The attestations can be verified by having the latest version of GitHub CLI installed in your system. Then, execute the following command:

gh attestation verify PATH/TO/YOUR/PARSEABLE/ARTIFACT-BINARY -R parseablehq/parseable

Editable frontend

The React/TypeScript frontend and reusable component library live in frontend/, with Logs Explore, SQL, datasets, dashboards, Storybook, and browser tests. Billing and payment collection are excluded.

Parseable embeds only this native frontend and serves it at /. Links to the earlier /next/ mount redirect to the same path at the root. Classic Prism assets are no longer downloaded or embedded; P_UI_ROOT=classic is rejected with migration instructions. Docker images and build/release workflows build native assets with Node 24 and embed frontend/dist through NEXT_ASSETS_PATH. To build locally from the repository root:

(cd frontend && npm ci && npm run build)
NEXT_ASSETS_PATH="$PWD/frontend/dist" cargo build --release

Missing native assets fail the build. For backend-only debug development, explicitly use P_BUILD_NO_UI=1 cargo run -- local-store with NEXT_ASSETS_PATH unset; UI requests then explain that the UI is disabled. This mode cannot be used for release builds. /api/v1/about reports native-<server version> or disabled as the UI version, alongside the server commit/build metadata.

To run this checkout in Docker:

docker compose -f compose.yaml -f compose.source.yaml up -d --build

docker compose up alone uses the prebuilt upstream image from compose.yaml; it does not build this checkout and can still show the old UI. The source override uses the existing named data/staging volumes.

See the native UI retirement notes and frontend build instructions for compatibility, backend development and PowerShell details.