feat(db): attribute Postgres connections by runtime via application_name

[TheodoreSpeaks]

Summary

• Set Postgres application_name per runtime so DB activity is attributable: sim-app (Next.js), sim-trigger (trigger.dev workers), sim-realtime (socket server). Previously every connection reported the postgres.js library default, making it impossible to tell whether lock spikes / advisory-lock contention / statement timeouts come from a request-path query or a background task.
• packages/db/db.ts + apps/realtime/src/database/operations.ts: add connection: { application_name } from DB_APP_NAME (defaults sim-app / sim-realtime).
• apps/sim/trigger.config.ts: declare DB_APP_NAME=sim-trigger in code via the syncEnvVars build extension — the worker's separate deploy scope gets the label without a manual dashboard step or sniffing trigger internals (the obvious TRIGGER_* vars are present in the web app too).
• apps/sim/lib/core/config/env.ts: document DB_APP_NAME.

Validated against prod (sim-studio-production): application_name propagates through PlanetScale's pooler to the backend, and a pg_locks ⋈ pg_stat_activity join attributes locks by app on the existing pg_read_all_data role (no pg_monitor needed). Finer-grained per-request/run comment tags, lock-victim logging, and a Grafana Alloy lock-by-app metric are deferred follow-ups.

Type of Change

• New feature (observability)

Testing

Tested manually. bun run lint, bun run check:api-validation:strict, and the @sim/db vitest suite (19/19) pass. Live verification post-deploy via SELECT application_name, count(*) FROM pg_stat_activity GROUP BY 1.

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
docs	Skipped		Jun 25, 2026 8:15pm

[cursor]

PR Summary

Low Risk
Observability-only connection metadata with safe defaults; no query or auth behavior changes.

Overview
Adds Postgres application_name attribution so pg_stat_activity and lock joins can distinguish Next.js (sim-app), Trigger.dev workers (sim-trigger), realtime (sim-realtime), and migrations (sim-migrate).

Shared @sim/db pools (including replica) now pass connection.application_name from optional DB_APP_NAME, defaulting to sim-app. Realtime sets DB_APP_NAME in bootstrap (before @sim/db loads) and on dev/start scripts; its dedicated socketDb pool uses the same env. Trigger workers get DB_APP_NAME=sim-trigger via syncEnvVars in trigger.config.ts. The migrate script uses a fixed sim-migrate label. DB_APP_NAME is documented in app env config.

^{Reviewed by Cursor Bugbot for commit 5cd82ba. Bugbot is set up for automated code reviews on this repo. Configure here.}

[greptile-apps]

Greptile Summary

This PR adds application_name attribution to all three Postgres connection pools — the Next.js app (sim-app), the trigger.dev background workers (sim-trigger), and the Socket.IO realtime server (sim-realtime) — by passing connection: { application_name } to postgres.js. The value is sourced from a DB_APP_NAME env var with per-service defaults; trigger.dev workers get the label injected unconditionally via syncEnvVars so the worker deploy scope never needs a manual dashboard step.

• packages/db/db.ts: adds application_name to the shared poolOptions object, which propagates to both the primary pool and the read-replica pool.
• apps/realtime/src/database/operations.ts: adds the same connection option to the socket-server pool, defaulting to sim-realtime.
• apps/sim/trigger.config.ts: uses syncEnvVars to inject DB_APP_NAME=sim-trigger at build time so trigger.dev workers are always labelled correctly.
• apps/sim/lib/core/config/env.ts: documents DB_APP_NAME for discoverability, noting it is consumed directly from process.env in @sim/db.

Confidence Score: 5/5

All four changes are additive and observability-only — no query logic, auth, or data path is touched.

The diff is four small additions: a single connection option added to two postgres.js pool configs, a hardcoded env var injected via syncEnvVars into the trigger.dev worker build, and an optional schema entry added for discoverability. The application_name parameter is a standard PostgreSQL connection attribute with no effect on query execution or access control. The author validated the label in production and the trigger/realtime defaults are clearly named. There is no risk of data corruption, security regression, or behavioral change.

No files require special attention.

