Redux dispatch action taking too long despite reducer function attached executes fast - performance

I am getting chrome violation errors of type setTimeout handler took 500+ ms which are seriously blocking my app, and some happen every time an action attached to a websocket payload comes and is processed. I tried debugging it with the chrome profiler, and this is what it shows exactly at the point when payload is processed.
https://imgur.com/a/ZnS0ZlG
The (anonymous) function is the one in the reducer and the time ran coincides with the error.
Here is some code.
// ACTION
const someAction = (data): Thunk => async dispatch => {
try {
const t = performance.now();
dispatch(someAction(data));
console.log('after dispatching cellReceived', performance.now() - t);
// logs 800+ ms and is consistent with chrome violation errors (setTimeout handler took <N> ms
}
}
// REDUCER
export default(state: State, action: Actions) {
switch(action.type) {
...
case ActionType.someAction: {
const { data } = action.payload;
const t = performance.now();
(... do calculations here)
console.log(performance.now() - t) // logs 30ms
}
}
}
I would greatly appreciate any help, I must have spent over 20 hours this week reading about this issue and trying to debug it. I didn't find any good resources on how to properly debug with chrome's profiler.

It doesn't actually need to be the dispatch or reducer. Under certain circumstances, React will start a rerender synchronously, as a direct result of the dispatch - so before your console.log('after dispatching cellReceived', performance.now() - t); line
So this could also be a very slow React render.
If you want to make sure:
import { batch } from 'react-redux'
const someAction = (data): Thunk => async dispatch => {
try {
batch(() => {
const t = performance.now();
dispatch(someAction(data));
console.log('after dispatching cellReceived', performance.now() - t);
})
} catch {/*...*/}
}

Related

Expecting a Promise *not* to complete, in Jest

I have the following need to test whether something does not happen.
While testing something like that may be worth a discussion (how long wait is long enough?), I hope there would exist a better way in Jest to integrate with test timeouts. So far, I haven't found one, but let's begin with the test.
test ('User information is not distributed to a project where the user is not a member', async () => {
// Write in 'userInfo' -> should NOT turn up in project 1.
//
await collection("userInfo").doc("xyz").set({ displayName: "blah", photoURL: "https://no-such.png" });
// (firebase-jest-testing 0.0.3-beta.3)
await expect( eventually("projects/1/userInfo/xyz", o => !!o, 800 /*ms*/) ).resolves.toBeUndefined();
// ideally:
//await expect(prom).not.toComplete; // ..but with cancelling such a promise
}, 9999 /*ms*/ );
The eventually returns a Promise and I'd like to check that:
within the test's normal timeout...
such a Promise does not complete (resolve or reject)
Jest provides .resolves and .rejects but nothing that would combine the two.
Can I create the anticipated .not.toComplete using some Jest extension mechanism?
Can I create a "run just before the test would time out" (with ability to make the test pass or fail) trigger?
I think the 2. suggestion might turn handy, and can create a feature request for such, but let's see what comments this gets..
Edit: There's a further complexity in that JS Promises cannot be cancelled from outside (but they can time out, from within).
I eventually solved this with a custom matcher:
/*
* test-fns/matchers/timesOut.js
*
* Usage:
* <<
* expect(prom).timesOut(500);
* <<
*/
import { expect } from '#jest/globals'
expect.extend({
async timesOut(prom, ms) { // (Promise of any, number) => { message: () => string, pass: boolean }
// Wait for either 'prom' to complete, or a timeout.
//
const [resolved,error] = await Promise.race([ prom, timeoutMs(ms) ])
.then(x => [x])
.catch(err => [undefined,err] );
const pass = (resolved === TIMED_OUT);
return pass ? {
message: () => `expected not to time out in ${ms}ms`,
pass: true
} : {
message: () => `expected to time out in ${ms}ms, but ${ error ? `rejected with ${error}`:`resolved with ${resolved}` }`,
pass: false
}
}
})
const timeoutMs = (ms) => new Promise((resolve) => { setTimeout(resolve, ms); })
.then( _ => TIMED_OUT);
const TIMED_OUT = Symbol()
source
The good side is, this can be added to any Jest project.
The down side is, one needs to separately mention the delay (and guarantee Jest's time out does not happen before).
Makes the question's code become:
await expect( eventually("projects/1/userInfo/xyz") ).timesOut(300)
Note for Firebase users:
Jest does not exit to OS level if Firestore JS SDK client listeners are still active. You can prevent it by unsubscribing to them in afterAll - but this means keeping track of which listeners are alive and which not. The firebase-jest-testing library does this for you, under the hood. Also, this will eventually ;) get fixed by Firebase.

