fix(workflow): set durations to seconds

packages/workflow-executor/README.md

@@ -24,8 +24,8 @@ npm install -g @forestadmin/workflow-executor
 | `DATABASE_URL` | ✓* | — | Postgres connection string (*not needed with `--in-memory`) |
 | `HTTP_PORT` | — | `3400` | Port for the executor HTTP server |
 | `FOREST_SERVER_URL` | — | `https://api.forestadmin.com` | Orchestrator URL |
-| `POLLING_INTERVAL_MS` | — | `5000` | Poll cadence for pending steps |
-| `STOP_TIMEOUT_MS` | — | `30000` | Graceful shutdown deadline |
+| `POLLING_INTERVAL_S` | — | `30` | Poll cadence for pending steps |
+| `STOP_TIMEOUT_S` | — | `30` | Graceful shutdown deadline |
 
 Optional AI configuration (all-or-nothing — falls back to server AI if any is missing):
 
@@ -44,7 +44,7 @@ forest-workflow-executor
 You should see (pretty format when stdout is a TTY):
 
 ```
-13:33:42 info  Workflow executor starting mode="database" forestServerUrl="https://api.forestadmin.com" agentUrl="http://localhost:3351" httpPort=3400 pollingIntervalMs=5000 aiConfig="server fallback"
+13:33:42 info  Workflow executor starting mode="database" forestServerUrl="https://api.forestadmin.com" agentUrl="http://localhost:3351" httpPort=3400 pollingIntervalS=5 aiConfig="server fallback"
 13:33:42 info  Workflow executor ready url="http://localhost:3400"
 13:33:47 info  Poll cycle completed fetched=0 dispatching=0
 ```
@@ -77,7 +77,7 @@ curl http://localhost:3400/health
 ### Graceful shutdown
 
 Send `SIGTERM` or `SIGINT`. The executor drains in-flight steps, closes the HTTP
-server, and exits with code `0`. Steps that don't drain within `STOP_TIMEOUT_MS`
+server, and exits with code `0`. Steps that don't drain within `STOP_TIMEOUT_S`
 are force-killed and the process exits with code `1`.
 
 ### Exit codes

packages/workflow-executor/example/.env.example

@@ -11,7 +11,7 @@ DATABASE_URL=postgres://executor:password@localhost:5459/workflow_executor
 # Optional — defaults shown
 HTTP_PORT=3400
 FOREST_SERVER_URL=https://api.development.forestadmin.com
-POLLING_INTERVAL_MS=5000
+POLLING_INTERVAL_S=5
 
 # Optional local AI (all-or-nothing — falls back to server AI if any is missing)
 # AI_PROVIDER=anthropic

packages/workflow-executor/example/.env.executors.example

@@ -24,4 +24,4 @@ NODE_TLS_REJECT_UNAUTHORIZED=0
 
 
 # Optional — default shown.
-POLLING_INTERVAL_MS=5000
+POLLING_INTERVAL_S=5

packages/workflow-executor/example/README.md

@@ -52,7 +52,7 @@ Expected output:
   Forest server    : https://api.forestadmin.com
   Agent URL        : http://localhost:3351
   HTTP port        : 3400
-  Polling interval : 5000ms
+  Polling interval : 5s
   AI config        : server fallback (no local AI)
 [forest-workflow-executor] Ready on http://localhost:3400
 {"message":"Poll cycle completed","timestamp":"...","fetched":0,"dispatching":0}
@@ -67,7 +67,7 @@ curl http://localhost:3400/health
 
 The executor will:
 - Auto-create the `workflow_step_executions` table via Umzug migrations
-- Poll the Forest Admin orchestrator every `POLLING_INTERVAL_MS` (5s default)
+- Poll the Forest Admin orchestrator every `POLLING_INTERVAL_S` (5s default)
 - Execute steps locally and report results back
 
