Supabase Log Drains Hit Pro: Full-Stack Observability Without the Enterprise Price Tag

Essential Points

• Supabase Log Drains, previously restricted to Team and Enterprise tiers, are now available as a Pro plan add-on
• Supported destinations include Datadog, Loki, Sentry, Axiom, AWS S3, OTLP, and any generic HTTP endpoint
• Logs from Postgres, Auth, Storage, Edge Functions, Realtime, and the API Gateway are all supported
• Each drain costs $60 per month, plus $0.20 per million log events and $0.09 per GB egress

Supabase just removed one of the biggest reasons developers outgrow the Pro plan before they outgrow the product. Log Drains, a feature that continuously pushes your entire infrastructure log stack into the observability tool your team already uses, moved from Team and Enterprise to Pro on March 4, 2026. This guide breaks down exactly what changed, what it costs, how to configure each destination, and what the pricing means in practice.

What Supabase Log Drains Actually Do

A Log Drain is a continuous export pipeline. Instead of manually querying Supabase’s built-in Logs Explorer, you configure a drain to push logs automatically to an external destination the moment they are generated. Logs are batched in groups of up to 250 events or flushed every second, whichever comes first, and compressed with gzip when the destination supports it.

This matters because production debugging rarely lives in a single tool. Your application errors are in Sentry, your infrastructure metrics are in Datadog, and your database queries were previously isolated inside Supabase Studio. Log Drains consolidate all of that into one place.

Also Read Supabase Storage Rebuilt From the Ground Up: Faster, More Secure, More Reliable

Which Services Send Logs Through a Drain

Every major Supabase service participates in the drain pipeline:

• Postgres: Query logs, connection events, database-level activity
• Auth: Login attempts, token events, failure reasons
• Storage: File operation logs and access patterns
• Edge Functions: Invocation traces and execution errors
• Realtime: Connection and subscription activity
• API Gateway: Request paths, status codes, and latency data

Previously, correlating a 500 error from your API Gateway with a Postgres event required switching between separate log tabs in Supabase Studio and matching timestamps manually. With a drain configured, both events land in the same dashboard your team already monitors.

All 7 Supported Destinations

Supabase currently supports six named destinations plus a universal generic HTTP endpoint:

The OTLP destination is compatible with OpenTelemetry Collector, Grafana Cloud, New Relic, Honeycomb, Datadog OTLP ingestion, Elastic, and more. The generic HTTP endpoint option allows teams to route logs through an Edge Function, filter or restructure them, and forward to any tool not natively listed.

How to Set Up Each Drain

All drain configuration happens inside the Supabase dashboard under Project Settings > Log Drains No CLI commands or external scripts are required.

Datadog: Generate a Datadog API key and select your Datadog site region. Logs are gzip-compressed before sending and tagged with the service name on the service field .

Loki: Provide your Loki HTTP API URL and any required headers. Loki must be configured to accept structured metadata. Supabase recommends increasing the default maximum structured metadata fields to at least 500 to accommodate large log payloads.

Sentry: Grab your DSN from Sentry project settings (format: {PROTOCOL}://{PUBLIC_KEY}:{SECRET_KEY}@{HOST}{PATH}/{PROJECT_ID}). Logs are ingested as Sentry Logs, not as errors. Self-hosted Sentry requires version 25.9.0 or later.

Axiom setup steps:

1. Create a dataset in Axiom Console under Datasets
2. Generate an Axiom API token with ingest permission for that dataset
3. Enter the dataset name and API token in the Supabase dashboard
4. Verify events appear in the Axiom Console Stream panel

Amazon S3 required fields:

• S3 Bucket name (must already exist)
• AWS Region where the bucket is located
• Access Key ID and Secret Access Key with write permissions
• Batch Timeout in milliseconds (recommended: 2000 to 5000ms)

OTLP required fields:

• Endpoint URL ending in /v1/logs
• Protocol: currently http/protobuf only
• Gzip: recommended enabled to reduce bandwidth
• Optional authentication headers such as Authorization or X-API-Key 

Pricing Breakdown: What $60/Month Gets You

The cost structure is additive:

• Base cost: $60 per drain per month
• Event volume: $0.20 per million log events processed
• Egress: $0.09 per GB of log data transferred

This is the same pricing structure that launched with the feature for Team and Enterprise in August 2024. The Pro plan base cost is $25 per month. A team running two drains (for example, Datadog for alerting and S3 for archival) starts at $120/month in drain fees before factoring in event volume. For reference, the Pro plan includes 7-day log retention by default. Routing to S3 gives teams unlimited archival at S3 storage rates, which is relevant for audit trail and compliance needs.

How Log Drains Are Built: The Logflare Architecture

Log Drains are built into Logflare, the analytics and observability server underlying the Supabase platform. Supabase rewrote Logflare’s architecture using a multi-backend V2 pipeline to enable efficient, scalable log dispatching to multiple simultaneous destinations. The BEAM runtime powering Logflare enables soft-realtime dispatching, meaning any configured drain receives log events as fast as, or faster than, they appear in the Supabase Logs UI.

When Log Drains Were First Available

Supabase first launched Log Drains in August 2024, exclusively for Team and Enterprise customers. The initial release supported only two destinations: Datadog and a generic HTTP endpoint. Since that launch, Supabase expanded the destination list to seven options, and as of March 4, 2026, opened access to Pro plan customers as an add-on.

Limitations Worth Knowing

Log Drains are priced as add-ons, not included in the Pro plan base cost. Teams running multiple drains on high-event-volume projects will see costs scale accordingly. The Free plan has no access to Log Drains. Generic HTTP endpoint requests are currently sent unsigned, though Supabase has noted that signed requests are planned for a future update.

Also Read AI Agents Are Breaking Everything: What the “Agents of Chaos” Study Actually Found

Frequently Asked Questions (FAQs)