Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Response Ops][Task Manager] Emitting error metric when task update fails #191307

Merged
merged 3 commits into from
Aug 28, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -2061,6 +2061,97 @@ describe('TaskManagerRunner', () => {
);
expect(onTaskEvent).toHaveBeenCalledTimes(2);
});

test('emits TaskEvent when failing to update a recurring task', async () => {
const id = _.random(1, 20).toString();
const runAt = minutesFromNow(_.random(5));
const onTaskEvent = jest.fn();
const { runner, instance, store } = await readyToRunStageSetup({
onTaskEvent,
instance: {
id,
schedule: { interval: '1m' },
},
definitions: {
bar: {
title: 'Bar!',
createTaskRunner: () => ({
async run() {
return { runAt, state: {} };
},
}),
},
},
});

const error = new Error('fail');

store.update.mockImplementation(() => {
throw error;
});

await expect(runner.run()).rejects.toThrowError('fail');

expect(onTaskEvent).toHaveBeenCalledWith(
withAnyTiming(
asTaskRunEvent(
id,
asErr({
task: instance,
persistence: TaskPersistence.Recurring,
result: TaskRunResult.Failed,
isExpired: false,
error,
})
)
)
);
});

test('emits TaskEvent when failing to update a non-recurring task', async () => {
const id = _.random(1, 20).toString();
const runAt = minutesFromNow(_.random(5));
const onTaskEvent = jest.fn();
const { runner, instance, store } = await readyToRunStageSetup({
onTaskEvent,
instance: {
id,
},
definitions: {
bar: {
title: 'Bar!',
createTaskRunner: () => ({
async run() {
return { runAt, state: {} };
},
}),
},
},
});

const error = new Error('fail');

store.update.mockImplementation(() => {
throw error;
});

await expect(runner.run()).rejects.toThrowError('fail');

expect(onTaskEvent).toHaveBeenCalledWith(
withAnyTiming(
asTaskRunEvent(
id,
asErr({
task: instance,
persistence: TaskPersistence.NonRecurring,
result: TaskRunResult.Failed,
isExpired: false,
error,
})
)
)
);
});
});

test('does not update saved object if task expires', async () => {
Expand Down
64 changes: 39 additions & 25 deletions x-pack/plugins/task_manager/server/task_running/task_runner.ts
Original file line number Diff line number Diff line change
Expand Up @@ -719,40 +719,54 @@ export class TaskManagerRunner implements TaskRunner {
await eitherAsync(
result,
async ({ runAt, schedule, taskRunError }: SuccessfulRunResult) => {
const processedResult = {
task,
persistence:
schedule || task.schedule ? TaskPersistence.Recurring : TaskPersistence.NonRecurring,
result: await (runAt || schedule || task.schedule
? this.processResultForRecurringTask(result)
: this.processResultWhenDone()),
};

// Alerting task runner returns SuccessfulRunResult with taskRunError
// when the alerting task fails, so we check for this condition in order
// to emit the correct task run event for metrics collection
// taskRunError contains the "source" (TaskErrorSource) data
if (!!taskRunError) {
debugLogger.debug(`Emitting task run failed event for task ${this.taskType}`);
const taskPersistence =
schedule || task.schedule ? TaskPersistence.Recurring : TaskPersistence.NonRecurring;
try {
const processedResult = {
task,
persistence: taskPersistence,
result: await (runAt || schedule || task.schedule
? this.processResultForRecurringTask(result)
: this.processResultWhenDone()),
};

// Alerting task runner returns SuccessfulRunResult with taskRunError
// when the alerting task fails, so we check for this condition in order
// to emit the correct task run event for metrics collection
// taskRunError contains the "source" (TaskErrorSource) data
if (!!taskRunError) {
debugLogger.debug(`Emitting task run failed event for task ${this.taskType}`);
this.onTaskEvent(
asTaskRunEvent(
this.id,
asErr({ ...processedResult, isExpired: taskHasExpired, error: taskRunError }),
taskTiming
)
);
} else {
this.onTaskEvent(
asTaskRunEvent(
this.id,
asOk({ ...processedResult, isExpired: taskHasExpired }),
taskTiming
)
);
}
} catch (err) {
this.onTaskEvent(
asTaskRunEvent(
this.id,
asErr({
...processedResult,
task,
persistence: taskPersistence,
result: TaskRunResult.Failed,
isExpired: taskHasExpired,
error: taskRunError,
error: err,
}),
taskTiming
)
);
} else {
this.onTaskEvent(
asTaskRunEvent(
this.id,
asOk({ ...processedResult, isExpired: taskHasExpired }),
taskTiming
)
);
throw err;
}
},
async ({ error }: FailedRunResult) => {
Expand Down