 ## Available scripts

packages/workflow-executor/example/docker-compose.executors.yml

@@ -19,7 +19,7 @@ x-executor-env: &executor-env
   AGENT_URL: ${EXECUTOR_AGENT_URL:-http://host.docker.internal:3351}
   FOREST_SERVER_URL: ${FOREST_SERVER_URL:-https://api.development.forestadmin.com}
   DATABASE_URL: ${EXECUTOR_DATABASE_URL}
-  POLLING_INTERVAL_MS: ${POLLING_INTERVAL_MS:-5000}
+  POLLING_INTERVAL_S: ${POLLING_INTERVAL_S:-5}
   HTTP_PORT: "3400"
   NODE_TLS_REJECT_UNAUTHORIZED: 0
 

packages/workflow-executor/src/build-workflow-executor.ts

@@ -14,19 +14,19 @@ import ForestServerWorkflowPort from './adapters/forest-server-workflow-port';
 import ForestadminClientActivityLogPortFactory from './adapters/forestadmin-client-activity-log-port-factory';
 import ServerAiAdapter from './adapters/server-ai-adapter';
 import {
-  DEFAULT_AI_INVOKE_TIMEOUT_MS,
+  DEFAULT_AI_INVOKE_TIMEOUT_S,
   DEFAULT_FOREST_SERVER_URL,
-  DEFAULT_POLLING_INTERVAL_MS,
-  DEFAULT_SCHEMA_CACHE_TTL_MS,
-  DEFAULT_STEP_TIMEOUT_MS,
+  DEFAULT_POLLING_INTERVAL_S,
+  DEFAULT_SCHEMA_CACHE_TTL_S,
+  DEFAULT_STEP_TIMEOUT_S,
 } from './defaults';
 import ExecutorHttpServer from './http/executor-http-server';
 import Runner from './runner';
 import SchemaCache from './schema-cache';
 import DatabaseStore from './stores/database-store';
 import InMemoryStore from './stores/in-memory-store';
 
-const FORCE_EXIT_DELAY_MS = 5000;
+const FORCE_EXIT_DELAY_S = 5;
 
 export interface WorkflowExecutor {
   start(): Promise<void>;
@@ -41,15 +41,15 @@ export interface ExecutorOptions {
   httpPort: number;
   forestServerUrl?: string;
   aiConfigurations?: AiConfiguration[];
-  pollingIntervalMs?: number;
+  pollingIntervalS?: number;
   logger?: Logger;
-  stopTimeoutMs?: number;
-  stepTimeoutMs?: number;
-  aiInvokeTimeoutMs?: number;
+  stopTimeoutS?: number;
+  stepTimeoutS?: number;
+  aiInvokeTimeoutS?: number;
   // Max auto-chained steps per entry (see RunnerConfig.maxChainDepth). 0 disables chaining.
   maxChainDepth?: number;
-  // Collection schema cache TTL in ms. Lower it to pick up orchestrator schema changes sooner.
-  schemaCacheTtlMs?: number;
+  // Collection schema cache TTL in seconds. Lower it to pick up orchestrator schema changes sooner.
+  schemaCacheTtlS?: number;
   // Dev only: makes every AI call fail immediately so error paths can be exercised locally.
   forceAiError?: boolean;
 }
@@ -96,7 +96,7 @@ function buildCommonDependencies(options: ExecutorOptions) {
 
   // A TTL of 0/negative/non-finite would silently make the cache always-stale, so fall back.
   const schemaCache = new SchemaCache(
-    positiveOrDefault(options.schemaCacheTtlMs, DEFAULT_SCHEMA_CACHE_TTL_MS),
+    positiveOrDefault(options.schemaCacheTtlS, DEFAULT_SCHEMA_CACHE_TTL_S),
   );
 
   const agentPort = new AgentClientAgentPort({
@@ -121,12 +121,12 @@ function buildCommonDependencies(options: ExecutorOptions) {
     aiModelPort,
     activityLogPortFactory,
     logger,
-    pollingIntervalMs: options.pollingIntervalMs ?? DEFAULT_POLLING_INTERVAL_MS,
+    pollingIntervalS: options.pollingIntervalS ?? DEFAULT_POLLING_INTERVAL_S,
     envSecret: options.envSecret,
     authSecret: options.authSecret,
-    stopTimeoutMs: options.stopTimeoutMs,
-    stepTimeoutMs: positiveOrDefault(options.stepTimeoutMs, DEFAULT_STEP_TIMEOUT_MS),
-    aiInvokeTimeoutMs: positiveOrDefault(options.aiInvokeTimeoutMs, DEFAULT_AI_INVOKE_TIMEOUT_MS),
+    stopTimeoutS: options.stopTimeoutS,
+    stepTimeoutS: positiveOrDefault(options.stepTimeoutS, DEFAULT_STEP_TIMEOUT_S),
+    aiInvokeTimeoutS: positiveOrDefault(options.aiInvokeTimeoutS, DEFAULT_AI_INVOKE_TIMEOUT_S),
     maxChainDepth: options.maxChainDepth,
   };
 }
@@ -169,7 +169,7 @@ function createWorkflowExecutor(
     setTimeout(() => {
       logger.error('Process did not exit after shutdown — forcing exit', {});
       process.exit(process.exitCode ?? 1);
-    }, FORCE_EXIT_DELAY_MS).unref();
+    }, FORCE_EXIT_DELAY_S * 1000).unref();
   };
 
