n8n-io/n8n · warning · StaleResumeError

No suspended run found for runId: ${this.runId}

Error message

No suspended run found for runId: ${this.runId}

What it means

Thrown when an HTTP Request node creates a brand-new credential of a 'plain' generic auth type (httpBearerAuth, httpHeaderAuth, httpQueryAuth, httpCustomAuth). Credential setup rejects these for new credentials, so the validator steers you to httpTemplatedCustomAuth, whose template can express 'Authorization: Bearer <token>'. Plain generic types are only valid when referencing an already-existing credential id.

Source

Thrown at packages/@n8n/agents/src/runtime/loop/agent-runtime.ts:360

	async resume(
		method: 'generate',
		data: unknown,
		options: ResumeOptions & ExecutionOptions,
	): Promise<GenerateResult>;
	async resume(
		method: 'stream',
		data: unknown,
		options: ResumeOptions & ExecutionOptions,
	): Promise<StreamResult>;
	async resume(
		method: 'generate' | 'stream',
		data: unknown,
		options: ResumeOptions & ExecutionOptions,
	): Promise<GenerateResult | StreamResult> {
		this.runId = options.runId;
		const state = await this.runState.resume(this.runId);
		if (!state) {
			throw new StaleResumeError(`No suspended run found for runId: ${this.runId}`);
		}

		const toolCall = state.pendingToolCalls[options.toolCallId];
		if (!toolCall) {
			throw new StaleResumeError(`No tool call found for toolCallId: ${options.toolCallId}`);
		}

		const list = AgentMessageList.deserialize(state.messageList);
		this.context.hydrateDeferredToolsFromList(list);
		await hydrateFileParts(list.messages(), this.config.fileStore, {
			threadId: state.persistence?.threadId,
		});

		const tool = this.context
			.getCurrentTools(state.persistence)
			.find((t) => t.name === toolCall.toolName);
		if (!tool) throw new Error(`Tool ${toolCall.toolName} not found`);

View on GitHub (pinned to 5ac6606e81)

Solutions

1. Switch genericAuthType to 'httpTemplatedCustomAuth' and supply a template like {"headers":{"Authorization":"Bearer {{api_key}}"}} (or {"query":{...}} / {"body":{...}} as the provider requires).
2. If you intend to reuse an existing credential, pass its id in credentials[genericAuthType] so it is not treated as new.
3. Remove genericAuthType entirely and use a dedicated credential type if the provider has a first-class one.

Example fix

// before
httpRequest({
  name: 'Call API',
  authentication: 'genericCredentialType',
  genericAuthType: 'httpBearerAuth',
  credentials: { httpBearerAuth: newCredential('My Bearer') },
});

// after
httpRequest({
  name: 'Call API',
  authentication: 'genericCredentialType',
  genericAuthType: 'httpTemplatedCustomAuth',
  credentials: { httpTemplatedCustomAuth: newCredential('My API Auth') },
});

const TEMPLATABLE = new Set(['httpBearerAuth', 'httpHeaderAuth', 'httpQueryAuth', 'httpCustomAuth']);

function isPlainNewGenericAuth(params: Record<string, unknown>, credentials?: Record<string, unknown>): boolean {
  const t = params.genericAuthType;
  return params.authentication === 'genericCredentialType' && typeof t === 'string' && TEMPLATABLE.has(t) &&
    (credentials?.[t] === undefined || isNewCredentialSentinel(credentials?.[t]));
}

function isNewCredentialSentinel(v: unknown): boolean {
  return typeof v === 'object' && v !== null && '__newCredential' in v && !(v as { id?: string }).id;
}

function isNewCredentialSentinel(v: unknown): boolean {
  return typeof v === 'object' && v !== null && '__newCredential' in v && !(v as { id?: string }).id;
}

When it happens

Trigger: params.authentication === 'genericCredentialType' AND params.genericAuthType is one of the four TEMPLATABLE_PLAIN_AUTH_TYPES, AND node.config.credentials[genericAuthType] is either undefined or a new-credential sentinel (an object with __newCredential but no id). Any of those conditions for a brand-new credential trips it.

Common situations: An AI agent sees 'Authorization: Bearer <token>' in provider docs and picks genericAuthType='httpBearerAuth' with newCredential('Bearer Auth'); upgrading a workflow where the old plain type used to be accepted; the model reusing a pattern from a different provider that genuinely supports bearer auth credentials.

Related errors

• LangSmithTelemetry creates its own tracer — do not use .otlp
• Invalid resume payload: ${parseResult.error}
• Failed to retrieve access token
• Failed to retrieve OAuth2 access token
• API Key is missing in the selected Azure OpenAI API credenti

AI-assisted analysis of n8n-io/n8n@5ac6606e81 (2026-08-12). Data as JSON: /api/errors/e4e1e3529353dff2. Report an issue: GitHub.