Async/Await to simulate busy server that delay update to client browser

MERN noob here. Trying to learn Async/Await by simulating a busy server where client browser only get the update > 3 seconds later (i will manually refresh localhost:3000, after 3 seconds. I only need help on Node.js/ server side for this question)
Could you help rectify codes below? Kindly avoid proposing other ways/methods but show me how to achieve using below example. Thanks in advance.
const app = require('express')()
async function getData() {
//Purpose: to simulate a busy server that returns data back to browser, after 3 seconds delay
await setTimeout(() => {
return 'After waiting 3 seconds, return this sentense as the required data to the browser.'
}, 3000);
}
app.get('/', async (req, res) => {
try {
const data = await getData()
await res.status(200).send(`${data}`)
} catch (err) {
await res.status(400).send(`Unable to get data. Error message, "${err}"`)
}
})
app.listen(3000)
The problem here is that setTimeout doesn't return a promise so you can't use await with it. It just executes the given function after 3 seconds. You can get what you want by wrapping it in a Promise like this:
const app = require('express')()
function getData() {
//Purpose: to simulate a busy server that returns data back to browser, after 3 seconds delay
return new Promise((resolve) => {
setTimeout(() => {
resolve('After waiting 3 seconds, return this sentense as the required data to the browser.');
}, 3000);
});
}
app.get('/', async (req, res) => {
try {
const data = await getData()
await res.status(200).send(`${data}`)
} catch (err) {
await res.status(400).send(`Unable to get data. Error message, "${err}"`)
}
})
app.listen(3008)
Note that you need a return statement to return the promise inside of getData. You didn't have a return statement originally which means the function returns undefined (or if marked as async it gives a Promise that resolves to undefined).
Here we don't need to use async/await because you're not needing to use await until in the app.get. Using async/await in getData could be added but it would be redundant.
Realize that aync/await uses Promises - it's just an easy way to work with Promises. So you can't await on anything but a Promise. Note that async really just means "this function returns a Promise and we'll wrap any result in a Promise if it isn't already a Promise". So you cannot use async/await without having a Promise at some point.
But if you really want to use async/await for some reason maybe this example would help you:
async function getData() {
//Purpose: to simulate a busy server that returns data back to browser, after 3 seconds delay
const result = await new Promise((resolve) => {
setTimeout(() => {
resolve('After waiting 3 seconds, return this sentense as the required data to the browser.');
}, 3000);
});
console.log('We are done waiting 3 seconds');
return result; // <-- this returns a Promise that resolves to the result string
}

