feat(core): Add error reporting for workflow evaluation (no-changelog) (#12838)

Co-authored-by: oleg <me@olegivaniv.com>

cypress/e2e/233-AI-switch-to-logs-on-error.cy.ts

@@ -1,4 +1,4 @@
-import type { ExecutionError } from 'n8n-workflow/src';
+import type { ExecutionError } from 'n8n-workflow';
 
 import {
 	closeManualChatModal,

packages/@n8n/nodes-langchain/nodes/chains/InformationExtractor/test/InformationExtraction.node.test.ts

@@ -1,7 +1,7 @@
 import type { BaseLanguageModel } from '@langchain/core/language_models/base';
 import { FakeLLM, FakeListChatModel } from '@langchain/core/utils/testing';
 import get from 'lodash/get';
-import type { IDataObject, IExecuteFunctions } from 'n8n-workflow/src';
+import type { IDataObject, IExecuteFunctions } from 'n8n-workflow';
 
 import { makeZodSchemaFromAttributes } from '../helpers';
 import { InformationExtractor } from '../InformationExtractor.node';

packages/cli/src/commands/base-command.ts

@@ -19,6 +19,7 @@ import * as CrashJournal from '@/crash-journal';
 import * as Db from '@/db';
 import { getDataDeduplicationService } from '@/deduplication';
 import { DeprecationService } from '@/deprecation/deprecation.service';
+import { TestRunnerService } from '@/evaluation.ee/test-runner/test-runner.service.ee';
 import { MessageEventBus } from '@/eventbus/message-event-bus/message-event-bus';
 import { TelemetryEventRelay } from '@/events/relays/telemetry.event-relay';
 import { initExpressionEvaluator } from '@/expression-evaluator';
@@ -259,6 +260,10 @@ export abstract class BaseCommand extends Command {
 		Container.get(WorkflowHistoryManager).init();
 	}
 
+	async cleanupTestRunner() {
+		await Container.get(TestRunnerService).cleanupIncompleteRuns();
+	}
+
 	async finally(error: Error | undefined) {
 		if (inTest || this.id === 'start') return;
 		if (Db.connectionState.connected) {

packages/cli/src/commands/start.ts

@@ -222,6 +222,11 @@ export class Start extends BaseCommand {
 		this.initWorkflowHistory();
 		this.logger.debug('Workflow history init complete');
 
+		if (!isMultiMainEnabled) {
+			await this.cleanupTestRunner();
+			this.logger.debug('Test runner cleanup complete');
+		}
+
 		if (!this.globalConfig.endpoints.disableUi) {
 			await this.generateStaticAssets();
 		}

packages/cli/src/databases/entities/test-case-execution.ee.ts

@@ -1,4 +1,5 @@
 import { Column, Entity, ManyToOne, OneToOne } from '@n8n/typeorm';
+import type { IDataObject } from 'n8n-workflow';
 
 import {
 	datetimeColumnType,
@@ -7,9 +8,19 @@ import {
 } from '@/databases/entities/abstract-entity';
 import type { ExecutionEntity } from '@/databases/entities/execution-entity';
 import { TestRun } from '@/databases/entities/test-run.ee';
+import type { TestCaseExecutionErrorCode } from '@/evaluation.ee/test-runner/errors.ee';
 
 export type TestCaseRunMetrics = Record<string, number | boolean>;
 
+export type TestCaseExecutionStatus =
+	| 'new' // Test case execution was created and added to the test run, but has not been started yet
+	| 'running' // Workflow under test is running
+	| 'evaluation_running' // Evaluation workflow is running
+	| 'success' // Both workflows have completed successfully
+	| 'error' // An error occurred during the execution of workflow under test or evaluation workflow
+	| 'warning' // There were warnings during the execution of workflow under test or evaluation workflow. Used only to signal possible issues to user, not to indicate a failure.
+	| 'cancelled';
+
 /**
  * This entity represents the linking between the test runs and individual executions.
  * It stores status, links to past, new and evaluation executions, and metrics produced by individual evaluation wf executions
@@ -49,7 +60,7 @@ export class TestCaseExecution extends WithStringId {
 	evaluationExecutionId: string | null;
 
 	@Column()
-	status: 'new' | 'running' | 'evaluation_running' | 'success' | 'error' | 'cancelled';
+	status: TestCaseExecutionStatus;
 
 	@Column({ type: datetimeColumnType, nullable: true })
 	runAt: Date | null;
@@ -58,10 +69,10 @@ export class TestCaseExecution extends WithStringId {
 	completedAt: Date | null;
 
 	@Column('varchar', { nullable: true })
-	errorCode: string | null;
+	errorCode: TestCaseExecutionErrorCode | null;
 
 	@Column(jsonColumnType, { nullable: true })
-	errorDetails: Record<string, unknown>;
+	errorDetails: IDataObject | null;
 
 	@Column(jsonColumnType, { nullable: true })
 	metrics: TestCaseRunMetrics;

packages/cli/src/databases/entities/test-run.ee.ts

@@ -1,13 +1,17 @@
-import { Column, Entity, Index, ManyToOne, RelationId } from '@n8n/typeorm';
+import { Column, Entity, Index, ManyToOne, OneToMany, RelationId } from '@n8n/typeorm';
+import type { IDataObject } from 'n8n-workflow';
 
 import {
 	datetimeColumnType,
 	jsonColumnType,
 	WithTimestampsAndStringId,
 } from '@/databases/entities/abstract-entity';
+import type { TestCaseExecution } from '@/databases/entities/test-case-execution.ee';
 import { TestDefinition } from '@/databases/entities/test-definition.ee';
+import type { TestRunFinalResult } from '@/databases/repositories/test-run.repository.ee';
+import type { TestRunErrorCode } from '@/evaluation.ee/test-runner/errors.ee';
 
-type TestRunStatus = 'new' | 'running' | 'completed' | 'error' | 'cancelled';
+export type TestRunStatus = 'new' | 'running' | 'completed' | 'error' | 'cancelled';
 
 export type AggregatedTestRunMetrics = Record<string, number | boolean>;
 
@@ -54,4 +58,26 @@ export class TestRun extends WithTimestampsAndStringId {
 	 */
 	@Column('integer', { nullable: true })
 	failedCases: number;
+
+	/**
+	 * This will contain the error code if the test run failed.
+	 * This is used for test run level errors, not for individual test case errors.
+	 */
+	@Column('varchar', { nullable: true, length: 255 })
+	errorCode: TestRunErrorCode | null;
+
+	/**
+	 * Optional details about the error that happened during the test run
+	 */
+	@Column(jsonColumnType, { nullable: true })
+	errorDetails: IDataObject | null;
+
+	@OneToMany('TestCaseExecution', 'testRun')
+	testCaseExecutions: TestCaseExecution[];
+
+	/**
+	 * Calculated property to determine the final result of the test run
+	 * depending on the statuses of test case executions
+	 */
+	finalResult?: TestRunFinalResult | null;
 }

packages/cli/src/databases/migrations/common/1737715421462-AddErrorColumnsToTestRuns.ts

@@ -0,0 +1,24 @@
+import type { MigrationContext, ReversibleMigration } from '@/databases/types';
+
+// We have to use raw query migration instead of schemaBuilder helpers,
+// because the typeorm schema builder implements addColumns by a table recreate for sqlite
+// which causes weird issues with the migration
+export class AddErrorColumnsToTestRuns1737715421462 implements ReversibleMigration {
+	async up({ escape, runQuery }: MigrationContext) {
+		const tableName = escape.tableName('test_run');
+		const errorCodeColumnName = escape.columnName('errorCode');
+		const errorDetailsColumnName = escape.columnName('errorDetails');
+
+		await runQuery(`ALTER TABLE ${tableName} ADD COLUMN ${errorCodeColumnName} VARCHAR(255);`);
+		await runQuery(`ALTER TABLE ${tableName} ADD COLUMN ${errorDetailsColumnName} TEXT;`);
+	}
+
+	async down({ escape, runQuery }: MigrationContext) {
+		const tableName = escape.tableName('test_run');
+		const errorCodeColumnName = escape.columnName('errorCode');
+		const errorDetailsColumnName = escape.columnName('errorDetails');
+
+		await runQuery(`ALTER TABLE ${tableName} DROP COLUMN ${errorCodeColumnName};`);
+		await runQuery(`ALTER TABLE ${tableName} DROP COLUMN ${errorDetailsColumnName};`);
+	}
+}

packages/cli/src/databases/migrations/mysqldb/index.ts

@@ -78,6 +78,7 @@ import { AddMockedNodesColumnToTestDefinition1733133775640 } from '../common/173
 import { AddManagedColumnToCredentialsTable1734479635324 } from '../common/1734479635324-AddManagedColumnToCredentialsTable';
 import { AddStatsColumnsToTestRun1736172058779 } from '../common/1736172058779-AddStatsColumnsToTestRun';
 import { CreateTestCaseExecutionTable1736947513045 } from '../common/1736947513045-CreateTestCaseExecutionTable';
+import { AddErrorColumnsToTestRuns1737715421462 } from '../common/1737715421462-AddErrorColumnsToTestRuns';
 