   return {

packages/workflow-executor/src/cli-core.ts

@@ -13,14 +13,14 @@ import {
   type WorkflowExecutor,
 } from './build-workflow-executor';
 import {
-  DEFAULT_AI_INVOKE_TIMEOUT_MS,
+  DEFAULT_AI_INVOKE_TIMEOUT_S,
   DEFAULT_FOREST_SERVER_URL,
   DEFAULT_HTTP_PORT,
   DEFAULT_MAX_CHAIN_DEPTH,
-  DEFAULT_POLLING_INTERVAL_MS,
-  DEFAULT_SCHEMA_CACHE_TTL_MS,
-  DEFAULT_STEP_TIMEOUT_MS,
-  DEFAULT_STOP_TIMEOUT_MS,
+  DEFAULT_POLLING_INTERVAL_S,
+  DEFAULT_SCHEMA_CACHE_TTL_S,
+  DEFAULT_STEP_TIMEOUT_S,
+  DEFAULT_STOP_TIMEOUT_S,
 } from './defaults';
 import { ConfigurationError, extractErrorMessage } from './errors';
 
@@ -157,12 +157,12 @@ export function readEnvConfig(env: NodeJS.ProcessEnv, args: CliArgs): CliConfig
     agentUrl: env.AGENT_URL as string,
     httpPort: parsePositiveIntEnv('HTTP_PORT', env.HTTP_PORT) ?? DEFAULT_HTTP_PORT,
     forestServerUrl: env.FOREST_SERVER_URL,
-    pollingIntervalMs: parsePositiveIntEnv('POLLING_INTERVAL_MS', env.POLLING_INTERVAL_MS),
-    stopTimeoutMs: parsePositiveIntEnv('STOP_TIMEOUT_MS', env.STOP_TIMEOUT_MS),
-    stepTimeoutMs: parsePositiveIntEnv('STEP_TIMEOUT_MS', env.STEP_TIMEOUT_MS),
-    aiInvokeTimeoutMs: parsePositiveIntEnv('AI_INVOKE_TIMEOUT_MS', env.AI_INVOKE_TIMEOUT_MS),
+    pollingIntervalS: parsePositiveIntEnv('POLLING_INTERVAL_S', env.POLLING_INTERVAL_S),
+    stopTimeoutS: parsePositiveIntEnv('STOP_TIMEOUT_S', env.STOP_TIMEOUT_S),
+    stepTimeoutS: parsePositiveIntEnv('STEP_TIMEOUT_S', env.STEP_TIMEOUT_S),
+    aiInvokeTimeoutS: parsePositiveIntEnv('AI_INVOKE_TIMEOUT_S', env.AI_INVOKE_TIMEOUT_S),
     maxChainDepth: parsePositiveIntEnv('MAX_CHAIN_DEPTH', env.MAX_CHAIN_DEPTH),
-    schemaCacheTtlMs: parsePositiveIntEnv('SCHEMA_CACHE_TTL_MS', env.SCHEMA_CACHE_TTL_MS),
+    schemaCacheTtlS: parsePositiveIntEnv('SCHEMA_CACHE_TTL_S', env.SCHEMA_CACHE_TTL_S),
     ...(aiConfigurations && { aiConfigurations }),
     ...(env.FORCE_AI_ERROR === 'true' && { forceAiError: true }),
   };
@@ -195,12 +195,12 @@ Required environment variables:
 Optional environment variables:
   HTTP_PORT              Default: ${DEFAULT_HTTP_PORT}
   FOREST_SERVER_URL      Default: ${DEFAULT_FOREST_SERVER_URL}
-  POLLING_INTERVAL_MS    Default: ${DEFAULT_POLLING_INTERVAL_MS}
-  STOP_TIMEOUT_MS        Default: ${DEFAULT_STOP_TIMEOUT_MS}
-  STEP_TIMEOUT_MS        Max duration of a step in ms (default: ${DEFAULT_STEP_TIMEOUT_MS})
-  AI_INVOKE_TIMEOUT_MS   Max duration of a single AI provider invocation in ms (default: ${DEFAULT_AI_INVOKE_TIMEOUT_MS})
+  POLLING_INTERVAL_S    Default: ${DEFAULT_POLLING_INTERVAL_S}
+  STOP_TIMEOUT_S        Default: ${DEFAULT_STOP_TIMEOUT_S}
+  STEP_TIMEOUT_S        Max duration of a step in seconds (default: ${DEFAULT_STEP_TIMEOUT_S})
+  AI_INVOKE_TIMEOUT_S   Max duration of a single AI provider invocation in seconds (default: ${DEFAULT_AI_INVOKE_TIMEOUT_S})
   MAX_CHAIN_DEPTH        Max steps auto-executed per run before yielding (default: ${DEFAULT_MAX_CHAIN_DEPTH})
-  SCHEMA_CACHE_TTL_MS    Collection schema cache TTL in ms (default: ${DEFAULT_SCHEMA_CACHE_TTL_MS})
+  SCHEMA_CACHE_TTL_S    Collection schema cache TTL in seconds (default: ${DEFAULT_SCHEMA_CACHE_TTL_S})
   NO_COLOR               Set to any value to disable ANSI colors in pretty logs
   FORCE_AI_ERROR         Set to "true" to make every AI call fail (dev only, to test error paths)
 
