github-featured/triggerdotdev--trigger.dev
2
0
Fork 0
Code Issues Pull Requests Deployments Models Agent Notes Packages Projects Releases Wiki Activity

feat(webapp): per-client database pool and connect timeout overrides (#4515)

Browse Source 
## Summary

Follow-on to #4513. The database connect timeout is now honored, but a
single global value has to serve three separate databases at once
(control-plane, legacy run-ops, and run-ops). This adds optional
per-client overrides for the Prisma pool and connect timeouts, one pair
for the writer and one for the read replica of each of the three
databases, each falling back to the shared `DATABASE_POOL_TIMEOUT` /
`DATABASE_CONNECTION_TIMEOUT` when unset.

That lets one database's clients run a fail-fast connect timeout (with a
bounded pool wait) while another keeps more headroom, without a single
knob forcing the same tradeoff everywhere. No behavior change until an
override is set.

It also tags each client's queries with its specific datasource
(`control-plane` / `legacy-run-ops` / `run-ops`, writer or replica) via
the `db.datasource` span attribute, so telemetry can attribute
connection behavior to a specific database instead of just
writer-vs-replica.
This commit is contained in:
Eric Allam
2026-08-05 17:26:41 +01:00
committed by GitHub
parent 1a16d61a37
commit 58bf4e2833
4 changed files with 97 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.server-changes/per-client-connection-timeouts.md
+6
View File
@@ -0,0 +1,6 @@
---
area: webapp
type: improvement
---
Allow the database connection pool and connect timeouts to be tuned separately for each database's writer and read replica, falling back to the shared defaults when unset.
apps/webapp/app/db.server.ts
+68 -18
View File
@@ -124,7 +124,15 @@ async function $transactionInner<R>(
export { Prisma };
function tagDatasource<T extends PrismaClient>(datasource: "writer" | "replica", client: T): T {
type DatasourceLabel =
| "control-plane-writer"
| "control-plane-replica"
| "legacy-run-ops-writer"
| "legacy-run-ops-replica"
| "run-ops-writer"
| "run-ops-replica";
function tagDatasource<T extends PrismaClient>(datasource: DatasourceLabel, client: T): T {
return client.$extends({
name: "datasource-tagger",
query: {
@@ -142,7 +150,7 @@ function tagDatasource<T extends PrismaClient>(datasource: "writer" | "replica",
// Same extension as tagDatasource but typed for RunOpsPrismaClient (different
// generated package — does not extend @trigger.dev/database.PrismaClient).
function tagDatasourceRunOps(
datasource: "writer" | "replica",
datasource: DatasourceLabel,
client: RunOpsPrismaClient
): RunOpsPrismaClient {
return client.$extends({
@@ -168,7 +176,7 @@ function captureInfraErrorsRunOps(client: RunOpsPrismaClient): RunOpsPrismaClien
}
export const prisma = singleton("prisma", () =>
captureInfrastructureErrors(tagDatasource("writer", getClient()))
captureInfrastructureErrors(tagDatasource("control-plane-writer", getClient()))
);
export const $replica: PrismaReplicaClient = singleton("replica", () => {
@@ -176,7 +184,9 @@ export const $replica: PrismaReplicaClient = singleton("replica", () => {
// Brand ONLY a real replica so the run-store routing layer keeps replica reads off the primary.
// No replica configured → fall back to the writer `prisma`, which must stay UNBRANDED.