 export const mysqlMigrations: Migration[] = [
 	InitialMigration1588157391238,
@@ -158,4 +159,5 @@ export const mysqlMigrations: Migration[] = [
 	AddProjectIcons1729607673469,
 	AddStatsColumnsToTestRun1736172058779,
 	CreateTestCaseExecutionTable1736947513045,
+	AddErrorColumnsToTestRuns1737715421462,
 ];

packages/cli/src/databases/migrations/postgresdb/index.ts

@@ -78,6 +78,7 @@ import { AddMockedNodesColumnToTestDefinition1733133775640 } from '../common/173
 import { AddManagedColumnToCredentialsTable1734479635324 } from '../common/1734479635324-AddManagedColumnToCredentialsTable';
 import { AddStatsColumnsToTestRun1736172058779 } from '../common/1736172058779-AddStatsColumnsToTestRun';
 import { CreateTestCaseExecutionTable1736947513045 } from '../common/1736947513045-CreateTestCaseExecutionTable';
+import { AddErrorColumnsToTestRuns1737715421462 } from '../common/1737715421462-AddErrorColumnsToTestRuns';
 
 export const postgresMigrations: Migration[] = [
 	InitialMigration1587669153312,
@@ -158,4 +159,5 @@ export const postgresMigrations: Migration[] = [
 	AddProjectIcons1729607673469,
 	AddStatsColumnsToTestRun1736172058779,
 	CreateTestCaseExecutionTable1736947513045,
+	AddErrorColumnsToTestRuns1737715421462,
 ];

packages/cli/src/databases/migrations/sqlite/index.ts

@@ -75,6 +75,7 @@ import { AddMockedNodesColumnToTestDefinition1733133775640 } from '../common/173
 import { AddManagedColumnToCredentialsTable1734479635324 } from '../common/1734479635324-AddManagedColumnToCredentialsTable';
 import { AddStatsColumnsToTestRun1736172058779 } from '../common/1736172058779-AddStatsColumnsToTestRun';
 import { CreateTestCaseExecutionTable1736947513045 } from '../common/1736947513045-CreateTestCaseExecutionTable';
+import { AddErrorColumnsToTestRuns1737715421462 } from '../common/1737715421462-AddErrorColumnsToTestRuns';
 
 const sqliteMigrations: Migration[] = [
 	InitialMigration1588102412422,
@@ -152,6 +153,7 @@ const sqliteMigrations: Migration[] = [
 	AddProjectIcons1729607673469,
 	AddStatsColumnsToTestRun1736172058779,
 	CreateTestCaseExecutionTable1736947513045,
+	AddErrorColumnsToTestRuns1737715421462,
 ];
 
 export { sqliteMigrations };

packages/cli/src/databases/repositories/test-case-execution.repository.ee.ts

@@ -2,8 +2,35 @@ import { Service } from '@n8n/di';
 import type { EntityManager } from '@n8n/typeorm';
 import { DataSource, In, Not, Repository } from '@n8n/typeorm';
 import type { DeepPartial } from '@n8n/typeorm/common/DeepPartial';
+import type { IDataObject } from 'n8n-workflow';
 
 import { TestCaseExecution } from '@/databases/entities/test-case-execution.ee';
+import type { TestCaseExecutionErrorCode } from '@/evaluation.ee/test-runner/errors.ee';
+
+type StatusUpdateOptions = {
+	testRunId: string;
+	pastExecutionId: string;
+	trx?: EntityManager;
+};
+
+type MarkAsFailedOptions = StatusUpdateOptions & {
+	errorCode?: TestCaseExecutionErrorCode;
+	errorDetails?: IDataObject;
+};
+
+type MarkAsWarningOptions = MarkAsFailedOptions;
+
+type MarkAsRunningOptions = StatusUpdateOptions & {
+	executionId: string;
+};
+
+type MarkAsEvaluationRunningOptions = StatusUpdateOptions & {
+	evaluationExecutionId: string;
+};
+
+type MarkAsCompletedOptions = StatusUpdateOptions & {
+	metrics: Record<string, number>;
+};
 
 @Service()
 export class TestCaseExecutionRepository extends Repository<TestCaseExecution> {
@@ -27,8 +54,11 @@ export class TestCaseExecutionRepository extends Repository<TestCaseExecution> {
 		return await this.save(mappings);
 	}
 
-	async markAsRunning(testRunId: string, pastExecutionId: string, executionId: string) {
+	async markAsRunning({ testRunId, pastExecutionId, executionId, trx }: MarkAsRunningOptions) {
-		return await this.update(
+		trx = trx ?? this.manager;
+
+		return await trx.update(
+			TestCaseExecution,
 			{ testRun: { id: testRunId }, pastExecutionId },
 			{
 				status: 'running',
@@ -38,12 +68,16 @@ export class TestCaseExecutionRepository extends Repository<TestCaseExecution> {
 		);
 	}
 
-	async markAsEvaluationRunning(
+	async markAsEvaluationRunning({
-		testRunId: string,
+		testRunId,
-		pastExecutionId: string,
+		pastExecutionId,
-		evaluationExecutionId: string,
+		evaluationExecutionId,
-	) {
+		trx,
-		return await this.update(
+	}: MarkAsEvaluationRunningOptions) {
+		trx = trx ?? this.manager;
+
+		return await trx.update(
+			TestCaseExecution,
 			{ testRun: { id: testRunId }, pastExecutionId },
 			{
 				status: 'evaluation_running',
@@ -52,12 +86,7 @@ export class TestCaseExecutionRepository extends Repository<TestCaseExecution> {
 		);
 	}
 
-	async markAsCompleted(
+	async markAsCompleted({ testRunId, pastExecutionId, metrics, trx }: MarkAsCompletedOptions) {
-		testRunId: string,
-		pastExecutionId: string,
-		metrics: Record<string, number>,
-		trx?: EntityManager,
-	) {
 		trx = trx ?? this.manager;
 
 		return await trx.update(
@@ -84,7 +113,13 @@ export class TestCaseExecutionRepository extends Repository<TestCaseExecution> {
 		);
 	}
 
-	async markAsFailed(testRunId: string, pastExecutionId: string, trx?: EntityManager) {
+	async markAsFailed({
+		testRunId,
+		pastExecutionId,
+		errorCode,
+		errorDetails,
+		trx,
+	}: MarkAsFailedOptions) {
 		trx = trx ?? this.manager;
 
 		return await trx.update(
@@ -93,6 +128,25 @@ export class TestCaseExecutionRepository extends Repository<TestCaseExecution> {
 			{
 				status: 'error',
 				completedAt: new Date(),
+				errorCode,
+				errorDetails,
+			},
+		);
+	}
+
+	async markAsWarning({
+		testRunId,
+		pastExecutionId,
+		errorCode,
+		errorDetails,
+	}: MarkAsWarningOptions) {
+		return await this.update(
+			{ testRun: { id: testRunId }, pastExecutionId },
+			{
+				status: 'warning',
+				completedAt: new Date(),
+				errorCode,
+				errorDetails,
 			},
 		);
 	}

packages/cli/src/databases/repositories/test-run.repository.ee.ts

@@ -1,11 +1,21 @@
 import { Service } from '@n8n/di';
 import type { EntityManager, FindManyOptions } from '@n8n/typeorm';
-import { DataSource, Repository } from '@n8n/typeorm';
+import { DataSource, In, Repository } from '@n8n/typeorm';
+import type { IDataObject } from 'n8n-workflow';
 
 import type { AggregatedTestRunMetrics } from '@/databases/entities/test-run.ee';
 import { TestRun } from '@/databases/entities/test-run.ee';
+import { NotFoundError } from '@/errors/response-errors/not-found.error';
+import type { TestRunErrorCode } from '@/evaluation.ee/test-runner/errors.ee';
+import { getTestRunFinalResult } from '@/evaluation.ee/test-runner/utils.ee';
 import type { ListQuery } from '@/requests';
 
+export type TestRunFinalResult = 'success' | 'error' | 'warning';
+
+export type TestRunSummary = TestRun & {
+	finalResult: TestRunFinalResult | null;
+};
+
 @Service()
 export class TestRunRepository extends Repository<TestRun> {
 	constructor(dataSource: DataSource) {
@@ -40,6 +50,21 @@ export class TestRunRepository extends Repository<TestRun> {
 		return await trx.update(TestRun, id, { status: 'cancelled' });
 	}
 
