[Alerting] Handle when an Alerting Task fails due to its Alert object being deleted mid flight (#63093) (#63564)

Detects if a task run failed due to the task SO being deleted mid flight and if so writes debug logs instead of warnings.

Detects if an Alerting task run failed due to the alert SO being deleted mid flight of the task and if so ensures the task doesn't reschedule itself (as it usually would with other types of tasks).

Ensures that the operation of deleting or disabling an Alert won't fail if it fails to delete an already deleted task (a task might preemptively self delete if its underlying alert object was deleted, even if the overall delete operation wasn't deleted).

Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>

Author: gmmorris

x-pack/plugins/alerting/server/alerts_client.ts

@@ -34,6 +34,7 @@ import {
 import { EncryptedSavedObjectsPluginStart } from '../../../plugins/encrypted_saved_objects/server';
 import { TaskManagerStartContract } from '../../../plugins/task_manager/server';
 import { taskInstanceToAlertTaskInstance } from './task_runner/alert_task_instance';
+import { deleteTaskIfItExists } from './lib/delete_task_if_it_exists';
 
 type NormalizedAlertAction = Omit<AlertAction, 'actionTypeId'>;
 export type CreateAPIKeyResult =
@@ -268,7 +269,7 @@ export class AlertsClient {
     const removeResult = await this.savedObjectsClient.delete('alert', id);
 
     await Promise.all([
-      taskIdToRemove ? this.taskManager.remove(taskIdToRemove) : null,
+      taskIdToRemove ? deleteTaskIfItExists(this.taskManager, taskIdToRemove) : null,
       apiKeyToInvalidate ? this.invalidateApiKey({ apiKey: apiKeyToInvalidate }) : null,
     ]);
 
@@ -510,7 +511,9 @@ export class AlertsClient {
       );
 
       await Promise.all([
-        attributes.scheduledTaskId ? this.taskManager.remove(attributes.scheduledTaskId) : null,
+        attributes.scheduledTaskId
+          ? deleteTaskIfItExists(this.taskManager, attributes.scheduledTaskId)
+          : null,
         apiKeyToInvalidate ? this.invalidateApiKey({ apiKey: apiKeyToInvalidate }) : null,
       ]);
     }

x-pack/plugins/alerting/server/lib/delete_task_if_it_exists.test.ts

@@ -0,0 +1,44 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+
+import uuid from 'uuid';
+import { taskManagerMock } from '../../../task_manager/server/mocks';
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
+import { deleteTaskIfItExists } from './delete_task_if_it_exists';
+
+describe('deleteTaskIfItExists', () => {
+  test('removes the task by its ID', async () => {
+    const tm = taskManagerMock.createStart();
+    const id = uuid.v4();
+
+    expect(await deleteTaskIfItExists(tm, id)).toBe(undefined);
+
+    expect(tm.remove).toHaveBeenCalledWith(id);
+  });
+
+  test('handles 404 errors caused by the task not existing', async () => {
+    const tm = taskManagerMock.createStart();
+    const id = uuid.v4();
+
+    tm.remove.mockRejectedValue(SavedObjectsErrorHelpers.createGenericNotFoundError('task', id));
+
+    expect(await deleteTaskIfItExists(tm, id)).toBe(undefined);
+
+    expect(tm.remove).toHaveBeenCalledWith(id);
+  });
+
+  test('throws if any other errro is caused by task removal', async () => {
+    const tm = taskManagerMock.createStart();
+    const id = uuid.v4();
+
+    const error = SavedObjectsErrorHelpers.createInvalidVersionError(uuid.v4());
+    tm.remove.mockRejectedValue(error);
+
+    expect(deleteTaskIfItExists(tm, id)).rejects.toBe(error);
+
+    expect(tm.remove).toHaveBeenCalledWith(id);
+  });
+});