Apollo server subscription not recognizing Async Iterable

I'm having an issue with Apollo GraphQL's subscription. When attempting to start the subscription I'm getting this in return:
"Subscription field must return Async Iterable. Received: { pubsub: { ee: [EventEmitter], subscriptions: {}, subIdCounter: 0 }, pullQueue: [], pushQueue: [], running: true, allSubscribed: null, eventsArray: [\"H-f_mUvS\"], return: [function return] }"
I have other subscriptions setup and are completely functional - so I can confirm the webserver is setup correctly.
I'm just curious if anyone else has ever ran onto this issue before.
Source code in PR diff (it's an open source project):
https://github.com/astronomer/houston-api/pull/165/files
error in playground
I don't think this is an issue specific to the PR you posted. I'd be surprised if any of the subscriptions were working as is.
Your subscribe function should return an AsyncIterable, as the error states. Since it returns a call to createPoller, createPoller should return an AsyncIterable. But here's what that function looks like:
export default function createPoller(
func,
pubsub,
interval = 5000, // Poll every 5 seconds
timeout = 3600000 // Kill after 1 hour
) {
// Gernate a random internal topic.
const topic = shortid.generate();
// Create an async iterator. This is what a subscription resolver expects to be returned.
const iterator = pubsub.asyncIterator(topic);
// Wrap the publish function on the pubsub object, pre-populating the topic.
const publish = bind(curry(pubsub.publish, 2)(topic), pubsub);
// Call the function once to get initial dataset.
func(publish);
// Then set up a timer to call the passed function. This is the poller.
const poll = setInterval(partial(func, publish), interval);
// If we are passed a timeout, kill subscription after that interval has passed.
const kill = setTimeout(iterator.return, timeout);
// Create a typical async iterator, but overwrite the return function
// and cancel the timer. The return function gets called by the apollo server
// when a subscription is cancelled.
return {
...iterator,
return: () => {
log.info(`Disconnecting subscription ${topic}`);
clearInterval(poll);
clearTimeout(kill);
return iterator.return();
}
};
}
So createPoller creates an AsyncIterable, but then creates a shallow copy of it and returns that. graphql-subscriptions uses iterall's isAsyncIterable for the check that's producing the error you're seeing. Because of the way isAsyncIterable works, a shallow copy won't fly. You can see this for yourself:
const { PubSub } = require('graphql-subscriptions')
const { isAsyncIterable } = require('iterall')
const pubSub = new PubSub()
const iterable = pubSub.asyncIterator('test')
const copy = { ...iterable }
console.log(isAsyncIterable(iterable)) // true
console.log(isAsyncIterable(copy)) // false
So, instead of returning a shallow copy, createPoller should just mutate the return method directly:
export default function createPoller(...) {
...
iterator.return = () => { ... }
return iterator
}

What is the right way to log a request time using the apollo graphql client?

I'm doing some application profiling on a react native application and I'm seeing pretty big differences between the request times that I'm getting using the apollo-link-logger (as well as the links I've rolled on my own) and the Android profiler's networking lane. For a ~600ms request in the profiler I'm seeing upwards of 2 seconds from the middle ware that uses the apollo link system.
There isn't much going on in my own link (below)
const metricsLink = new ApolloLink((operation, forward) => {
const { operationName } = operation
const startTime = new Date().getTime()
const observable = forward(operation)
observable.subscribe({
complete: () => {
const elapsed = new Date().getTime() - startTime
console.warn(`[METRICS][${operationName}] (${elapsed}) complete`)
}
})
return observable
})
It seems like its going to end up factoring in the time it takes apollo to manage this request chain as well. I've confirmed that this isn't a general issue with the app by fetching from other endpoints directly with fetch and comparing those times with the profiler times (which match).
Should I expect this to actually reflect the request time? Where might the rest of the time be going between the native network request and the time I'm seeing in the apollo client?
From the Apollo Link docs, there is an example to measure time using link composition
Note: I've modified the example to use performance.now() instead of new Date()based on this other SO answer
const timeStartLink = new ApolloLink((operation, forward) => {
operation.setContext({ start: performance.now() })
return forward(operation)
})
const logTimeLink = new ApolloLink((operation, forward) => {
return forward(operation).map(data => {
// data from a previous link
const time = performance.now() - operation.getContext().start
console.log(`operation ${operation.operationName} took ${time} to complete`)
return data
})
})
const link = timeStartLink.concat(logTimeLink)
The issue (at least in the version I was using) was that the Zen Observable (used by Apollo in these links) executes its logic every time the subscribe call is made. That meant that I got double requests being sent because I had two links and each one called subscribe and forwarded the already created observable. My workaround was to create new observables each time and hook them up to the observable that the prior link passed down.
const metricsLink = new ApolloLink((operation, forward) => {
const { operationName } = operation
const startTime = new Date().getTime()
const observable = forward(operation)
// Return a new observable so no other links can call .subscribe on the
// the one that we were passsed.
return new Observable(observer => {
observable.subscribe({
complete: () => {
const elapsed = new Date().getTime() - startTime
console.warn(`[METRICS][${operationName}] (${elapsed}) complete`)
observer.complete()
},
next: observer.next.bind(observer),
error: error => {
// ...
observer.error(error)
}
})
})
})
This seems like an oversight to me, but at least I have a workaround.