+	async markAsError(id: string, errorCode: TestRunErrorCode, errorDetails?: IDataObject) {
+		return await this.update(id, {
+			status: 'error',
+			errorCode,
+			errorDetails,
+		});
+	}
+
+	async markAllIncompleteAsFailed() {
+		return await this.update(
+			{ status: In(['new', 'running']) },
+			{ status: 'error', errorCode: 'INTERRUPTED' },
+		);
+	}
+
 	async incrementPassed(id: string, trx?: EntityManager) {
 		trx = trx ?? this.manager;
 		return await trx.increment(TestRun, { id }, 'passedCases', 1);
@@ -51,9 +76,11 @@ export class TestRunRepository extends Repository<TestRun> {
 	}
 
 	async getMany(testDefinitionId: string, options: ListQuery.Options) {
+		// FIXME: optimize fetching final result of each test run
 		const findManyOptions: FindManyOptions<TestRun> = {
 			where: { testDefinition: { id: testDefinitionId } },
 			order: { createdAt: 'DESC' },
+			relations: ['testCaseExecutions'],
 		};
 
 		if (options?.take) {
@@ -61,6 +88,37 @@ export class TestRunRepository extends Repository<TestRun> {
 			findManyOptions.take = options.take;
 		}
 
-		return await this.find(findManyOptions);
+		const testRuns = await this.find(findManyOptions);
+
+		return testRuns.map(({ testCaseExecutions, ...testRun }) => {
+			const finalResult =
+				testRun.status === 'completed' ? getTestRunFinalResult(testCaseExecutions) : null;
+			return { ...testRun, finalResult };
+		});
+	}
+
+	/**
+	 * Test run summary is a TestRun with a final result.
+	 * Final result is calculated based on the status of all test case executions.
+	 * E.g. Test Run is considered successful if all test case executions are successful.
+	 * Test Run is considered failed if at least one test case execution is failed.
+	 */
+	async getTestRunSummaryById(
+		testDefinitionId: string,
+		testRunId: string,
+	): Promise<TestRunSummary> {
+		const testRun = await this.findOne({
+			where: { id: testRunId, testDefinition: { id: testDefinitionId } },
+			relations: ['testCaseExecutions'],
+		});
+
+		if (!testRun) {
+			throw new NotFoundError('Test run not found');
+		}
+
+		testRun.finalResult =
+			testRun.status === 'completed' ? getTestRunFinalResult(testRun.testCaseExecutions) : null;
+
+		return testRun as TestRunSummary;
 	}
 }

packages/cli/src/evaluation.ee/test-definition.service.ee.ts

@@ -126,13 +126,10 @@ export class TestDefinitionService {
 			);
 			const existingNodeIds = new Map(existingTestDefinition.workflow.nodes.map((n) => [n.id, n]));
 
-			attrs.mockedNodes.forEach((node) => {
+			// If some node was previously mocked and then removed from the workflow, it should be removed from the mocked nodes
-				if (!existingNodeIds.has(node.id) || (node.name && !existingNodeNames.has(node.name))) {
+			attrs.mockedNodes = attrs.mockedNodes.filter(
-					throw new BadRequestError(
+				(node) => existingNodeIds.has(node.id) || (node.name && existingNodeNames.has(node.name)),
-						`Pinned node not found in the workflow: ${node.id} (${node.name})`,
+			);
-					);
-				}
-			});
 
 			// Update the node names OR node ids if they are not provided
 			attrs.mockedNodes = attrs.mockedNodes.map((node) => {

packages/cli/src/evaluation.ee/test-runner/__tests__/evaluation-metrics.ee.test.ts

@@ -13,29 +13,23 @@ describe('EvaluationMetrics', () => {
 		expect(aggregatedMetrics).toEqual({ metric1: 0.75, metric2: 0.1 });
 	});
 
-	test('should aggregate only numbers', () => {
+	test('should throw when metric value is not number', () => {
 		const testMetricNames = new Set(['metric1', 'metric2']);
 		const metrics = new EvaluationMetrics(testMetricNames);
 
-		metrics.addResults({ metric1: 1, metric2: 0 });
+		expect(() => metrics.addResults({ metric1: 1, metric2: 0 })).not.toThrow();
-		metrics.addResults({ metric1: '0.5', metric2: 0.2 });
+		expect(() => metrics.addResults({ metric1: '0.5', metric2: 0.2 })).toThrow('INVALID_METRICS');
-		metrics.addResults({ metric1: 'not a number', metric2: [1, 2, 3] });
+		expect(() => metrics.addResults({ metric1: 'not a number', metric2: [1, 2, 3] })).toThrow(
-
+			'INVALID_METRICS',
-		const aggregatedUpMetrics = metrics.getAggregatedMetrics();
+		);
-
-		expect(aggregatedUpMetrics).toEqual({ metric1: 1, metric2: 0.1 });
 	});
 
-	test('should handle missing values', () => {
+	test('should throw when missing values', () => {
 		const testMetricNames = new Set(['metric1', 'metric2']);
 		const metrics = new EvaluationMetrics(testMetricNames);
 
-		metrics.addResults({ metric1: 1 });
+		expect(() => metrics.addResults({ metric1: 1 })).toThrow('METRICS_MISSING');
-		metrics.addResults({ metric2: 0.2 });
+		expect(() => metrics.addResults({ metric2: 0.2 })).toThrow('METRICS_MISSING');
-
-		const aggregatedMetrics = metrics.getAggregatedMetrics();
-
-		expect(aggregatedMetrics).toEqual({ metric1: 1, metric2: 0.2 });
 	});
 
 	test('should handle empty metrics', () => {
@@ -69,4 +63,19 @@ describe('EvaluationMetrics', () => {
 
 		expect(aggregatedMetrics).toEqual({ metric1: 0.75 });
 	});
+
+	test('should report info on added metrics', () => {
+		const testMetricNames = new Set(['metric1']);
+		const metrics = new EvaluationMetrics(testMetricNames);
+		let info;
+
+		expect(() => (info = metrics.addResults({ metric1: 1, metric2: 0 }))).not.toThrow();
+
+		expect(info).toBeDefined();
+		expect(info).toHaveProperty('unknownMetrics');
+		expect(info!.unknownMetrics).toEqual(new Set(['metric2']));
+
+		expect(info).toHaveProperty('addedMetrics');
+		expect(info!.addedMetrics).toEqual({ metric1: 1 });
+	});
 });

packages/cli/src/evaluation.ee/test-runner/__tests__/get-test-run-final-result.ee.test.ts

@@ -0,0 +1,42 @@
+import { mock } from 'jest-mock-extended';
+
+import type { TestCaseExecution } from '@/databases/entities/test-case-execution.ee';
+import { getTestRunFinalResult } from '@/evaluation.ee/test-runner/utils.ee';
+
+function mockTestCaseExecutions(statuses: Array<TestCaseExecution['status']>) {
+	return statuses.map((status) => mock<TestCaseExecution>({ status }));
+}
+
+describe('getTestRunFinalResult', () => {
+	test('should return success if all test cases are successful', () => {
+		const result = getTestRunFinalResult(
+			mockTestCaseExecutions(['success', 'success', 'success', 'success', 'success']),
+		);
+
+		expect(result).toEqual('success');
+	});
+
+	test('should return error if at least one test case is errored', () => {
+		const result = getTestRunFinalResult(
+			mockTestCaseExecutions(['success', 'error', 'success', 'success', 'success']),
+		);
+
+		expect(result).toEqual('error');
+	});
+
+	test('should return warning if at least one test case is warned', () => {
+		const result = getTestRunFinalResult(
+			mockTestCaseExecutions(['success', 'warning', 'success', 'success', 'success']),
+		);
+
+		expect(result).toEqual('warning');
+	});
+
+	test('should return error if there are errors and warnings', () => {
+		const result = getTestRunFinalResult(
+			mockTestCaseExecutions(['success', 'error', 'warning', 'success', 'success']),
+		);
+
+		expect(result).toEqual('error');
+	});
+});

packages/cli/src/evaluation.ee/test-runner/__tests__/mock-data/execution-data-renamed-nodes.json

