development

[umbral zealot]

uhhhh sorry I meant discord.com
Sorry, professional deformation

[opal plank]

you can talk to eacthother

[drifting wedge]

talk in dumb people words

[modest maple]

ehhhhhhhhhhhhh you're getting massively un-reasonable for python

[drifting wedge]

like 4 letter words max

[opal plank]

those are the terms though

you know what IPC is right?

[modest maple]

using pipes like that is a really bad idea in python's async world

[drifting wedge]

yes

inter proccess communication

[opal plank]

okay, you know what process is ?

[drifting wedge]

yes

[opal plank]

okay, good

process.send() is a function that you can use to talk with IPC

it works like an emitter

let me get my example from leviathan rq

[drifting wedge]

alr

[jovial ore]

I literally have Discord open?? LOL @umbral zealot I don't get the issue.

[quiet badger]

just posted my new bot on the site

lmao

I hope it doesnt get rejected

[umbral zealot]

can you give me your bot's ID, lemme check something

[quartz kindle]

as long as it works, it wont be rejected

[quiet badger]

okk do they accept semi inapropriate bots

[modest maple]

Erwin this is Python not JS

[opal plank]

take a look at this example

wait what

[umbral zealot]

read the #rules-and-info

[modest maple]

bro

did you

[opal plank]

oh, if its python

[modest maple]

dude

[opal plank]

welllll

nvm then LMAO

[modest maple]

I told you like

[opal plank]

xD

[quartz kindle]

lmao

[opal plank]

oopsie

[quartz kindle]

erwin be blind

too much caipirinhna

[modest maple]

[opal plank]

im legally blind, i'll let you know sir

[quiet badger]

Alright I read the rule

[quartz kindle]

or maybe caipiroska

[quiet badger]

but my bot is kinda on the limit

[opal plank]

that too

drunkness causes blindness

[drifting wedge]

okay so why not @modest maple

[opal plank]

causality

[quiet badger]

can I dm someone about what my bot does and you tell me if that's not allowed

[drifting wedge]

can u EXPLAIn why not?

so basically

[quiet badger]

or show in vc

[drifting wedge]

2 options

[umbral zealot]

Rules are extremely clear, there are no grey areas

and your bot will be reviewed when it reaches the top of the queue not before

[drifting wedge]

i run a specific instance of the bot for dashboard, or i can add ipc to all the bot clusters

and then just find a way to ask the specfic cluster to do set thing

[modest maple]

Because python web-servers work via a pre-forking moddle each worker is a new stateless process which is a new ipc client mess per server instance which makes your code even more spaghetti

[quartz kindle]

how do you spawn your clusters anyway? doesnt py have a master process? like a shard manager

[modest maple]

sharding in python is still 1 thread

so you can normally support upto 15-20k servers per instance on your average server

my pc can handle about 60k servers per instance

[quiet badger]

rip

[quartz kindle]

my understanding is hes already running multiple instance

[modest maple]

ik

[quartz kindle]

at least thats what i understand as "clustering"

[modest maple]

which he shouldnt be doing

[jovial ore]

Fixed

[opal plank]

master, heheheh

[quartz kindle]

so im asking, how are you spawning these instances

[drifting wedge]

well the way im doing it is have each cluster be an instance then making the instance run specific shards

[quartz kindle]

are they spawned by a master process?

or do you run 2 different files?

[drifting wedge]

2 dif files

[quartz kindle]

ah

[modest maple]

okay

so

i REALLY gotta repeat that your shouldnt be clustering

like no offence

but this is way out of your skill level from what i can gather for now

[midnight blaze]

guys pls give me some hints how to solve this problem. I am sure I am not the first who had this, I was on google and all I could find was topics with v11 js

djs

[quartz kindle]

i mean

[modest maple]

1 instance is fine to use for the ext-ipc module

[quartz kindle]

just run an http server on each of your processes

no need for ipc

[opal plank]

you could try just switching libs

[modest maple]

Then you cant access cache from each process

[quartz kindle]

you can

[modest maple]

only the cache local to the server

which you cant control

[drifting wedge]

huh

[opal plank]

jk, lets see whats up

[quartz kindle]

exactly

[drifting wedge]

prob would work

[lusty quest]

async stuff is annoying AF. returning data from a Promise that runs inside a promise, i hate it.