@@ -234,7 +234,7 @@ export function logStartup(logger: Logger, config: CliConfig): void {
     forestServerUrl: opts.forestServerUrl ?? DEFAULT_FOREST_SERVER_URL,
     agentUrl: opts.agentUrl,
     httpPort: opts.httpPort,
-    pollingIntervalMs: opts.pollingIntervalMs ?? DEFAULT_POLLING_INTERVAL_MS,
+    pollingIntervalS: opts.pollingIntervalS ?? DEFAULT_POLLING_INTERVAL_S,
     aiConfig: aiLabel,
   });
 }

packages/workflow-executor/src/defaults.ts

@@ -1,8 +1,8 @@
 export const DEFAULT_HTTP_PORT = 3400;
 export const DEFAULT_FOREST_SERVER_URL = 'https://api.forestadmin.com';
-export const DEFAULT_POLLING_INTERVAL_MS = 30_000;
-export const DEFAULT_STEP_TIMEOUT_MS = 5 * 60_000;
-export const DEFAULT_AI_INVOKE_TIMEOUT_MS = 30_000;
-export const DEFAULT_STOP_TIMEOUT_MS = 30_000;
+export const DEFAULT_POLLING_INTERVAL_S = 30;
+export const DEFAULT_STEP_TIMEOUT_S = 5 * 60;
+export const DEFAULT_AI_INVOKE_TIMEOUT_S = 30;
+export const DEFAULT_STOP_TIMEOUT_S = 30;
 export const DEFAULT_MAX_CHAIN_DEPTH = 50;
-export const DEFAULT_SCHEMA_CACHE_TTL_MS = 10 * 60_000;
+export const DEFAULT_SCHEMA_CACHE_TTL_S = 10 * 60;

packages/workflow-executor/src/errors.ts

@@ -212,9 +212,9 @@ export class ActivityLogCreationError extends WorkflowExecutorError {
 }
 
 export class StepTimeoutError extends WorkflowExecutorError {
-  constructor(timeoutMs: number) {
+  constructor(timeoutS: number) {
     super(
-      `Step execution exceeded timeout of ${timeoutMs}ms`,
+      `Step execution exceeded timeout of ${timeoutS}s`,
       'The step took too long to complete. Please try again, or contact your administrator if the problem persists.',
     );
   }
@@ -224,9 +224,9 @@ export class StepTimeoutError extends WorkflowExecutorError {
 // StepTimeoutError so we can surface a provider-specific message and tune the AI timeout
 // independently of the step timeout (AI hangs are common; record fetches are not).
 export class AiInvokeTimeoutError extends WorkflowExecutorError {
-  constructor(timeoutMs: number) {
+  constructor(timeoutS: number) {
     super(
-      `AI provider did not respond within ${timeoutMs}ms`,
+      `AI provider did not respond within ${timeoutS}s`,
       'The AI provider did not respond in time. Please try again, or contact your administrator if the problem persists.',
     );
   }

