Deploy and Host Blazer on Railway

Blazer is a business intelligence tool for people who know SQL: write a query, get a table, a line or bar chart, a map or a forecast, put queries on dashboards with a shared date-range picker, and turn any query into a check that emails or posts to Slack when it starts failing or spots an anomaly. It connects to Postgres, MySQL, ClickHouse, SQL Server, BigQuery, Snowflake and more. A lightweight Metabase or Redash alternative, and this template runs it on Railway behind a password.

About Hosting Blazer

Two services:

Blazer: the official ankane/blazer image, pinned to a release, on your Railway domain.
Postgres: Blazer's own database for queries, dashboards, checks and the audit log. Private, with a generated password and its own volume.

What the template adds:

Password from the first second. Basic auth with a generated password; only /health is open.
Databases are variables. BLAZER_DATABASE_URL is the data source called "main"; each BLAZER_DATABASE_URL_ adds a data source called ``. clickhouse://user:pass@host:8123/db URLs use Blazer's ClickHouse adapter. No config file to edit, no rebuild.
Checks that actually run. Blazer's checks need a scheduler; here it runs inside the web server ("5 minutes", "1 hour" and "1 day" schedules, plus the daily failing-checks email), so there is no second service to pay for.
Forecasts and anomaly detection in the box. Prophet forecasting and anomaly checks run locally, no data goes to an outside API.

Common Use Cases

SQL dashboards on your production database or a read replica
Alerts when a metric drops, a queue backs up or a number looks wrong
Sharing saved queries with a team instead of pasting SQL in chat

Dependencies for Blazer Hosting

Postgres 17 (included, private network only) for Blazer's own data
The databases you want to query, reachable from Railway

Deployment Dependencies

Implementation Details

Sign in at the Blazer service's domain with BLAZER_USERNAME and BLAZER_PASSWORD from its Variables tab.

Add a database. Set BLAZER_DATABASE_URL on the Blazer service to your database's URL, for example ${{Postgres-App.DATABASE_URL}} for another Railway database in the same project. Use a read-only user: Blazer runs whatever SQL its users type. Until you set one, "main" is Blazer's own database, a place to try it out. More databases: BLAZER_DATABASE_URL_ANALYTICS, BLAZER_DATABASE_URL_BILLING and so on. BLAZER_TIMEOUT (seconds) caps every query.

Variables in queries. {start_time} and {end_time} in a query give it a date-range picker, any other {name} gives it a text box.

Check emails. Set SMTP_URL (smtp://user:password@host:587, or smtps:// for port 465) and BLAZER_FROM_EMAIL. For Slack, set BLAZER_SLACK_WEBHOOK_URL.

Time zone. BLAZER_TIME_ZONE (default Etc/UTC), for example Europe/Warsaw.

Memory. About 0.15 GB for Blazer and 0.05 GB for Postgres at rest.

Upgrades. The template's repository pins the Blazer release in its Dockerfile; a new release there reaches your service on its next redeploy, and migrations run on start.

Why Deploy Blazer on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Blazer on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.