[quartz kindle]

same thing with ipc

[opal plank]

@midnight blaze how good is ur internet and are you sharding?

[modest maple]

exactly

[quartz kindle]

your dashboard will need to send a message to all processes

and have the right process respond

[umbral zealot]

How many guilds do you have, and also, what's your sharding code?

[drifting wedge]

ye

[quartz kindle]

like discord.js does

[drifting wedge]

i can like find what cluster a guild is in

ye?

like it asks all the clusters

[modest maple]

not really

[quartz kindle]

you can find what shard it is in

[drifting wedge]

is that guild in your list of guilds?

[quartz kindle]

if you have the total shards and the guild id

[drifting wedge]

@quartz kindle yea

that wou;d work

[modest maple]

each cluster will only have a set amount of guilds

send to all then have only one respond

but you cant use the ipc module for this btw

[opal plank]

not really needed tho

what tim said is right

theres a formula to calculate the guild

[quartz kindle]

if you know which shard ids each process has, then you can send it only to that one

[drifting wedge]

ye?

[opal plank]

(guild id >> 22)

erm

[modest maple]

true

[opal plank]

fuck i forgot the rest

[latent heron]

programmer moment

[opal plank]

og right

[quartz kindle]

pwogwammer

[latent heron]

uWU pWoGrammR WoWment

[opal plank]

(GUILD_ID >> 22n) % SHARD_COUNT

thats the formula to see which shard your guild is in

so like tim said, yeah, you can directly fetch the shard the guild is suppose to be in

[drifting wedge]

yep

and just have the dashboard send a request to http server

[quartz kindle]

yes

[opal plank]

you would need to have a master though

[drifting wedge]

that bot cluster gets and acts on

[zenith terrace]

Pwowogwammer

[drifting wedge]

yea i wanted that

[opal plank]

or somewhere so you can fetch how many shards your bot has

[drifting wedge]

ye ye

ill just have a config.json

file

[opal plank]

oof, hardcoded shards

[drifting wedge]

@opal plank OR

i could do guildcount / 2,5k

then round that

[opal plank]

well, i guess you could just read the file and send it via http

[quartz kindle]

guildcount would still be hardcoded

[drifting wedge]

wdym?

[opal plank]

when someoena dds ur bot

[quartz kindle]

just send a request to /gateway/bot

[opal plank]

then your number is already wrong

[drifting wedge]

true

[midnight blaze]

6000, should I really share my code? It aint special

[opal plank]

oh yeah

[drifting wedge]

well then what can i do?

[opal plank]

i think you can fetch via API

./me should work no?

[drifting wedge]

how wouoldu have dynamic shards doe?

[opal plank]

or is it on the gateway?

[modest maple]

gateway iirc

[midnight blaze]

internet? It is a vultr vps?

[drifting wedge]

dont u have to specify shardcount on bot start?

[umbral zealot]

unless you've literally duplicated the exact shardingManager code, I'm pretty sure it would be helpful

[quartz kindle]

send a request to /gateway/bot with your bot's token, and you will get the recommended shard count

use that to know how many shards

[drifting wedge]

@quartz kindle ye but i still will need to CHANGE the shard count

like restart the bot

[quartz kindle]

yes, but you dont need to hard code it

[drifting wedge]

oh ye

i mean how does the api deal with it>

[quartz kindle]

every time your bot starts, it should be getting the right number of shards automatically

[opal plank]

reason why i ask is thats usually due to poor internet which makes discord send too much data and you miss ACK's and heartbeats before they become ready. Caching all users can cause that, if your bot is pretty big

[midnight blaze]

ah ok got it

[modest maple]

manual chunking™️

[quartz kindle]

manual thonking

[opal plank]