packages/workflow-executor/src/executors/base-step-executor.ts

@@ -71,7 +71,7 @@ export default abstract class BaseStepExecutor<TStep extends StepDefinition = St
       if (error instanceof StepTimeoutError) {
         this.context.logger.error(error.message, {
           ...this.logCtx,
-          timeoutMs: this.context.stepTimeoutMs,
+          timeoutS: this.context.stepTimeoutS,
         });
 
         return this.buildOutcomeResult({ status: 'error', error: error.userMessage });
@@ -112,8 +112,8 @@ export default abstract class BaseStepExecutor<TStep extends StepDefinition = St
   // on execPromise must be attached BEFORE the race so a late rejection doesn't trigger
   // UnhandledPromiseRejection. Late resolutions are silently discarded.
   private async runWithTimeout(): Promise<StepExecutionResult> {
-    const timeoutMs = this.context.stepTimeoutMs;
-    if (!timeoutMs || timeoutMs <= 0) return this.doExecute();
+    const timeoutS = this.context.stepTimeoutS;
+    if (!timeoutS || timeoutS <= 0) return this.doExecute();
 
     let timer: NodeJS.Timeout | undefined;
     let hasTimeoutFired = false;
@@ -133,8 +133,8 @@ export default abstract class BaseStepExecutor<TStep extends StepDefinition = St
         new Promise<never>((_, reject) => {
           timer = setTimeout(() => {
             hasTimeoutFired = true;
-            reject(new StepTimeoutError(timeoutMs));
-          }, timeoutMs);
+            reject(new StepTimeoutError(timeoutS));
+          }, timeoutS * 1000);
         }),
       ]);
     } finally {
@@ -314,9 +314,9 @@ export default abstract class BaseStepExecutor<TStep extends StepDefinition = St
     BaseStepExecutor.assertNoMidArraySystemMessages(messages);
     const modelWithTools = this.context.model.bindTools(tools, { tool_choice: 'any' });
     const preparedMessages = BaseStepExecutor.mergeLeadingSystemMessages(messages);
-    const aiTimeoutMs = this.context.aiInvokeTimeoutMs;
-    const timeoutMs = aiTimeoutMs && aiTimeoutMs > 0 ? aiTimeoutMs : undefined;
-    const signal = timeoutMs ? AbortSignal.timeout(timeoutMs) : undefined;
+    const aiTimeoutS = this.context.aiInvokeTimeoutS;
+    const timeoutS = aiTimeoutS && aiTimeoutS > 0 ? aiTimeoutS : undefined;
+    const signal = timeoutS ? AbortSignal.timeout(timeoutS * 1000) : undefined;
 
     let response;
 
@@ -326,7 +326,7 @@ export default abstract class BaseStepExecutor<TStep extends StepDefinition = St
     } catch (err) {
       // Detect the timeout via our own signal, not the thrown error's name: providers wrap an
       // aborted request differently (AbortError, TimeoutError, APIUserAbortError, …).
-      if (timeoutMs !== undefined && signal?.aborted) throw new AiInvokeTimeoutError(timeoutMs);
+      if (timeoutS !== undefined && signal?.aborted) throw new AiInvokeTimeoutError(timeoutS);
       throw err;
     }
 

packages/workflow-executor/src/executors/step-executor-factory.ts

@@ -48,8 +48,8 @@ export interface StepContextConfig {
   runStore: RunStore;
   schemaCache: SchemaCache;
   logger: Logger;
-  stepTimeoutMs?: number;
-  aiInvokeTimeoutMs?: number;
+  stepTimeoutS?: number;
+  aiInvokeTimeoutS?: number;
 }
 
 export default class StepExecutorFactory {
@@ -159,8 +159,8 @@ export default class StepExecutorFactory {
       schemaResolver,
       logger: cfg.logger,
       incomingPendingData,
-      stepTimeoutMs: cfg.stepTimeoutMs,
-      aiInvokeTimeoutMs: cfg.aiInvokeTimeoutMs,
+      stepTimeoutS: cfg.stepTimeoutS,
+      aiInvokeTimeoutS: cfg.aiInvokeTimeoutS,
     };
   }
 }