@@ -0,0 +1,171 @@
+{
+	"startData": {},
+	"resultData": {
+		"runData": {
+			"Manual Run": [
+				{
+					"hints": [],
+					"startTime": 1731079118048,
+					"executionTime": 0,
+					"source": [],
+					"executionStatus": "success",
+					"data": {
+						"main": [
+							[
+								{
+									"json": {
+										"query": "First item"
+									},
+									"pairedItem": {
+										"item": 0
+									}
+								},
+								{
+									"json": {
+										"query": "Second item"
+									},
+									"pairedItem": {
+										"item": 0
+									}
+								},
+								{
+									"json": {
+										"query": "Third item"
+									},
+									"pairedItem": {
+										"item": 0
+									}
+								}
+							]
+						]
+					}
+				}
+			],
+			"Set Attribute": [
+				{
+					"hints": [],
+					"startTime": 1731079118049,
+					"executionTime": 0,
+					"source": [
+						{
+							"previousNode": "Manual Run"
+						}
+					],
+					"executionStatus": "success",
+					"data": {
+						"main": [
+							[
+								{
+									"json": {
+										"foo": "bar"
+									},
+									"pairedItem": {
+										"item": 0
+									}
+								},
+								{
+									"json": {
+										"foo": "bar"
+									},
+									"pairedItem": {
+										"item": 1
+									}
+								},
+								{
+									"json": {
+										"foo": "bar"
+									},
+									"pairedItem": {
+										"item": 2
+									}
+								}
+							]
+						]
+					}
+				}
+			],
+			"Code": [
+				{
+					"hints": [],
+					"startTime": 1731079118049,
+					"executionTime": 3,
+					"source": [
+						{
+							"previousNode": "Set Attribute"
+						}
+					],
+					"executionStatus": "success",
+					"data": {
+						"main": [
+							[
+								{
+									"json": {
+										"foo": "bar",
+										"random": 0.6315509336851373
+									},
+									"pairedItem": {
+										"item": 0
+									}
+								},
+								{
+									"json": {
+										"foo": "bar",
+										"random": 0.3336315687359024
+									},
+									"pairedItem": {
+										"item": 1
+									}
+								},
+								{
+									"json": {
+										"foo": "bar",
+										"random": 0.4241870158917733
+									},
+									"pairedItem": {
+										"item": 2
+									}
+								}
+							]
+						]
+					}
+				}
+			]
+		},
+		"pinData": {
+			"Manual Run": [
+				{
+					"json": {
+						"query": "First item"
+					},
+					"pairedItem": {
+						"item": 0
+					}
+				},
+				{
+					"json": {
+						"query": "Second item"
+					},
+					"pairedItem": {
+						"item": 0
+					}
+				},
+				{
+					"json": {
+						"query": "Third item"
+					},
+					"pairedItem": {
+						"item": 0
+					}
+				}
+			]
+		},
+		"lastNodeExecuted": "Code"
+	},
+	"executionData": {
+		"contextData": {},
+		"nodeExecutionStack": [],
+		"metadata": {},
+		"waitingExecution": {},
+		"waitingExecutionSource": {}
+	}
+}

packages/cli/src/evaluation.ee/test-runner/__tests__/test-runner.service.ee.test.ts

@@ -4,6 +4,7 @@ import { readFileSync } from 'fs';
 import { mock, mockDeep } from 'jest-mock-extended';
 import type { ErrorReporter } from 'n8n-core';
 import type { ExecutionError, GenericValue, IRun } from 'n8n-workflow';
+import type { ITaskData } from 'n8n-workflow';
 import path from 'path';
 
 import type { ActiveExecutions } from '@/active-executions';
@@ -54,6 +55,12 @@ const executionDataJson = JSON.parse(
 	readFileSync(path.join(__dirname, './mock-data/execution-data.json'), { encoding: 'utf-8' }),
 );
 
+const executionDataRenamedNodesJson = JSON.parse(
+	readFileSync(path.join(__dirname, './mock-data/execution-data-renamed-nodes.json'), {
+		encoding: 'utf-8',
+	}),
+);
+
 const executionDataMultipleTriggersJson = JSON.parse(
 	readFileSync(path.join(__dirname, './mock-data/execution-data.multiple-triggers.json'), {
 		encoding: 'utf-8',
@@ -81,7 +88,7 @@ const executionMocks = [
 		workflowId: 'workflow-under-test-id',
 		status: 'success',
 		executionData: {
-			data: stringify(executionDataJson),
+			data: stringify(executionDataRenamedNodesJson),
 			workflowData: wfUnderTestRenamedNodesJson,
 		},
 	}),
@@ -91,7 +98,12 @@ function mockExecutionData() {
 	return mock<IRun>({
 		data: {
 			resultData: {
-				runData: {},
+				runData: {
+					'When clicking ‘Test workflow’': mock<ITaskData[]>(),
+				},
+				// error is an optional prop, but jest-mock-extended will mock it by default,
+				// which affects the code logic. So, we need to explicitly set it to undefined.
+				error: undefined,
 			},
 		},
 	});
@@ -292,7 +304,7 @@ describe('TestRunnerService', () => {
 
 		activeExecutions.getPostExecutePromise
 			.calledWith('some-execution-id-4')
-			.mockResolvedValue(mockEvaluationExecutionData({ metric1: 0.5 }));
+			.mockResolvedValue(mockEvaluationExecutionData({ metric1: 0.5, metric2: 100 }));
 
 		await testRunnerService.runTest(
 			mock<User>(),
@@ -322,7 +334,7 @@ describe('TestRunnerService', () => {
 		// Check evaluation workflow was executed
 		expect(workflowRunner.run).toHaveBeenCalledWith(
 			expect.objectContaining({
-				executionMode: 'evaluation',
+				executionMode: 'integrated',
 				executionData: expect.objectContaining({
 					executionData: expect.objectContaining({
 						nodeExecutionStack: expect.arrayContaining([
@@ -343,7 +355,7 @@ describe('TestRunnerService', () => {
 		expect(testRunRepository.markAsCompleted).toHaveBeenCalledTimes(1);
 		expect(testRunRepository.markAsCompleted).toHaveBeenCalledWith('test-run-id', {
 			metric1: 0.75,
-			metric2: 0,
+			metric2: 50,
 		});
 
 		expect(testRunRepository.incrementPassed).toHaveBeenCalledTimes(2);
@@ -624,6 +636,7 @@ describe('TestRunnerService', () => {
 		const startNodesData = (testRunnerService as any).getStartNodesData(
 			wfMultipleTriggersJson,
 			executionDataMultipleTriggersJson,
+			wfMultipleTriggersJson, // Test case where workflow didn't change
 		);
 
 		expect(startNodesData).toEqual({
@@ -652,6 +665,7 @@ describe('TestRunnerService', () => {
 		const startNodesData = (testRunnerService as any).getStartNodesData(
 			wfMultipleTriggersJson,
 			executionDataMultipleTriggersJson2,
+			wfMultipleTriggersJson, // Test case where workflow didn't change
 		);
 
 		expect(startNodesData).toEqual({
@@ -662,6 +676,35 @@ describe('TestRunnerService', () => {
 		});
 	});
 
+	test('should properly choose trigger when it was renamed', async () => {
+		const testRunnerService = new TestRunnerService(
+			logger,
+			telemetry,
+			workflowRepository,
+			workflowRunner,
+			executionRepository,
+			activeExecutions,
+			testRunRepository,
+			testCaseExecutionRepository,
+			testMetricRepository,
+			mockNodeTypes,
+			errorReporter,
+		);
+
+		const startNodesData = (testRunnerService as any).getStartNodesData(
+			wfUnderTestRenamedNodesJson, // Test case where workflow didn't change
+			executionDataJson,
+			wfUnderTestJson,
+		);
+
+		expect(startNodesData).toEqual({
+			startNodes: expect.arrayContaining([expect.objectContaining({ name: 'Set attribute' })]),
+			triggerToStartFrom: expect.objectContaining({
+				name: 'Manual Run',
+			}),
+		});
+	});
+
 	describe('Test Run cancellation', () => {
 		beforeAll(() => {
 			jest.useFakeTimers();

packages/cli/src/evaluation.ee/test-runner/errors.ee.ts

@@ -0,0 +1,39 @@
+import { ApplicationError } from 'n8n-workflow';
+
+export type TestCaseExecutionErrorCode =
+	| 'MOCKED_NODE_DOES_NOT_EXIST'
+	| 'TRIGGER_NO_LONGER_EXISTS'
+	| 'FAILED_TO_EXECUTE_WORKFLOW'
+	| 'EVALUATION_WORKFLOW_DOES_NOT_EXIST'
+	| 'FAILED_TO_EXECUTE_EVALUATION_WORKFLOW'
+	| 'METRICS_MISSING'
+	| 'UNKNOWN_METRICS'
+	| 'INVALID_METRICS'
+	| 'PAYLOAD_LIMIT_EXCEEDED'
+	| 'UNKNOWN_ERROR';
+
+export class TestCaseExecutionError extends ApplicationError {
+	readonly code: TestCaseExecutionErrorCode;
+
+	constructor(code: TestCaseExecutionErrorCode, extra: Record<string, unknown> = {}) {
+		super('Test Case execution failed with code ' + code, { extra });
+
+		this.code = code;
+	}
+}
+
+export type TestRunErrorCode =
+	| 'PAST_EXECUTIONS_NOT_FOUND'
+	| 'EVALUATION_WORKFLOW_NOT_FOUND'
+	| 'INTERRUPTED'
+	| 'UNKNOWN_ERROR';
+
+export class TestRunError extends ApplicationError {
+	readonly code: TestRunErrorCode;
+
+	constructor(code: TestRunErrorCode, extra: Record<string, unknown> = {}) {
+		super('Test Run failed with code ' + code, { extra });
+
+		this.code = code;
+	}
+}

packages/cli/src/evaluation.ee/test-runner/evaluation-metrics.ee.ts

@@ -1,5 +1,15 @@
+import difference from 'lodash/difference';
 import type { IDataObject } from 'n8n-workflow';
 
