Observability

Monitoring

Monitoring

Live traffic, errors and database load for one Supabase project, minute by minute.

Monitoring shows the current activity of a single Supabase project at minute resolution: traffic, errors, and database load.

Location: Monitoring in the sidebar, then select a project. Available to workspace owners and admins.

The tabs

Services

One tab per Supabase service in use, such as the Data API, Auth, Storage, Realtime, and Edge Functions. Each tab shows requests, errors, busiest routes, and slowest queries.

Monitoring PostgREST tab with request volume and latency

Errors

Postgres errors grouped by SQLSTATE code, such as statement timeouts, deadlocks, permission denied, and too many connections. Dreambase stores error counts only; error messages are never stored because they can contain row values.

Monitoring Errors tab with Postgres errors grouped by SQLSTATE

Database

Database server metrics, sampled every minute:

  • CPU, memory and disk

  • Connections and poolers, and connections by role

  • WAL and transaction ID age

  • Query load, deadlocks and temp files

  • Data API pool timeouts and session totals

Monitoring Database tab with CPU, memory, disk and connection charts

Ranges and freshness

Available ranges are 1H, 24H, and 7D. A range is offered only when the project's history covers it.

Server metrics arrive every minute. Logs are read hourly; the newest log data can lag by up to about 90 minutes.

Compare periods

+ Compare opens Benchmark, which compares two periods, such as before and after a deploy, and lists the statements that became busier or slower.

Tool reference

get_supabase_metrics

MCP and REST · health-reports:read

{ workspaceId, connectionId, view?, range?, end?, branchRef?, previous? }

The project's server metrics over 1H, 24H or 7D. Pass branchRef for a branch, and previous: true to get the period before for comparison.

get_supabase_workloads

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to? }

The project's busiest and slowest statements. range is 1H, 24H, 7D, 30D or 90D, or give from and to.

get_supabase_routes

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to?, limit?, cursor? }

The busiest API routes over the last 24 hours by default, with their callers, error counts and p95 latency, and the routes failing most.

compare_supabase_periods

MCP and REST · health-reports:read

{ workspaceId, connectionId, aFrom, aTo, bFrom, bTo, by?, order?, queryId?, role?, limit? }

Compares two periods by statement or segment, ordered by busiest or change. Narrow to one statement with queryId and role.

get_supabase_events

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to?, kinds?, limit?, cursor? }

What changed on the project over the last 7 days by default: migrations, schema changes, function deploys, resizes, upgrades, setting changes, backups, restarts, out-of-memory kills and branches.

Read metrics through the API

curl "https://api.dreambase.com/v1/workspaces/<workspace-id>/connections/supabase/<connection-id>/metrics?range=24H" \
  -H "Authorization: Bearer <access-token>"

REST paths sit under /workspaces/{workspace_id}/connections/supabase/{connection_id}/.

  • Observability — How Observability collects and grades.

  • Projects — Health, recommendations and alerts.

  • Triggers — Alert when a signal crosses a threshold.

Important boundaries

  • Server metrics require a project on a paid Supabase plan.

  • Minute data is retained for 14 days, hourly rollups for 90 days, and daily rollups for 13 months.

  • Log data can lag by up to about 90 minutes.

  • Error messages are never stored; only codes and counts are kept.

Monitoring shows the current activity of a single Supabase project at minute resolution: traffic, errors, and database load.

Location: Monitoring in the sidebar, then select a project. Available to workspace owners and admins.

The tabs

Services

One tab per Supabase service in use, such as the Data API, Auth, Storage, Realtime, and Edge Functions. Each tab shows requests, errors, busiest routes, and slowest queries.

Monitoring PostgREST tab with request volume and latency

Errors

Postgres errors grouped by SQLSTATE code, such as statement timeouts, deadlocks, permission denied, and too many connections. Dreambase stores error counts only; error messages are never stored because they can contain row values.

Monitoring Errors tab with Postgres errors grouped by SQLSTATE

Database

Database server metrics, sampled every minute:

  • CPU, memory and disk

  • Connections and poolers, and connections by role

  • WAL and transaction ID age

  • Query load, deadlocks and temp files

  • Data API pool timeouts and session totals

Monitoring Database tab with CPU, memory, disk and connection charts

Ranges and freshness

Available ranges are 1H, 24H, and 7D. A range is offered only when the project's history covers it.

Server metrics arrive every minute. Logs are read hourly; the newest log data can lag by up to about 90 minutes.

Compare periods

+ Compare opens Benchmark, which compares two periods, such as before and after a deploy, and lists the statements that became busier or slower.