return replica
? markReadReplicaClient(captureInfrastructureErrors(tagDatasource("replica", replica)))
? markReadReplicaClient(
captureInfrastructureErrors(tagDatasource("control-plane-replica", replica))
)
: prisma;
});
@@ -296,7 +306,7 @@ const runOpsTopology: RunOpsTopology = singleton("runOpsTopology", () => {
controlPlane: { writer: prisma, replica: $replica },
buildNewWriter: (url, clientType) =>
captureInfraErrorsRunOps(
tagDatasourceRunOps("writer", buildRunOpsWriterClient({ url, clientType }))
tagDatasourceRunOps("run-ops-writer", buildRunOpsWriterClient({ url, clientType }))
),
// Brand the run-ops replica (only built for a real replica URL) so routed replica reads stay
// off the primary. When no replica URL is set, selectRunOpsTopology reuses the writer here —
@@ -304,19 +314,35 @@ const runOpsTopology: RunOpsTopology = singleton("runOpsTopology", () => {
buildNewReplica: (url, clientType) =>
markReadReplicaClient(
captureInfraErrorsRunOps(
tagDatasourceRunOps("replica", buildRunOpsReplicaClient({ url, clientType }))
tagDatasourceRunOps("run-ops-replica", buildRunOpsReplicaClient({ url, clientType }))
)
),
// Legacy client shares the exact control-plane wrapper stack (the legacy DB carries the full
// control-plane schema); markReadReplicaClient only on a real replica URL, as with the NEW replica.
buildLegacyWriter: (url, clientType) =>
captureInfrastructureErrors(
tagDatasource("writer", buildWriterClient({ url, clientType }))
tagDatasource(
"legacy-run-ops-writer",
buildWriterClient({
url,
clientType,
poolTimeout: env.RUN_OPS_LEGACY_DATABASE_WRITER_POOL_TIMEOUT,
connectTimeout: env.RUN_OPS_LEGACY_DATABASE_WRITER_CONNECTION_TIMEOUT,
})
)
),
buildLegacyReplica: (url, clientType) =>
markReadReplicaClient(
captureInfrastructureErrors(
tagDatasource("replica", buildReplicaClient({ url, clientType }))
tagDatasource(
"legacy-run-ops-replica",
buildReplicaClient({
url,
clientType,
poolTimeout: env.RUN_OPS_LEGACY_DATABASE_READ_REPLICA_POOL_TIMEOUT,
connectTimeout: env.RUN_OPS_LEGACY_DATABASE_READ_REPLICA_CONNECTION_TIMEOUT,
})
)
)
),
}
@@ -383,7 +409,12 @@ function getClient() {
const url = env.CONTROL_PLANE_DATABASE_URL ?? env.DATABASE_URL;
invariant(typeof url === "string", "neither CONTROL_PLANE_DATABASE_URL nor DATABASE_URL is set");
return buildWriterClient({ url, clientType: "writer" });
return buildWriterClient({
url,
clientType: "writer",
poolTimeout: env.DATABASE_WRITER_POOL_TIMEOUT,
connectTimeout: env.DATABASE_WRITER_CONNECTION_TIMEOUT,
});
}
// Generalized writer builder shared by the control-plane client and the run-ops
@@ -392,14 +423,18 @@ function getClient() {
export function buildWriterClient({
url,
clientType,
poolTimeout,
connectTimeout,
}: {
url: string;
clientType: string;
poolTimeout?: number;
connectTimeout?: number;
}): PrismaClient {
const databaseUrl = buildPrismaConnectionUrl(url, {
connectionLimit: env.DATABASE_CONNECTION_LIMIT.toString(),
poolTimeout: env.DATABASE_POOL_TIMEOUT.toString(),
connectTimeout: env.DATABASE_CONNECTION_TIMEOUT.toString(),
poolTimeout: (poolTimeout ?? env.DATABASE_POOL_TIMEOUT).toString(),
connectTimeout: (connectTimeout ?? env.DATABASE_CONNECTION_TIMEOUT).toString(),
applicationName: env.SERVICE_NAME,
});
@@ -530,7 +565,12 @@ function getReplicaClient() {
return;
}
return buildReplicaClient({ url, clientType: "reader" });
return buildReplicaClient({
url,