+import { TestCaseExecutionError } from '@/evaluation.ee/test-runner/errors.ee';
+
+export interface EvaluationMetricsAddResultsInfo {
+	addedMetrics: Record<string, number>;
+	missingMetrics: Set<string>;
+	unknownMetrics: Set<string>;
+	incorrectTypeMetrics: Set<string>;
+}
+
 export class EvaluationMetrics {
 	private readonly rawMetricsByName = new Map<string, number[]>();
 
@@ -9,17 +19,41 @@ export class EvaluationMetrics {
 		}
 	}
 
-	addResults(result: IDataObject): Record<string, number> {
+	addResults(result: IDataObject): EvaluationMetricsAddResultsInfo {
-		const addedMetrics: Record<string, number> = {};
+		const addResultsInfo: EvaluationMetricsAddResultsInfo = {
+			addedMetrics: {},
+			missingMetrics: new Set<string>(),
+			unknownMetrics: new Set<string>(),
+			incorrectTypeMetrics: new Set<string>(),
+		};
 
 		for (const [metricName, metricValue] of Object.entries(result)) {
-			if (typeof metricValue === 'number' && this.metricNames.has(metricName)) {
+			if (this.metricNames.has(metricName)) {
-				addedMetrics[metricName] = metricValue;
+				if (typeof metricValue === 'number') {
-				this.rawMetricsByName.get(metricName)!.push(metricValue);
+					addResultsInfo.addedMetrics[metricName] = metricValue;
+					this.rawMetricsByName.get(metricName)!.push(metricValue);
+				} else {
+					throw new TestCaseExecutionError('INVALID_METRICS', {
+						metricName,
+						metricValue,
+					});
+				}
+			} else {
+				addResultsInfo.unknownMetrics.add(metricName);
 			}
 		}
 
-		return addedMetrics;
+		// Check that result contains all expected metrics
+		if (
+			difference(Array.from(this.metricNames), Object.keys(addResultsInfo.addedMetrics)).length > 0
+		) {
+			throw new TestCaseExecutionError('METRICS_MISSING', {
+				expectedMetrics: Array.from(this.metricNames).sort(),
+				receivedMetrics: Object.keys(addResultsInfo.addedMetrics).sort(),
+			});
+		}
+
+		return addResultsInfo;
 	}
 
 	getAggregatedMetrics() {

packages/cli/src/evaluation.ee/test-runner/test-runner.service.ee.ts

@@ -24,6 +24,7 @@ import { TestMetricRepository } from '@/databases/repositories/test-metric.repos
 import { TestRunRepository } from '@/databases/repositories/test-run.repository.ee';
 import { WorkflowRepository } from '@/databases/repositories/workflow.repository';
 import * as Db from '@/db';
+import { TestCaseExecutionError, TestRunError } from '@/evaluation.ee/test-runner/errors.ee';
 import { NodeTypes } from '@/node-types';
 import { Telemetry } from '@/telemetry';
 import { getRunData } from '@/workflow-execute-additional-data';
@@ -67,12 +68,21 @@ export class TestRunnerService {
 		private readonly errorReporter: ErrorReporter,
 	) {}
 
+	/**
+	 * As Test Runner does not have a recovery mechanism, it can not resume Test Runs interrupted by the server restart.
+	 * All Test Runs in incomplete state will be marked as cancelled.
+	 */
+	async cleanupIncompleteRuns() {
+		await this.testRunRepository.markAllIncompleteAsFailed();
+	}
+
 	/**
 	 * Prepares the start nodes and trigger node data props for the `workflowRunner.run` method input.
 	 */
 	private getStartNodesData(
 		workflow: WorkflowEntity,
 		pastExecutionData: IRunExecutionData,
+		pastExecutionWorkflowData: IWorkflowBase,
 	): Pick<IWorkflowExecutionDataProcess, 'startNodes' | 'triggerToStartFrom'> {
 		// Create a new workflow instance to use the helper functions (getChildNodes)
 		const workflowInstance = new Workflow({
@@ -82,21 +92,38 @@ export class TestRunnerService {
 			nodeTypes: this.nodeTypes,
 		});
 
+		// Create a map between node IDs and node names for the past workflow
+		const pastWorkflowNodeIdByName = new Map(
+			pastExecutionWorkflowData.nodes.map((node) => [node.name, node.id]),
+		);
+
+		// Create a map between node names and IDs for the up-to-date workflow
+		const workflowNodeNameById = new Map(workflow.nodes.map((node) => [node.id, node.name]));
+
 		// Determine the trigger node of the past execution
 		const pastExecutionTriggerNode = getPastExecutionTriggerNode(pastExecutionData);
 		assert(pastExecutionTriggerNode, 'Could not find the trigger node of the past execution');
 
+		const pastExecutionTriggerNodeId = pastWorkflowNodeIdByName.get(pastExecutionTriggerNode);
+		assert(pastExecutionTriggerNodeId, 'Could not find the trigger node ID of the past execution');
+
+		// Check the trigger is still present in the workflow
+		const triggerNode = workflowNodeNameById.get(pastExecutionTriggerNodeId);
+		if (!triggerNode) {
+			throw new TestCaseExecutionError('TRIGGER_NO_LONGER_EXISTS');
+		}
+
 		const triggerNodeData = pastExecutionData.resultData.runData[pastExecutionTriggerNode][0];
 		assert(triggerNodeData, 'Trigger node data not found');
 
 		const triggerToStartFrom = {
-			name: pastExecutionTriggerNode,
+			name: triggerNode,
 			data: triggerNodeData,
 		};
 
 		// Start nodes are the nodes that are connected to the trigger node
 		const startNodes = workflowInstance
-			.getChildNodes(pastExecutionTriggerNode, NodeConnectionType.Main, 1)
+			.getChildNodes(triggerNode, NodeConnectionType.Main, 1)
 			.map((nodeName) => ({
 				name: nodeName,
 				sourceData: { previousNode: pastExecutionTriggerNode },
@@ -135,7 +162,7 @@ export class TestRunnerService {
 
 		// Prepare the data to run the workflow
 		const data: IWorkflowExecutionDataProcess = {
-			...this.getStartNodesData(workflow, pastExecutionData),
+			...this.getStartNodesData(workflow, pastExecutionData, pastExecutionWorkflowData),
 			executionMode: 'evaluation',
 			runData: {},
 			pinData,
@@ -154,11 +181,11 @@ export class TestRunnerService {
 		});
 
 		// Update status of the test run execution mapping
-		await this.testCaseExecutionRepository.markAsRunning(
+		await this.testCaseExecutionRepository.markAsRunning({
-			metadata.testRunId,
+			testRunId: metadata.testRunId,
-			metadata.pastExecutionId,
+			pastExecutionId: metadata.pastExecutionId,
 			executionId,
-		);
+		});
 
 		// Wait for the execution to finish
 		const executePromise = this.activeExecutions.getPostExecutePromise(executionId);
@@ -192,7 +219,7 @@ export class TestRunnerService {
 
 		// Prepare the data to run the evaluation workflow
 		const data = await getRunData(evaluationWorkflow, [evaluationInputData]);
-		data.executionMode = 'evaluation';
+		data.executionMode = 'integrated';
 
 		// Trigger the evaluation workflow
 		const executionId = await this.workflowRunner.run(data);
@@ -204,11 +231,11 @@ export class TestRunnerService {
 		});
 
 		// Update status of the test run execution mapping
-		await this.testCaseExecutionRepository.markAsEvaluationRunning(
+		await this.testCaseExecutionRepository.markAsEvaluationRunning({
-			metadata.testRunId,
+			testRunId: metadata.testRunId,
-			metadata.pastExecutionId,
+			pastExecutionId: metadata.pastExecutionId,
-			executionId,
+			evaluationExecutionId: executionId,
-		);
+		});
 
 		// Wait for the execution to finish
 		const executePromise = this.activeExecutions.getPostExecutePromise(executionId);
@@ -256,9 +283,6 @@ export class TestRunnerService {
 		const workflow = await this.workflowRepository.findById(test.workflowId);
 		assert(workflow, 'Workflow not found');
 
-		const evaluationWorkflow = await this.workflowRepository.findById(test.evaluationWorkflowId);
-		assert(evaluationWorkflow, 'Evaluation workflow not found');
-
 		// 0. Create new Test Run
 		const testRun = await this.testRunRepository.createTestRun(test.id);
 		assert(testRun, 'Unable to create a test run');
@@ -276,6 +300,12 @@ export class TestRunnerService {
 
 		const abortSignal = abortController.signal;
 		try {
+			// Get the evaluation workflow
+			const evaluationWorkflow = await this.workflowRepository.findById(test.evaluationWorkflowId);
+			if (!evaluationWorkflow) {
+				throw new TestRunError('EVALUATION_WORKFLOW_NOT_FOUND');
+			}
+
 			///
 			// 1. Make test cases from previous executions
 			///
@@ -294,6 +324,10 @@ export class TestRunnerService {
 
 			this.logger.debug('Found past executions', { count: pastExecutions.length });
 