Tool reference

get_supabase_metrics

MCP and REST · health-reports:read

{ workspaceId, connectionId, view?, range?, end?, branchRef?, previous? }

The project's server metrics over 1H, 24H or 7D. Pass branchRef for a branch, and previous: true to get the period before for comparison.

get_supabase_workloads

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to? }

The project's busiest and slowest statements. range is 1H, 24H, 7D, 30D or 90D, or give from and to.

get_supabase_routes

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to?, limit?, cursor? }

The busiest API routes over the last 24 hours by default, with their callers, error counts and p95 latency, and the routes failing most.

compare_supabase_periods

MCP and REST · health-reports:read

{ workspaceId, connectionId, aFrom, aTo, bFrom, bTo, by?, order?, queryId?, role?, limit? }

Compares two periods by statement or segment, ordered by busiest or change. Narrow to one statement with queryId and role.

get_supabase_events

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to?, kinds?, limit?, cursor? }

What changed on the project over the last 7 days by default: migrations, schema changes, function deploys, resizes, upgrades, setting changes, backups, restarts, out-of-memory kills and branches.

Read metrics through the API

curl "https://api.dreambase.com/v1/workspaces/<workspace-id>/connections/supabase/<connection-id>/metrics?range=24H" \
  -H "Authorization: Bearer <access-token>"

REST paths sit under /workspaces/{workspace_id}/connections/supabase/{connection_id}/.

  • Observability — How Observability collects and grades.

  • Projects — Health, recommendations and alerts.

  • Triggers — Alert when a signal crosses a threshold.

Important boundaries

  • Server metrics require a project on a paid Supabase plan.

  • Minute data is retained for 14 days, hourly rollups for 90 days, and daily rollups for 13 months.

  • Log data can lag by up to about 90 minutes.

  • Error messages are never stored; only codes and counts are kept.

Monitoring shows the current activity of a single Supabase project at minute resolution: traffic, errors, and database load.

Location: Monitoring in the sidebar, then select a project. Available to workspace owners and admins.

The tabs

Services

One tab per Supabase service in use, such as the Data API, Auth, Storage, Realtime, and Edge Functions. Each tab shows requests, errors, busiest routes, and slowest queries.

Monitoring PostgREST tab with request volume and latency

Errors

Postgres errors grouped by SQLSTATE code, such as statement timeouts, deadlocks, permission denied, and too many connections. Dreambase stores error counts only; error messages are never stored because they can contain row values.

Monitoring Errors tab with Postgres errors grouped by SQLSTATE

Database

Database server metrics, sampled every minute:

  • CPU, memory and disk

  • Connections and poolers, and connections by role

  • WAL and transaction ID age

  • Query load, deadlocks and temp files

  • Data API pool timeouts and session totals

Monitoring Database tab with CPU, memory, disk and connection charts

Ranges and freshness

Available ranges are 1H, 24H, and 7D. A range is offered only when the project's history covers it.

Server metrics arrive every minute. Logs are read hourly; the newest log data can lag by up to about 90 minutes.

Compare periods

+ Compare opens Benchmark, which compares two periods, such as before and after a deploy, and lists the statements that became busier or slower.

Tool reference

get_supabase_metrics

MCP and REST · health-reports:read

{ workspaceId, connectionId, view?, range?, end?, branchRef?, previous? }

The project's server metrics over 1H, 24H or 7D. Pass branchRef for a branch, and previous: true to get the period before for comparison.

get_supabase_workloads

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to? }

The project's busiest and slowest statements. range is 1H, 24H, 7D, 30D or 90D, or give from and to.

get_supabase_routes

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to?, limit?, cursor? }

The busiest API routes over the last 24 hours by default, with their callers, error counts and p95 latency, and the routes failing most.

compare_supabase_periods

MCP and REST · health-reports:read

{ workspaceId, connectionId, aFrom, aTo, bFrom, bTo, by?, order?, queryId?, role?, limit? }

Compares two periods by statement or segment, ordered by busiest or change. Narrow to one statement with queryId and role.

get_supabase_events

MCP and REST · health-reports:read

{ workspaceId, connectionId, range?, from?, to?, kinds?, limit?, cursor? }

What changed on the project over the last 7 days by default: migrations, schema changes, function deploys, resizes, upgrades, setting changes, backups, restarts, out-of-memory kills and branches.

Read metrics through the API

curl "https://api.dreambase.com/v1/workspaces/<workspace-id>/connections/supabase/<connection-id>/metrics?range=24H" \
  -H "Authorization: Bearer <access-token>"

REST paths sit under /workspaces/{workspace_id}/connections/supabase/{connection_id}/.

  • Observability — How Observability collects and grades.

  • Projects — Health, recommendations and alerts.

  • Triggers — Alert when a signal crosses a threshold.

Important boundaries

  • Server metrics require a project on a paid Supabase plan.

  • Minute data is retained for 14 days, hourly rollups for 90 days, and daily rollups for 13 months.

  • Log data can lag by up to about 90 minutes.

  • Error messages are never stored; only codes and counts are kept.

Projects

API Connectors

© 2026 Dream, Inc. All rights reserved.