x-pack/plugins/alerting/server/lib/delete_task_if_it_exists.ts

@@ -0,0 +1,17 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+import { TaskManagerStartContract } from '../../../task_manager/server';
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
+
+export async function deleteTaskIfItExists(taskManager: TaskManagerStartContract, taskId: string) {
+  try {
+    await taskManager.remove(taskId);
+  } catch (err) {
+    if (!SavedObjectsErrorHelpers.isNotFoundError(err)) {
+      throw err;
+    }
+  }
+}

x-pack/plugins/alerting/server/lib/is_alert_not_found_error.test.ts

@@ -0,0 +1,31 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+
+import { isAlertSavedObjectNotFoundError } from './is_alert_not_found_error';
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
+import uuid from 'uuid';
+
+describe('isAlertSavedObjectNotFoundError', () => {
+  test('identifies SavedObjects Not Found errors', () => {
+    const id = uuid.v4();
+    // ensure the error created by SO parses as a string with the format we expect
+    expect(
+      `${SavedObjectsErrorHelpers.createGenericNotFoundError('alert', id)}`.includes(`alert/${id}`)
+    ).toBe(true);
+
+    const errorBySavedObjectsHelper = SavedObjectsErrorHelpers.createGenericNotFoundError(
+      'alert',
+      id
+    );
+
+    expect(isAlertSavedObjectNotFoundError(errorBySavedObjectsHelper, id)).toBe(true);
+  });
+
+  test('identifies generic errors', () => {
+    const id = uuid.v4();
+    expect(isAlertSavedObjectNotFoundError(new Error(`not found`), id)).toBe(false);
+  });
+});

x-pack/plugins/alerting/server/lib/is_alert_not_found_error.ts

@@ -0,0 +1,11 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
+
+export function isAlertSavedObjectNotFoundError(err: Error, alertId: string) {
+  return SavedObjectsErrorHelpers.isNotFoundError(err) && `${err}`.includes(alertId);
+}

x-pack/plugins/alerting/server/task_runner/task_runner.test.ts

@@ -16,6 +16,7 @@ import { PluginStartContract as ActionsPluginStart } from '../../../actions/serv
 import { actionsMock } from '../../../actions/server/mocks';
 import { eventLoggerMock } from '../../../event_log/server/event_logger.mock';
 import { IEventLogger } from '../../../event_log/server';
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
 
 const alertType = {
   id: 'test',
@@ -665,4 +666,36 @@ describe('Task Runner', () => {
       }
     `);
   });
+
+  test('avoids rescheduling a failed Alert Task Runner when it throws due to failing to fetch the alert', async () => {
+    savedObjectsClient.get.mockImplementation(() => {
+      throw SavedObjectsErrorHelpers.createGenericNotFoundError('task', '1');
+    });
+
+    const taskRunner = new TaskRunner(
+      alertType,
+      mockedTaskInstance,
+      taskRunnerFactoryInitializerParams
+    );
+
+    encryptedSavedObjectsPlugin.getDecryptedAsInternalUser.mockResolvedValueOnce({
+      id: '1',
+      type: 'alert',
+      attributes: {
+        apiKey: Buffer.from('123:abc').toString('base64'),
+      },
+      references: [],
+    });
+
+    const runnerResult = await taskRunner.run();
+
+    expect(runnerResult).toMatchInlineSnapshot(`
+      Object {
+        "runAt": undefined,
+        "state": Object {
+          "previousStartedAt": 1970-01-01T00:00:00.000Z,
+        },
+      }
+    `);
+  });
 });

x-pack/plugins/alerting/server/task_runner/task_runner.ts

@@ -26,12 +26,13 @@ import { taskInstanceToAlertTaskInstance } from './alert_task_instance';
 import { AlertInstances } from '../alert_instance/alert_instance';
 import { EVENT_LOG_ACTIONS } from '../plugin';
 import { IEvent, IEventLogger } from '../../../event_log/server';
+import { isAlertSavedObjectNotFoundError } from '../lib/is_alert_not_found_error';
 
 const FALLBACK_RETRY_INTERVAL: IntervalSchedule = { interval: '5m' };
 
 interface AlertTaskRunResult {
   state: AlertTaskState;
-  runAt: Date;
+  runAt: Date | undefined;
 }
 