+			if (pastExecutions.length === 0) {
+				throw new TestRunError('PAST_EXECUTIONS_NOT_FOUND');
+			}
+
 			// Add all past executions mappings to the test run.
 			// This will be used to track the status of each test case and keep the connection between test run and all related executions (past, current, and evaluation).
 			await this.testCaseExecutionRepository.createBatch(
@@ -365,20 +399,20 @@ export class TestRunnerService {
 					this.logger.debug('Test case execution finished', { pastExecutionId });
 
 					// In case of a permission check issue, the test case execution will be undefined.
-					// Skip them, increment the failed count and continue with the next test case
+					// If that happens, or if the test case execution produced an error, mark the test case as failed.
-					if (!testCaseExecution) {
+					if (!testCaseExecution || testCaseExecution.data.resultData.error) {
 						await Db.transaction(async (trx) => {
 							await this.testRunRepository.incrementFailed(testRun.id, trx);
-							await this.testCaseExecutionRepository.markAsFailed(testRun.id, pastExecutionId, trx);
+							await this.testCaseExecutionRepository.markAsFailed({
+								testRunId: testRun.id,
+								pastExecutionId,
+								errorCode: 'FAILED_TO_EXECUTE_WORKFLOW',
+								trx,
+							});
 						});
 						continue;
 					}
 
-					// Update status of the test case execution mapping entry in case of an error
-					if (testCaseExecution.data.resultData.error) {
-						await this.testCaseExecutionRepository.markAsFailed(testRun.id, pastExecutionId);
-					}
-
 					// Collect the results of the test case execution
 					const testCaseRunData = testCaseExecution.data.resultData.runData;
 
@@ -398,32 +432,67 @@ export class TestRunnerService {
 					this.logger.debug('Evaluation execution finished', { pastExecutionId });
 
 					// Extract the output of the last node executed in the evaluation workflow
-					const addedMetrics = metrics.addResults(this.extractEvaluationResult(evalExecution));
+					const { addedMetrics, unknownMetrics } = metrics.addResults(
+						this.extractEvaluationResult(evalExecution),
+					);
 
 					if (evalExecution.data.resultData.error) {
 						await Db.transaction(async (trx) => {
 							await this.testRunRepository.incrementFailed(testRun.id, trx);
-							await this.testCaseExecutionRepository.markAsFailed(testRun.id, pastExecutionId, trx);
+							await this.testCaseExecutionRepository.markAsFailed({
+								testRunId: testRun.id,
+								pastExecutionId,
+								errorCode: 'FAILED_TO_EXECUTE_EVALUATION_WORKFLOW',
+								trx,
+							});
 						});
 					} else {
 						await Db.transaction(async (trx) => {
 							await this.testRunRepository.incrementPassed(testRun.id, trx);
-							await this.testCaseExecutionRepository.markAsCompleted(
+
-								testRun.id,
+							// Add warning if the evaluation workflow produced an unknown metric
-								pastExecutionId,
+							if (unknownMetrics.size > 0) {
-								addedMetrics,
+								await this.testCaseExecutionRepository.markAsWarning({
-								trx,
+									testRunId: testRun.id,
-							);
+									pastExecutionId,
+									errorCode: 'UNKNOWN_METRICS',
+									errorDetails: { unknownMetrics: Array.from(unknownMetrics) },
+								});
+							} else {
+								await this.testCaseExecutionRepository.markAsCompleted({
+									testRunId: testRun.id,
+									pastExecutionId,
+									metrics: addedMetrics,
+									trx,
+								});
+							}
 						});
 					}
 				} catch (e) {
 					// In case of an unexpected error, increment the failed count and continue with the next test case
 					await Db.transaction(async (trx) => {
 						await this.testRunRepository.incrementFailed(testRun.id, trx);
-						await this.testCaseExecutionRepository.markAsFailed(testRun.id, pastExecutionId, trx);
-					});
 
-					this.errorReporter.error(e);
+						if (e instanceof TestCaseExecutionError) {
+							await this.testCaseExecutionRepository.markAsFailed({
+								testRunId: testRun.id,
+								pastExecutionId,
+								errorCode: e.code,
+								errorDetails: e.extra as IDataObject,
+								trx,
+							});
+						} else {
+							await this.testCaseExecutionRepository.markAsFailed({
+								testRunId: testRun.id,
+								pastExecutionId,
+								errorCode: 'UNKNOWN_ERROR',
+								trx,
+							});
+
+							// Report unexpected errors
+							this.errorReporter.error(e);
+						}
+					});
 				}
 			}
 
@@ -435,6 +504,7 @@ export class TestRunnerService {
 				});
 			} else {
 				const aggregatedMetrics = metrics.getAggregatedMetrics();
+
 				await this.testRunRepository.markAsCompleted(testRun.id, aggregatedMetrics);
 
 				this.logger.debug('Test run finished', { testId: test.id });
@@ -450,7 +520,10 @@ export class TestRunnerService {
 					await this.testRunRepository.markAsCancelled(testRun.id, trx);
 					await this.testCaseExecutionRepository.markAllPendingAsCancelled(testRun.id, trx);
 				});
+			} else if (e instanceof TestRunError) {
+				await this.testRunRepository.markAsError(testRun.id, e.code, e.extra as IDataObject);
 			} else {
+				await this.testRunRepository.markAsError(testRun.id, 'UNKNOWN_ERROR');
 				throw e;
 			}
 		} finally {

packages/cli/src/evaluation.ee/test-runner/utils.ee.ts

@@ -1,8 +1,11 @@
 import assert from 'assert';
 import type { IRunExecutionData, IPinData, IWorkflowBase } from 'n8n-workflow';
 
+import type { TestCaseExecution } from '@/databases/entities/test-case-execution.ee';
 import type { MockedNodeItem } from '@/databases/entities/test-definition.ee';
 import type { WorkflowEntity } from '@/databases/entities/workflow-entity';
+import type { TestRunFinalResult } from '@/databases/repositories/test-run.repository.ee';
+import { TestCaseExecutionError } from '@/evaluation.ee/test-runner/errors.ee';
 
 /**
  * Extracts the execution data from the past execution
@@ -41,6 +44,8 @@ export function createPinData(
 
 			if (nodeData?.[0]?.data?.main?.[0]) {
 				pinData[nodeName] = nodeData[0]?.data?.main?.[0];
+			} else {
+				throw new TestCaseExecutionError('MOCKED_NODE_DOES_NOT_EXIST');
 			}
 		}
 	}
@@ -58,3 +63,31 @@ export function getPastExecutionTriggerNode(executionData: IRunExecutionData) {
 		return !data[0].source || data[0].source.length === 0 || data[0].source[0] === null;
 	});
 }
+
+/**
+ * Returns the final result of the test run based on the test case executions.
+ * The final result is the most severe status among all test case executions' statuses.
+ */
+export function getTestRunFinalResult(testCaseExecutions: TestCaseExecution[]): TestRunFinalResult {
+	// Priority of statuses: error > warning > success
+	const severityMap: Record<TestRunFinalResult, number> = {
+		error: 3,
+		warning: 2,
+		success: 1,
+	};
+
+	let finalResult: TestRunFinalResult = 'success';
+
+	for (const testCaseExecution of testCaseExecutions) {
+		if (['error', 'warning'].includes(testCaseExecution.status)) {
+			if (
+				testCaseExecution.status in severityMap &&
+				severityMap[testCaseExecution.status as TestRunFinalResult] > severityMap[finalResult]
+			) {
+				finalResult = testCaseExecution.status as TestRunFinalResult;
+			}
+		}
+	}
+
+	return finalResult;
+}

packages/cli/src/evaluation.ee/test-runs.controller.ee.ts

@@ -73,9 +73,11 @@ export class TestRunsController {
 
 	@Get('/:testDefinitionId/runs/:id')
 	async getOne(req: TestRunsRequest.GetOne) {
+		const { testDefinitionId, id } = req.params;
+
 		await this.getTestDefinition(req);
 
-		return await this.getTestRun(req);
+		return await this.testRunRepository.getTestRunSummaryById(testDefinitionId, id);
 	}
 