Important Files Changed

Filename	Overview
packages/db/db.ts	Adds application_name to poolOptions; applies to both primary and replica pools since they spread the same object — correct and intentional.
apps/realtime/src/database/operations.ts	Adds connection.application_name inline to the socket DB pool, defaulting to sim-realtime; reads process.env.DB_APP_NAME directly, consistent with the documented pattern in env.ts.
apps/sim/trigger.config.ts	Uses syncEnvVars to unconditionally inject DB_APP_NAME=sim-trigger into the worker deploy scope; avoids relying on shared env vars that may already carry a different value from the web-app scope.
apps/sim/lib/core/config/env.ts	Adds DB_APP_NAME to the env schema as optional for discoverability; clearly documents that actual consumption is via process.env in @sim/db.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
ENV[DB_APP_NAME env var]
subgraph NextJS["apps/sim (Next.js)"]
DB_PKG["packages/db/db.ts\npoolOptions.connection.application_name\n= process.env.DB_APP_NAME ?? 'sim-app'"]
PRIMARY["Primary pool\napplication_name: sim-app"]
REPLICA["Replica pool\napplication_name: sim-app"]
end
subgraph Trigger["apps/sim (trigger.dev workers)"]
SYNC["syncEnvVars()\ninjects DB_APP_NAME=sim-trigger\nat build time"]
TRIG_POOL["Worker pool\napplication_name: sim-trigger"]
end
subgraph Realtime["apps/realtime (Socket.IO)"]
RT_POOL["socketDb pool\napplication_name: sim-realtime\n(process.env.DB_APP_NAME ?? 'sim-realtime')"]
end
PG[("PostgreSQL\npg_stat_activity\napplication_name column")]
ENV -->|optional override| DB_PKG
ENV -->|optional override| RT_POOL
SYNC -->|unconditional override| TRIG_POOL
DB_PKG --> PRIMARY
DB_PKG --> REPLICA
PRIMARY --> PG
REPLICA --> PG
TRIG_POOL --> PG
RT_POOL --> PG

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
ENV[DB_APP_NAME env var]
subgraph NextJS["apps/sim (Next.js)"]
DB_PKG["packages/db/db.ts\npoolOptions.connection.application_name\n= process.env.DB_APP_NAME ?? 'sim-app'"]
PRIMARY["Primary pool\napplication_name: sim-app"]
REPLICA["Replica pool\napplication_name: sim-app"]
end
subgraph Trigger["apps/sim (trigger.dev workers)"]
SYNC["syncEnvVars()\ninjects DB_APP_NAME=sim-trigger\nat build time"]
TRIG_POOL["Worker pool\napplication_name: sim-trigger"]
end
subgraph Realtime["apps/realtime (Socket.IO)"]
RT_POOL["socketDb pool\napplication_name: sim-realtime\n(process.env.DB_APP_NAME ?? 'sim-realtime')"]
end
PG[("PostgreSQL\npg_stat_activity\napplication_name column")]
ENV -->|optional override| DB_PKG
ENV -->|optional override| RT_POOL
SYNC -->|unconditional override| TRIG_POOL
DB_PKG --> PRIMARY
DB_PKG --> REPLICA
PRIMARY --> PG
REPLICA --> PG
TRIG_POOL --> PG
RT_POOL --> PG

Loading

_{Reviews (1): Last reviewed commit: "feat(db): attribute Postgres connections..." | Re-trigger Greptile}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

There are 2 total unresolved issues (including 1 from previous review).

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit b5c2995. Configure here.}

[TheodoreSpeaks]

Addressed the Bugbot finding (realtime traffic mislabeled sim-app) in 5cd82ba.

Root cause: the realtime process opens two Postgres clients — its own socketDb pool and the shared @sim/db client (used by handlers/{subblocks,workflow,variables}, database/preflight, middleware/permissions). Only socketDb was labeled, so the shared client fell back to sim-app.

Fix: set DB_APP_NAME=sim-realtime at the process level so both clients pick it up, mirroring how the trigger worker sets sim-trigger:

• bootstrap.ts (prod entry): process.env.DB_APP_NAME ??= 'sim-realtime' after loadRuntimeSecrets() and before the dynamic import('@/index'), so it lands before @sim/db reads it at module-eval time. ??= respects an explicit override.
• package.jsondev/start scripts: prefixed DB_APP_NAME=sim-realtime for the local path that runs src/index.ts directly (bypasses bootstrap).

Now every connection the realtime process opens reports sim-realtime. Verified with bun run --filter @sim/realtime type-check and bun run lint.