How fast / efficient is Bluebird's timeout?

The following example times out in most cases (outputs timed out):
Promise = require('bluebird');
new Promise(resolve => {
setTimeout(resolve, 1000);
})
.timeout(1001)
.then(() => {
console.log('finished');
})
.catch(error => {
if (error instanceof Promise.TimeoutError) {
console.log('timed out');
} else {
console.log('other error');
}
});
Does this mean that the Bluebird's promise overhead takes longer than 1ms?
I see it often time out even if I use .timeout(1002).
The main reason for asking - I'm trying to figure what the safe threshold is, which gets more important with smaller timeouts.
Using Bluebird 3.5.0, under Node.js 8.1.2
I have traced your bug in Bluebird's code. Consider this:
const p = new Promise(resolve => setTimeout(resolve, 1000));
const q = p.timeout(1001); // Bluebird spawns setTimeout(fn, 1001) deep inside
That looks rather innocent, yeah? Though, not in this case. Internally, Bluebird implemented it something like (not actually valid JS; timeout clearing logic is omitted):
Promise.prototype.timeout = function(ms) {
const original = this;
let result = original.then(); // Looks like noop
setTimeout(() => {
if result.isPending() {
make result rejected with TimeoutError; // Pseudocode
}
}, ms);
return result;
}
Bug was presence of line ret.isPending(). It resulted in brief time when original.isPending() === false and ret.isPending() === true because "resolved" status didn't propagate yet from original to children. Your code hit that extremely short period and BOOM, you had race condition.
I think what's going on here is that there's a race between the time the rest of the promise chain takes and the timer from the .timeout(). Since they are both so close in timing, sometimes one wins and sometimes the other wins - they are racy. When I run this code that logs the sequence of events, I get different ordering on different runs. The exact output order is unpredictable (e.g. racy).
const Promise = require('bluebird');
let buffer = [];
function log(x) {
buffer.push(x);
}
new Promise(resolve => {
setTimeout(() => {
log("hit my timeout");
resolve();
}, 1000);
}).timeout(1001).then(() => {
log('finished');
}).catch(error => {
if (error instanceof Promise.TimeoutError) {
log('timed out');
} else {
log('other error');
}
});
setTimeout(() => {
console.log(buffer.join("\n"));
}, 2000);
Sometimes this outputs:
hit my timeout
finished
And, sometimes it outputs:
hit my timeout
timed out
As has been mentioned in the comments, if .then() was always executed via microtask (which should precede any macrotasks), then one would think that the .then() would precede the setTimeout() from the .timeout(), but things are apparently not that simple.
Since the details of promise .then() scheduling is not mandated by specification (only that the stack is clear of application code) a code design should not assume a specific scheduling algorithm. Thus, a timeout this close to the execution of the async operation it is following can be racy and thus unpredictable.
If you could explain exactly what problem you're trying to solve, we could probably offer more concrete advice about what to do. No timers in Javascript are precise to the ms because JS is single threaded and all timer events have to go through the event queue and they only call their callbacks when their event gets serviced (not exactly when the timer fired). That said, timer events will always be served in order so a setTimeout(..., 1000) will always come before setTimeout(..., 1001), even though there may not be exactly 1ms delta between the executing of the two callbacks.

Resources