Supabase Integration
Monitor Postgres connection pooling, auth request metrics, and realtime subscription health across your Supabase projects. Get AI-powered root cause analysis spanning every layer of the Supabase stack.
How It Works
Connect via Supabase Metrics API
TigerOps connects to your Supabase project metrics endpoint using your service role key. Both self-hosted Supabase and Supabase Cloud (Pro plan and above) expose Prometheus-compatible metrics.
Enable Postgres Exporter
For self-hosted deployments, enable the bundled postgres_exporter in your Supabase stack config. For Supabase Cloud, TigerOps pulls PgBouncer and Postgres metrics via the platform metrics endpoint.
Configure Service-Level Monitoring
TigerOps auto-discovers GoTrue (Auth), PostgREST, Realtime, and Storage service metrics from the Supabase stack. Each service is tracked individually with per-endpoint latency and error rate visibility.
Set Pooler and Auth Alerts
Define thresholds for PgBouncer pool saturation, auth request p99 latency, and realtime channel subscription count. TigerOps correlates pooler exhaustion with upstream API Gateway traffic spikes.
What You Get Out of the Box
Postgres Connection Pool Monitoring
PgBouncer active/waiting client counts, pool saturation per database, transaction mode vs session mode pool efficiency, and server connection utilization. Alert before pooler exhaustion drops requests.
Auth Service Request Metrics
GoTrue endpoint request rates, p50/p95/p99 latency per auth route (signup, signin, token refresh), error rates, and active JWT session counts. Track auth degradation before users notice.
Realtime Subscription Health
Active channel subscriptions, message broadcast rates, subscription error rates, and Phoenix channel join latency. Detect realtime server memory pressure and connection churn automatically.
PostgREST API Performance
Per-endpoint request counts, response latency, schema cache hit ratio, and connection hold time. Identify which REST API routes are executing the most expensive Postgres queries.
Storage Bucket & Upload Metrics
Object storage request rates, upload/download throughput, multipart upload completion rates, and bucket size growth. Track storage ingestion patterns and CDN cache bypass rates.
AI Root Cause Analysis
When Supabase API latency spikes, TigerOps AI correlates PgBouncer pool depth, PostgREST schema cache staleness, and GoTrue database queries to identify whether the bottleneck is connection saturation or query performance.
TigerOps Config for Supabase
Connect TigerOps to your Supabase Cloud project or self-hosted deployment metrics endpoints.
# TigerOps Supabase integration config
# Place at /etc/tigerops/conf.d/supabase.yaml
integrations:
- name: supabase
type: supabase
config:
# Supabase Cloud: use platform metrics endpoint
# Self-hosted: use internal service URLs
project_ref: "${SUPABASE_PROJECT_REF}"
service_role_key: "${SUPABASE_SERVICE_ROLE_KEY}"
# Services to monitor
services:
postgres: true # Postgres + pg_stat_* metrics
pgbouncer: true # Connection pool metrics
gotrue: true # Auth service metrics
realtime: true # Websocket subscription metrics
postgrest: true # REST API metrics
storage: true # Object storage metrics
edge_functions: true # Deno runtime metrics
# Postgres monitoring config
postgres:
collect_pg_stat_statements: true
top_query_count: 25
track_io_timing: true
# PgBouncer alert thresholds
pgbouncer:
max_wait_seconds_alert: 5
pool_saturation_percent_alert: 80
scrape_interval: 15s
remote_write:
endpoint: https://ingest.atatus.net/api/v1/write
bearer_token: "${TIGEROPS_API_KEY}"Common Questions
Does TigerOps work with Supabase Cloud (managed)?
Yes. Supabase Cloud Pro plans and above expose a Prometheus metrics endpoint. TigerOps connects using your Supabase service role key to collect Postgres, PgBouncer, and service-level metrics without requiring SSH access to the underlying infrastructure.
How does TigerOps monitor PgBouncer connection pools?
TigerOps collects pgbouncer_pools metrics including cl_active, cl_waiting, sv_active, sv_idle, and maxwait. It computes pool saturation percentage and fires predictive alerts when the waiting queue growth rate indicates imminent connection refusals.
Can TigerOps monitor self-hosted Supabase deployments?
Yes. Self-hosted Supabase via Docker Compose or Kubernetes exposes the same Prometheus metrics as the Cloud platform. Point the TigerOps agent at your internal metrics endpoints and all service components are discovered automatically.
How are Supabase Realtime alerts handled?
TigerOps monitors Phoenix channel message rates, subscription counts, and process memory on Realtime nodes. Alerts fire when subscription counts exceed your configured limit or when message delivery latency exceeds your SLO — before clients start experiencing dropped messages.
Does TigerOps track Supabase Edge Function metrics?
Yes. TigerOps collects Deno runtime metrics from Supabase Edge Functions including invocation counts, cold start frequency, execution duration histograms, and memory usage per function. Slow or error-prone functions are surfaced automatically.
Get Complete Visibility Into Your Supabase Stack
Connection pool monitoring, auth metrics, realtime health, and AI root cause analysis. Deploy in 5 minutes.