 	@Get('/:testDefinitionId/runs/:id/cases')

packages/editor-ui/src/api/testDefinition.ee.ts

@@ -43,12 +43,15 @@ export interface UpdateTestResponse {
 export interface TestRunRecord {
 	id: string;
 	testDefinitionId: string;
-	status: 'new' | 'running' | 'completed' | 'error' | 'cancelled';
+	status: 'new' | 'running' | 'completed' | 'error' | 'cancelled' | 'warning' | 'success';
 	metrics?: Record<string, number>;
 	createdAt: string;
 	updatedAt: string;
 	runAt: string;
 	completedAt: string;
+	errorCode?: string;
+	errorDetails?: Record<string, unknown>;
+	finalResult?: 'success' | 'error' | 'warning';
 }
 
 interface GetTestRunParams {

packages/editor-ui/src/components/TestDefinition/ListRuns/TestRunsTable.vue

@@ -22,6 +22,17 @@ const locale = useI18n();
 const navigateToRunDetail = (run: TestRunRecord) => emit('getRunDetail', run);
 const selectedRows = ref<TestRunRecord[]>([]);
 
+// Combine test run statuses and finalResult to get the final status
+const runSummaries = computed(() => {
+	return props.runs.map(({ status, finalResult, ...run }) => {
+		if (status === 'completed' && finalResult) {
+			return { ...run, status: finalResult };
+		}
+
+		return { ...run, status };
+	});
+});
+
 const metrics = computed(() => {
 	return props.runs.reduce((acc, run) => {
 		const metricKeys = Object.keys(run.metrics ?? {});
@@ -29,6 +40,19 @@ const metrics = computed(() => {
 	}, [] as string[]);
 });
 
+const getErrorTooltipLinkRoute = (row: TestRunRecord) => {
+	if (row.errorCode === 'EVALUATION_WORKFLOW_NOT_FOUND') {
+		return {
+			name: VIEWS.TEST_DEFINITION_EDIT,
+			params: {
+				testId: row.testDefinitionId,
+			},
+		};
+	}
+
+	return undefined;
+};
+
 const columns = computed((): Array<TestTableColumn<TestRunRecord>> => {
 	return [
 		{
@@ -51,15 +75,17 @@ const columns = computed((): Array<TestTableColumn<TestRunRecord>> => {
 				{ text: locale.baseText('testDefinition.listRuns.status.error'), value: 'error' },
 				{ text: locale.baseText('testDefinition.listRuns.status.cancelled'), value: 'cancelled' },
 			],
+			errorRoute: getErrorTooltipLinkRoute,
 			filterMethod: (value: string, row: TestRunRecord) => row.status === value,
 		},
 		{
 			prop: 'date',
 			label: locale.baseText('testDefinition.listRuns.runDate'),
 			sortable: true,
-			formatter: (row: TestRunRecord) => convertToDisplayDate(new Date(row.runAt).getTime()),
+			formatter: (row: TestRunRecord) =>
+				convertToDisplayDate(new Date(row.runAt ?? row.createdAt).getTime()),
 			sortMethod: (a: TestRunRecord, b: TestRunRecord) =>
-				new Date(a.runAt).getTime() - new Date(b.runAt).getTime(),
+				new Date(a.runAt ?? a.createdAt).getTime() - new Date(b.runAt ?? b.createdAt).getTime(),
 		},
 
 		...metrics.value.map((metric) => ({
@@ -104,7 +130,7 @@ async function deleteRuns() {
 			</n8n-button>
 		</div>
 		<TestTableBase
-			:data="runs"
+			:data="runSummaries"
 			:columns="columns"
 			selectable
 			@row-click="navigateToRunDetail"

packages/editor-ui/src/components/TestDefinition/shared/TableCell.vue

@@ -1,5 +1,4 @@
 <script setup lang="ts" generic="T">
-import { useI18n } from '@/composables/useI18n';
 import type { TestTableColumn } from './TestTableBase.vue';
 import { useRouter } from 'vue-router';
 
@@ -12,35 +11,7 @@ defineEmits<{
 	click: [];
 }>();
 
-const locale = useI18n();
 const router = useRouter();
-interface WithStatus {
-	status: string;
-}
-
-function hasStatus(row: unknown): row is WithStatus {
-	return typeof row === 'object' && row !== null && 'status' in row;
-}
-
-const statusThemeMap: Record<string, string> = {
-	new: 'default',
-	running: 'warning',
-	evaluation_running: 'warning',
-	completed: 'success',
-	error: 'danger',
-	success: 'success',
-	cancelled: 'default',
-};
-
-const statusLabelMap: Record<string, string> = {
-	new: locale.baseText('testDefinition.listRuns.status.new'),
-	running: locale.baseText('testDefinition.listRuns.status.running'),
-	evaluation_running: locale.baseText('testDefinition.listRuns.status.evaluating'),
-	completed: locale.baseText('testDefinition.listRuns.status.completed'),
-	error: locale.baseText('testDefinition.listRuns.status.error'),
-	success: locale.baseText('testDefinition.listRuns.status.success'),
-	cancelled: locale.baseText('testDefinition.listRuns.status.cancelled'),
-};
 
 function hasProperty(row: unknown, prop: string): row is Record<string, unknown> {
 	return typeof row === 'object' && row !== null && prop in row;
@@ -64,14 +35,6 @@ const getCellContent = (column: TestTableColumn<T>, row: T) => {
 		</router-link>
 	</div>
 
-	<N8nBadge
-		v-else-if="column.prop === 'status' && hasStatus(row)"
-		:theme="statusThemeMap[row.status]"
-		class="mr-4xs"
-	>
-		{{ statusLabelMap[row.status] }}
-	</N8nBadge>
-
 	<div v-else>
 		{{ getCellContent(column, row) }}
 	</div>

packages/editor-ui/src/components/TestDefinition/shared/TableStatusCell.vue

@@ -0,0 +1,104 @@
+<script setup lang="ts" generic="T">
+import N8nTooltip from 'n8n-design-system/components/N8nTooltip';
+import type { BaseTextKey } from '@/plugins/i18n';
+import type { TestTableColumn } from '@/components/TestDefinition/shared/TestTableBase.vue';
+import { useI18n } from '@/composables/useI18n';
+import { useRouter } from 'vue-router';
+
+defineProps<{
+	column: TestTableColumn<T>;
+	row: T & { status: string };
+}>();
+
+const locale = useI18n();
+const router = useRouter();
+
+interface WithError {
+	errorCode: string;
+}
+
+function hasError(row: unknown): row is WithError {
+	return typeof row === 'object' && row !== null && 'errorCode' in row;
+}
+
+const errorTooltipMap: Record<string, BaseTextKey> = {
+	// Test case errors
+	MOCKED_NODE_DOES_NOT_EXIST: 'testDefinition.runDetail.error.mockedNodeMissing',
+	FAILED_TO_EXECUTE_EVALUATION_WORKFLOW: 'testDefinition.runDetail.error.evaluationFailed',
+	FAILED_TO_EXECUTE_WORKFLOW: 'testDefinition.runDetail.error.executionFailed',
+	TRIGGER_NO_LONGER_EXISTS: 'testDefinition.runDetail.error.triggerNoLongerExists',
+	METRICS_MISSING: 'testDefinition.runDetail.error.metricsMissing',
+	UNKNOWN_METRICS: 'testDefinition.runDetail.error.unknownMetrics',
+	INVALID_METRICS: 'testDefinition.runDetail.error.invalidMetrics',
+
+	// Test run errors
+	PAST_EXECUTIONS_NOT_FOUND: 'testDefinition.listRuns.error.noPastExecutions',
+	EVALUATION_WORKFLOW_NOT_FOUND: 'testDefinition.listRuns.error.evaluationWorkflowNotFound',
+};
+
+// FIXME: move status logic to a parent component
+const statusThemeMap: Record<string, string> = {
+	new: 'default',
+	running: 'warning',
+	evaluation_running: 'warning',
+	completed: 'success',
+	error: 'danger',
+	success: 'success',
+	warning: 'warning',
+	cancelled: 'default',
+};
+
+const statusLabelMap: Record<string, string> = {
+	new: locale.baseText('testDefinition.listRuns.status.new'),
+	running: locale.baseText('testDefinition.listRuns.status.running'),
+	evaluation_running: locale.baseText('testDefinition.listRuns.status.evaluating'),
+	completed: locale.baseText('testDefinition.listRuns.status.completed'),
+	error: locale.baseText('testDefinition.listRuns.status.error'),
+	success: locale.baseText('testDefinition.listRuns.status.success'),
+	warning: locale.baseText('testDefinition.listRuns.status.warning'),
+	cancelled: locale.baseText('testDefinition.listRuns.status.cancelled'),
+};
+
+function getErrorTooltip(column: TestTableColumn<T>, row: T): string | undefined {
+	if (hasError(row) && errorTooltipMap[row.errorCode]) {
+		const tooltipLinkUrl = getErrorTooltipUrl(column, row);
+
+		if (tooltipLinkUrl) {
+			return locale.baseText(errorTooltipMap[row.errorCode], {
+				interpolate: {
+					url: tooltipLinkUrl,
+				},
+			});
+		} else {
+			return locale.baseText(errorTooltipMap[row.errorCode]);
+		}
+	}
+
+	return undefined;
+}
+
+function getErrorTooltipUrl(column: TestTableColumn<T>, row: T): string | undefined {
+	if (hasError(row) && column.errorRoute?.(row)) {
+		return router.resolve(column.errorRoute(row)!).href;
+	}
+
+	return undefined;
+}
+</script>
+
+<template>
+	<N8nTooltip
+		placement="right"
+		:show-after="300"
+		:disabled="getErrorTooltip(column, row) === undefined"
+	>
+		<template #content>
+			<div v-n8n-html="getErrorTooltip(column, row)" />
+		</template>
+		<N8nBadge :theme="statusThemeMap[row.status]" class="mr-4xs">
+			{{ statusLabelMap[row.status] }}
+		</N8nBadge>
+	</N8nTooltip>
+</template>
+
+<style scoped lang="scss"></style>