Dynamic Cluster Gang ||and ddosing twitch in the process of learning||

[midnight blaze]

require("dotenv").config();
const AutoPoster = require('topgg-autoposter')

const Discord = require('discord.js');
const client = new Discord.Client();
const { ShardingManager } = require('discord.js')




const manager = new ShardingManager('./index.js', {
    totalShards: 'auto',
    token: process.env.TOKEN
});

const poster = AutoPoster('eyJhbGci----iWt_c', manager);

manager.on('shardCreate', shard => {
    console.log(`Launched shard ${shard.id}`)
    shard.on('ready', () => {
        console.log('Shard ready')
    })
    shard.on('disconnect', (a, b) => {
        console.log('Shard disconnected')
        console.log(a)
        console.log(b)
    })
    shard.on('reconnecting', (a, b) => {
        console.log('Shard reconnecting')
        console.log(a)
        console.log(b)
    })
    shard.on('death', (a, b) => {
        console.log('Shard died')
        client.destroy()
        client.login("token")
        console.log(a)
        console.log(b)
    })
})

setTimeout(() => {
    manager.spawn()
},15000)

[drifting wedge]

lol

whats the /gateway/bot url?

like how would i do it?

i STILL dont understand http stuff lol

[opal plank]

thats not what we asked lol

[drifting wedge]

shard.murder

[opal plank]

how many members/guilds is ur bot in and how fast is ur internet/processor

[midnight blaze]

they did tho..

[opal plank]

who?

[quartz kindle]

aiohttp

discord.com/api/gateway/bot

in the headers put Authorization: Bot tokenhere

[midnight blaze]

here 😛

[opal plank]

ah evie asked for it

well, still, theres 3 questions you didnt answer :p

[drifting wedge]

@opal plank @quartz kindle @modest maple ty all so much

[midnight blaze]

6000 servers

[opal plank]

that might be why

[midnight blaze]

I have already said that

😛

[quartz kindle]

running a client in the manager oof

[umbral zealot]

I feel like maybe now all the information has been gathered, but I don't do sharding

[opal plank]

are you caching all members on startup?

@quartz kindle d.js caches everything by default no?

[quiet badger]

Do you just get the bot developer role when your bot goes on the site?

[drifting wedge]

https://discord.com/api/gateway/bot is the /bot?

[quartz kindle]

delet this

const Discord = require('discord.js');
const client = new Discord.Client();

[midnight blaze]

no

ok

[drifting wedge]

bot token?

[umbral zealot]

yes

[drifting wedge]

like or do i need to put it in the header

or whatever

[quiet badger]

cool

[quartz kindle]

yes

but not members

[opal plank]

isnt there a chance he might be missing heartbeats due to discord sending all member data?

[quartz kindle]

nah

well

actually it is

[opal plank]

i remember cake having a similar problem on his

[quartz kindle]

because djs has a flawed fetchAllMembers

[opal plank]

indeed

[quartz kindle]

i had to fix it myself on djsl

[rugged girder]

So my bot died this morning because it sends GET requests to the Top.gg API checking if a user has voted (for certain commands). In order to solve this, I need to implement better error handling so if the top.gg API isn't working, it just bypasses and allows the user to use the command anyway.

However, with the issues, it takes around 10 seconds for it to receive the 504 error from the API. Because of this, I don't think it's feasible to just wait until the response comes back... what can I do instead? I've heard that webhooks might be a superior solution, however, how can I make my bot store the data that the user has voted within the last 12 hours (as I am not rewarding users, I am just force-voting on some special commands for the bot).