 interface AlertTaskInstance extends ConcreteTaskInstance {
@@ -328,22 +329,29 @@ export class TaskRunner {
           };
         },
         (err: Error) => {
-          this.logger.error(`Executing Alert "${alertId}" has resulted in Error: ${err.message}`);
+          const message = `Executing Alert "${alertId}" has resulted in Error: ${err.message}`;
+          if (isAlertSavedObjectNotFoundError(err, alertId)) {
+            this.logger.debug(message);
+          } else {
+            this.logger.error(message);
+          }
           return {
             ...originalState,
             previousStartedAt,
           };
         }
       ),
-      runAt: resolveErr<Date, Error>(runAt, () =>
-        getNextRunAt(
-          new Date(),
-          // if we fail at this point we wish to recover but don't have access to the Alert's
-          // attributes, so we'll use a default interval to prevent the underlying task from
-          // falling into a failed state
-          FALLBACK_RETRY_INTERVAL
-        )
-      ),
+      runAt: resolveErr<Date | undefined, Error>(runAt, err => {
+        return isAlertSavedObjectNotFoundError(err, alertId)
+          ? undefined
+          : getNextRunAt(
+              new Date(),
+              // if we fail at this point we wish to recover but don't have access to the Alert's
+              // attributes, so we'll use a default interval to prevent the underlying task from
+              // falling into a failed state
+              FALLBACK_RETRY_INTERVAL
+            );
+      }),
     };
   }
 }

x-pack/plugins/oss_telemetry/server/lib/tasks/visualizations/task_runner.test.ts

@@ -4,7 +4,6 @@
  * you may not use this file except in compliance with the Elastic License.
  */
 
-import moment from 'moment';
 import {
   getMockCallWithInternal,
   getMockConfig,
@@ -13,6 +12,7 @@ import {
 } from '../../../test_utils';
 import { visualizationsTaskRunner } from './task_runner';
 import { TaskInstance } from '../../../../../task_manager/server';
+import { getNextMidnight } from '../../get_next_midnight';
 
 describe('visualizationsTaskRunner', () => {
   let mockTaskInstance: TaskInstance;
@@ -41,12 +41,6 @@ describe('visualizationsTaskRunner', () => {
   });
 
   test('Summarizes visualization response data', async () => {
-    const getNextMidnight = () =>
-      moment()
-        .add(1, 'days')
-        .startOf('day')
-        .toDate();
-
     const runner = visualizationsTaskRunner(mockTaskInstance, getMockConfig(), getMockEs());
     const result = await runner();
 

x-pack/plugins/task_manager/server/lib/is_task_not_found_error.test.ts

@@ -0,0 +1,31 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+
+import { isTaskSavedObjectNotFoundError } from './is_task_not_found_error';
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
+import uuid from 'uuid';
+
+describe('isTaskSavedObjectNotFoundError', () => {
+  test('identifies SavedObjects Not Found errors', () => {
+    const id = uuid.v4();
+    // ensure the error created by SO parses as a string with the format we expect
+    expect(
+      `${SavedObjectsErrorHelpers.createGenericNotFoundError('task', id)}`.includes(`task/${id}`)
+    ).toBe(true);
+
+    const errorBySavedObjectsHelper = SavedObjectsErrorHelpers.createGenericNotFoundError(
+      'task',
+      id
+    );
+
+    expect(isTaskSavedObjectNotFoundError(errorBySavedObjectsHelper, id)).toBe(true);
+  });
+
+  test('identifies generic errors', () => {
+    const id = uuid.v4();
+    expect(isTaskSavedObjectNotFoundError(new Error(`not found`), id)).toBe(false);
+  });
+});

x-pack/plugins/task_manager/server/lib/is_task_not_found_error.ts

@@ -0,0 +1,11 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+
+import { SavedObjectsErrorHelpers } from '../../../../../src/core/server';
+
+export function isTaskSavedObjectNotFoundError(err: Error, taskId: string) {
+  return SavedObjectsErrorHelpers.isNotFoundError(err) && `${err}`.includes(taskId);
+}