clientType: "reader",
poolTimeout: env.DATABASE_READ_REPLICA_POOL_TIMEOUT,
connectTimeout: env.DATABASE_READ_REPLICA_CONNECTION_TIMEOUT,
});
}
// Generalized replica builder shared by the control-plane replica and the run-ops
@@ -539,14 +579,18 @@ function getReplicaClient() {
export function buildReplicaClient({
url,
clientType,
poolTimeout,
connectTimeout,
}: {
url: string;
clientType: string;
poolTimeout?: number;
connectTimeout?: number;
}): PrismaClient {
const replicaUrl = buildPrismaConnectionUrl(url, {
connectionLimit: env.DATABASE_CONNECTION_LIMIT.toString(),
poolTimeout: env.DATABASE_POOL_TIMEOUT.toString(),
connectTimeout: env.DATABASE_CONNECTION_TIMEOUT.toString(),
poolTimeout: (poolTimeout ?? env.DATABASE_POOL_TIMEOUT).toString(),
connectTimeout: (connectTimeout ?? env.DATABASE_CONNECTION_TIMEOUT).toString(),
applicationName: env.SERVICE_NAME,
});
@@ -675,8 +719,10 @@ function buildRunOpsWriterClient({
}): RunOpsPrismaClient {
const databaseUrl = buildPrismaConnectionUrl(url, {
connectionLimit: env.DATABASE_CONNECTION_LIMIT.toString(),
poolTimeout: env.DATABASE_POOL_TIMEOUT.toString(),
connectTimeout: env.DATABASE_CONNECTION_TIMEOUT.toString(),
poolTimeout: (env.RUN_OPS_DATABASE_WRITER_POOL_TIMEOUT ?? env.DATABASE_POOL_TIMEOUT).toString(),
connectTimeout: (
env.RUN_OPS_DATABASE_WRITER_CONNECTION_TIMEOUT ?? env.DATABASE_CONNECTION_TIMEOUT
).toString(),
applicationName: env.SERVICE_NAME,
});
@@ -728,8 +774,12 @@ function buildRunOpsReplicaClient({
connectionLimit: (
env.RUN_OPS_DATABASE_READ_REPLICA_CONNECTION_LIMIT ?? env.DATABASE_CONNECTION_LIMIT
).toString(),
poolTimeout: env.DATABASE_POOL_TIMEOUT.toString(),
connectTimeout: env.DATABASE_CONNECTION_TIMEOUT.toString(),
poolTimeout: (
env.RUN_OPS_DATABASE_READ_REPLICA_POOL_TIMEOUT ?? env.DATABASE_POOL_TIMEOUT
).toString(),
connectTimeout: (
env.RUN_OPS_DATABASE_READ_REPLICA_CONNECTION_TIMEOUT ?? env.DATABASE_CONNECTION_TIMEOUT
).toString(),
applicationName: env.SERVICE_NAME,
});
apps/webapp/app/env.server.ts
+19 -1
View File
@@ -108,6 +108,12 @@ const isNotInsecureSecret = (value: string) =>
const INSECURE_SECRET_MESSAGE =
"must not be a known-insecure published default; set a strong, unique value. If you cannot rotate it yet (e.g. it protects existing encrypted data or active sessions), set ALLOW_INSECURE_DEFAULT_SECRETS=1 to boot while you migrate.";
/** Optional int env var; blank/whitespace normalises to undefined (z.coerce turns "" into 0). */
const OptionalIntEnv = z.preprocess(
(v) => (typeof v === "string" && v.trim() === "" ? undefined : v),
z.coerce.number().int().optional()
);
const EnvironmentSchema = z
.object({
NODE_ENV: z.union([z.literal("development"), z.literal("production"), z.literal("test")]),
@@ -120,6 +126,10 @@ const EnvironmentSchema = z
DATABASE_CONNECTION_LIMIT: z.coerce.number().int().default(10),
DATABASE_POOL_TIMEOUT: z.coerce.number().int().default(60),
DATABASE_CONNECTION_TIMEOUT: z.coerce.number().int().default(20),
DATABASE_WRITER_POOL_TIMEOUT: OptionalIntEnv,
DATABASE_WRITER_CONNECTION_TIMEOUT: OptionalIntEnv,
DATABASE_READ_REPLICA_POOL_TIMEOUT: OptionalIntEnv,
DATABASE_READ_REPLICA_CONNECTION_TIMEOUT: OptionalIntEnv,
// Dashboard-agent conversation store. Cloud points this at a dedicated
// database; when unset it falls back to DATABASE_URL (OSS), where
// the tables live in the isolated `trigger_dashboard_agent` schema.