[quartz kindle]

top.gg is having issues

always catch your errors

[lusty quest]

catch the error

[rugged girder]

yes but that's the issue, as the catch takes like 10 seconds to work

because it is still waiting for the response

[opal plank]

like a champ, ts topgg.postStats().catch(() => {})

[rugged girder]

[opal plank]

why is that mmulu so small?

d

thats tiny af

[quartz kindle]

wait, you're fetching votes?

yeah dont do that

use webhooks

[rugged girder]

voteParse = await got(`https://top.gg/api/bots/695011795110461520/check?userId=${message.author.id}`, { headers: { "Authorization": process.env.TOPTOKEN } });

like i've just been sending requests

this is probably a bad idea right

[quartz kindle]

thats not a good idea for many reasons yes

[lusty quest]

yea better use webhooks

[earnest phoenix]

implement a cooldown system, this will not only prevent your users from spamming the commands but it will also prevent reexecution of the same command while the first one is still executing

[opal plank]

SUPER

[latent heron]

webhooks are usually more opt for this @rugged girder

[opal plank]

FUCKING thats ur name

i forgot yesterday

i dont remember why but someone was asking about something from you

[rugged girder]

do you all have a guide for the webhooks, as I don't think it's feasible to store the data and delete after 12 hours...?

[lusty quest]

lol im getting famous here?

[quartz kindle]

you need a database

[rugged girder]

and then delete after 12h?

[solemn latch]

most databases can delete after 12 hours fine

[lusty quest]

if you want to store temporary i can suggest use Redis

it also fires an event if the TTL is reached

so you can send a reminder with it

[quartz kindle]

technically, you dont even need it

if you already have a database which stores user info

[lusty quest]

would say its better than having every x minutes a db call

[rugged girder]

but to prevent a ton of database queries, then won't you have to cache like 1000 votes?

[quartz kindle]

you can just add a field that says lastVoted: timestamp

[opal plank]

i had to mention you yesterday for something, but i forgot your nick

and i now forgot the reason too

[lusty quest]

do i need to lurk around here more?

[midnight blaze]

that thing was there only because I tested something and forgot to delete it..well, but anyways, now it works for some reasons

[solemn latch]

caching 1000 votes wouldnt be a ton of data

[pale vessel]

yes

[solemn latch]

a few kilobytes if done properly

[opal plank]

prob not lmao, too much braincell leaks

[rugged girder]

ree

ok

[quartz kindle]

yes because your client was trying to login without sharding, so it was erroring

[leaden lake]

does anyone can help me ?
I am programming on discord.py, and the file I have to send guild count make my bot doesn't launch.
But the one important thing, is the file worked perfectly 2h ago, but not now.

It'n not working on both OS, Ubuntu LTS and Windows

[modest maple]

error?

[umbral zealot]

if you mean the top.gg api guild count, it's because there are errors right now on the system. See #site-status and #support 's last message.

[midnight blaze]

hmm..ok, I see, I guess you are right, I just dont get why it wasnt deleted completly and I missed it, thanks for your help

[lusty quest]

anyways i think i did something cool.
a Expess Tensorflow integration with automatic scaling Worker threads and load balancer.

[amber quiver]

someone have a code for deleting messages? (not bulk delete)

[umbral zealot]

no because don't delete messages individually. It's bad, it's API spam.

[amber quiver]

and somebody knows why for some reason when Im bulk deleting 14+ days old messages, the bot just stop working

[modest maple]

because you cant

[lusty quest]

bcs you cant

[amber quiver]

thx

[modest maple]

the limit is 2 weeks max

[amber quiver]

I know

[umbral zealot]

If you want to delete all messages in a channel, clone() the channel and delete the original one.

[modest maple]

and bulk deleting anything beyond that is API abuse

[umbral zealot]

it's 2 lines, can't be simpler.

[amber quiver]

but why its just stopping my bot

[umbral zealot]

Because you didn't code it right

[lusty quest]

catch the error

[amber quiver]

thx

[lusty quest]

if you use node unhandled errors will exit the process

[opal plank]

likely cuz u use node 15

[umbral zealot]

But, again, the way to do this is clone + delete. Super simple.