packages/editor-ui/src/components/TestDefinition/shared/TestTableBase.vue

@@ -1,6 +1,7 @@
 <script setup lang="ts" generic="T extends object">
 import type { RouteLocationRaw } from 'vue-router';
 import TableCell from './TableCell.vue';
+import TableStatusCell from './TableStatusCell.vue';
 import { ElTable, ElTableColumn } from 'element-plus';
 import { ref, watch, nextTick, onMounted, onUnmounted } from 'vue';
 import type { TableInstance } from 'element-plus';
@@ -22,6 +23,7 @@ export type TestTableColumn<TRow> = {
 	filters?: Array<{ text: string; value: string }>;
 	filterMethod?: (value: string, row: TRow) => boolean;
 	route?: (row: TRow) => RouteLocationRaw | undefined;
+	errorRoute?: (row: TRow) => RouteLocationRaw | undefined;
 	sortMethod?: (a: TRow, b: TRow) => number;
 	openInNewTab?: boolean;
 	formatter?: (row: TRow) => string;
@@ -29,6 +31,8 @@ export type TestTableColumn<TRow> = {
 
 type TableRow = T & { id: string };
 
+type TableRowWithStatus = TableRow & { status: string };
+
 const MIN_TABLE_HEIGHT = 350;
 const MAX_TABLE_HEIGHT = 1400;
 const props = withDefaults(
@@ -94,6 +98,10 @@ const computeTableHeight = () => {
 	tableHeight.value = `${height - 100}px`;
 };
 
+function hasStatus(row: unknown): row is TableRowWithStatus {
+	return typeof row === 'object' && row !== null && 'status' in row;
+}
+
 onMounted(() => {
 	computeTableHeight();
 
@@ -131,8 +139,14 @@ onUnmounted(() => {
 			data-test-id="table-column"
 		>
 			<template #default="{ row }">
+				<TableStatusCell
+					v-if="column.prop === 'status' && hasStatus(row)"
+					:column="column"
+					:row="row"
+				/>
 				<TableCell
-					:key="row.status"
+					v-else
+					:key="row.id + column.prop"
 					:column="column"
 					:row="row"
 					data-test-id="table-cell"

packages/editor-ui/src/plugins/i18n/locales/en.json

@@ -2875,6 +2875,7 @@
 	"testDefinition.listRuns.status.cancelled": "Cancelled",
 	"testDefinition.listRuns.status.error": "Error",
 	"testDefinition.listRuns.status.success": "Success",
+	"testDefinition.listRuns.status.warning": "Warning",
 	"testDefinition.listRuns.metricsOverTime": "Metrics over time",
 	"testDefinition.listRuns.status": "Status",
 	"testDefinition.listRuns.runNumber": "Run #",
@@ -2884,11 +2885,20 @@
 	"testDefinition.listRuns.noRuns.description": "Run a test to see the results here",
 	"testDefinition.listRuns.deleteRuns": "No runs to delete | Delete {count} run | Delete {count} runs",
 	"testDefinition.listRuns.noRuns.button": "Run Test",
+	"testDefinition.listRuns.error.noPastExecutions": "No executions added to the specified tag",
+	"testDefinition.listRuns.error.evaluationWorkflowNotFound": "Selected evaluation workflow does not exist. <br /><a href=\"{url}\">Fix test configuration</a>.",
 	"testDefinition.runDetail.ranAt": "Ran at",
 	"testDefinition.runDetail.testCase": "Test case",
 	"testDefinition.runDetail.testCase.id": "Test case ID",
 	"testDefinition.runDetail.testCase.status": "Test case status",
 	"testDefinition.runDetail.totalCases": "Total cases",
+	"testDefinition.runDetail.error.mockedNodeMissing": "Output for a mocked node does not exist in benchmark execution. <br /><a href=\"{url}\">Fix test configuration</a>.",
+	"testDefinition.runDetail.error.executionFailed": "Failed to execute workflow with benchmark trigger. <br /><a href=\"{url}\" target=\"_blank\">View execution</a>.",
+	"testDefinition.runDetail.error.evaluationFailed": "Failed to execute the evaluation workflow. <br /><a href=\"{url}\" target=\"_blank\">View evaluation execution</a>.",
+	"testDefinition.runDetail.error.triggerNoLongerExists": "Trigger in benchmark execution no longer exists in workflow. <br /><a href=\"{url}\" target=\"_blank\">View benchmark</a>.",
+	"testDefinition.runDetail.error.metricsMissing": "Metrics defined in test were not returned by evaluation workflow. <br /><a href=\"{url}\">Fix test configuration</a>.",
+	"testDefinition.runDetail.error.unknownMetrics": "Evaluation workflow defined metrics that are not defined in the test. <br /><a href=\"{url}\">Fix test configuration</a>.",
+	"testDefinition.runDetail.error.invalidMetrics": "Evaluation workflow returned invalid metrics. Only numeric values are expected. View evaluation execution. <br /><a href=\"{url}\" target=\"_blank\">View evaluation execution</a>.",
 	"testDefinition.runTest": "Run Test",
 	"testDefinition.cancelTestRun": "Cancel Test Run",
 	"testDefinition.notImplemented": "This feature is not implemented yet!",

packages/editor-ui/src/views/TestDefinition/TestDefinitionRunDetailView.vue

@@ -28,6 +28,65 @@ const filteredTestCases = computed(() => {
 	return testCases.value;
 });
 
+const getErrorTooltipLinkRoute = (row: TestCaseExecutionRecord) => {
+	if (row.errorCode === 'FAILED_TO_EXECUTE_EVALUATION_WORKFLOW') {
+		return {
+			name: VIEWS.EXECUTION_PREVIEW,
+			params: {
+				name: test.value?.evaluationWorkflowId,
+				executionId: row.evaluationExecutionId,
+			},
+		};
+	} else if (row.errorCode === 'MOCKED_NODE_DOES_NOT_EXIST') {
+		return {
+			name: VIEWS.TEST_DEFINITION_EDIT,
+			params: {
+				testId: testId.value,
+			},
+		};
+	} else if (row.errorCode === 'FAILED_TO_EXECUTE_WORKFLOW') {
+		return {
+			name: VIEWS.EXECUTION_PREVIEW,
+			params: {
+				name: test.value?.workflowId,
+				executionId: row.executionId,
+			},
+		};
+	} else if (row.errorCode === 'TRIGGER_NO_LONGER_EXISTS') {
+		return {
+			name: VIEWS.EXECUTION_PREVIEW,
+			params: {
+				name: test.value?.workflowId,
+				executionId: row.pastExecutionId,
+			},
+		};
+	} else if (row.errorCode === 'METRICS_MISSING') {
+		return {
+			name: VIEWS.TEST_DEFINITION_EDIT,
+			params: {
+				testId: testId.value,
+			},
+		};
+	} else if (row.errorCode === 'UNKNOWN_METRICS') {
+		return {
+			name: VIEWS.TEST_DEFINITION_EDIT,
+			params: {
+				testId: testId.value,
+			},
+		};
+	} else if (row.errorCode === 'INVALID_METRICS') {
+		return {
+			name: VIEWS.EXECUTION_PREVIEW,
+			params: {
+				name: test.value?.evaluationWorkflowId,
+				executionId: row.evaluationExecutionId,
+			},
+		};
+	}
+
+	return undefined;
+};
+
 const columns = computed(
 	(): Array<TestTableColumn<TestCaseExecutionRecord>> => [
 		{
@@ -60,6 +119,7 @@ const columns = computed(
 				{ text: locale.baseText('testDefinition.listRuns.status.success'), value: 'success' },
 				{ text: locale.baseText('testDefinition.listRuns.status.error'), value: 'error' },
 			],
+			errorRoute: getErrorTooltipLinkRoute,
 			filterMethod: (value: string, row: TestCaseExecutionRecord) => row.status === value,
 		},
 		...Object.keys(run.value?.metrics ?? {}).map((metric) => ({