@@ -185,7 +195,15 @@ const EnvironmentSchema = z
.refine(isValidDatabaseUrl, "RUN_OPS_LEGACY_DATABASE_READ_REPLICA_URL is invalid")
.optional(),
// Optional cap for the unpooled new run-ops read replica. Unset falls back to DATABASE_CONNECTION_LIMIT.
RUN_OPS_DATABASE_READ_REPLICA_CONNECTION_LIMIT: z.coerce.number().int().optional(),
RUN_OPS_DATABASE_READ_REPLICA_CONNECTION_LIMIT: OptionalIntEnv,
RUN_OPS_DATABASE_WRITER_POOL_TIMEOUT: OptionalIntEnv,
RUN_OPS_DATABASE_WRITER_CONNECTION_TIMEOUT: OptionalIntEnv,
RUN_OPS_DATABASE_READ_REPLICA_POOL_TIMEOUT: OptionalIntEnv,
RUN_OPS_DATABASE_READ_REPLICA_CONNECTION_TIMEOUT: OptionalIntEnv,
RUN_OPS_LEGACY_DATABASE_WRITER_POOL_TIMEOUT: OptionalIntEnv,
RUN_OPS_LEGACY_DATABASE_WRITER_CONNECTION_TIMEOUT: OptionalIntEnv,
RUN_OPS_LEGACY_DATABASE_READ_REPLICA_POOL_TIMEOUT: OptionalIntEnv,
RUN_OPS_LEGACY_DATABASE_READ_REPLICA_CONNECTION_TIMEOUT: OptionalIntEnv,
// Direct DSN for applying the full @trigger.dev/database migrations to the LEGACY run-ops DB, keeping
// its schema current after the control plane moves off it. Direct, not pooled — migrations never run
// over a pooler. Optional; unset -> the entrypoint's legacy migrate step is skipped.
docs/self-hosting/env/webapp.mdx
Vendored
+4
View File
@@ -26,7 +26,11 @@ mode: "wide"
| `DATABASE_CONNECTION_LIMIT` | No | 10 | Max DB connections. |
| `DATABASE_POOL_TIMEOUT` | No | 60 | DB pool timeout (s). |
| `DATABASE_CONNECTION_TIMEOUT` | No | 20 | DB connect timeout (s). |
| `DATABASE_WRITER_POOL_TIMEOUT` | No | `DATABASE_POOL_TIMEOUT` | Writer pool timeout (s); overrides the shared default for the writer only. |
| `DATABASE_WRITER_CONNECTION_TIMEOUT` | No | `DATABASE_CONNECTION_TIMEOUT` | Writer connect timeout (s); overrides the shared default for the writer only. |
| `DATABASE_READ_REPLICA_URL` | No | `DATABASE_URL` | Read-replica DB string. |
| `DATABASE_READ_REPLICA_POOL_TIMEOUT` | No | `DATABASE_POOL_TIMEOUT` | Read-replica pool timeout (s); overrides the shared default for the replica only. |
| `DATABASE_READ_REPLICA_CONNECTION_TIMEOUT` | No | `DATABASE_CONNECTION_TIMEOUT` | Read-replica connect timeout (s); overrides the shared default for the replica only. |
| **Redis** | | | |
| `REDIS_HOST` | Yes | — | Redis host. |
| `REDIS_PORT` | Yes | — | Redis port. |