[amber quiver]

yeah thx for the advice

[opal plank]

which now, thankfully, crashes the process now

finally™️

[solemn latch]

pog

[opal plank]

imagine not handling your errors

imagine not handling and counting them

and logging

[solemn latch]

why did it not crash the process before

[opal plank]

cuz pre node 14 it just dont

[leaden lake]

thanks !! the problem isn't from me. oof 😂

[opal plank]

they advertised it for a long ass time

[umbral zealot]

exactly 😄

[opal plank]

that it would crash the process

[modest maple]

ultimate logging™️

[opal plank]

PFFF

pleb

[modest maple]

im docker so i not got those colours

[opal plank]

imagine not having console variables

[modest maple]

i gotta work out where df that EOF error is coming from though

[lusty quest]

welp my bot got seizures. https://i.s8s.app/Wdw1bmdd most informative error ive ever got

[modest maple]

5 different systems that can raise that message

[topaz lynx]

How do I know my bot has been reviewed

[amber quiver]

I have another question, how can I make my bot to delete the specific amount of messages that I declare to him

[modest maple]

you get DMd by luca

[latent heron]

i feel like anyone can make a custom console error logger

[modest maple]

apply a limit to the bulk delete

[opal plank]

my console pogs

[opal plank]

indeed, but not many do

[amber quiver]

ok and...

[latent heron]

you just need to have a little bit of Bash knowledge

fair

[modest maple]

and what

[amber quiver]

thats it?

[modest maple]

just give the bulk delete a limit parameter

[amber quiver]

ok

thx

[topaz lynx]

I Checked IN MORNING, MY BOT HAD 50%
Now, it is getting reviewed
I had to do final try for myself
Wtf should I do now

[quartz kindle]

nothing

[modest maple]

if your bot is 50% in queue its not getting reviewed now

[quartz kindle]

just wait

[topaz lynx]

I said I checked in morning, it was 50%
Now, is getting reviewed

[modest maple]

its not

[cinder patio]

#site-status message

[modest maple]

trust me

its not

[topaz lynx]

It is showing it is in top.gg

[opal plank]

@slim heart //@ts-nocheck > //@ts-ignore

[modest maple]

again, trust us its not read what @cinder patio sent you

[slim heart]

i dont use ts-ignore even

[opal plank]

lmao

i know

[slim heart]

except for twice in rose

cuz fuckin

[opal plank]

just 300IQ and go full on

[slim heart]

discord-api-types poopoo

[quartz kindle]

or just dont use ts in the first place

:^)

[slim heart]

shut up tim

smh

[quartz kindle]

lmao

[opal plank]

frick u tim

[slim heart]

[modest maple]

But the alternative is JS

just use NIM

[opal plank]

scratch is clearly the superior

[lusty quest]

ts doesnt offer any performance improvement over js iirc

[slim heart]

its all fun and games til discord.js moves to ts tim

[modest maple]

again, read the message in #site-status

[quartz kindle]

i'll stop working on djsl when they do

[opal plank]

then tim will come to detritus

[modest maple]

couldnt possibly become anyworse tbh

[quartz kindle]

or just go raw

[slim heart]

i mean

[opal plank]

thats what she said

[topaz lynx]

So, we can't see percentage?
Bruh

[slim heart]

djs is adding customizable caching soontm

[quartz kindle]

soon™️

[modest maple]

read the message

[topaz lynx]

Ye

[slim heart]

they've already added methods to stores that can do api requests without cache

[solemn latch]

the percentage is just an estimation anyway

it doesnt really mean much

[opal plank]

or just use an alraedy good library that doesnt need to wait until v13 to have BASIC FUCKING FEATURES like caching customization.

cough cough detritus cough cough

[modest maple]

true

discord.py has had all that since 2017

[slim heart]

oK bUt ImAgInE fUnNeLlInG

poopy lib

[opal plank]

they already explained to you

its 8 per thread

[topaz lynx]

@modest maple can u suggest me a good bot except dank memer, mee6, dyno, carl

That has games in it

[modest maple]

🤨

[topaz lynx]

?

[opal plank]

my bot shameless ads

jk

[modest maple]

[opal plank]

this channel really isnt the palce to ask tbh

[topaz lynx]

This is development

[opal plank]

yeah, BOT DEVELOPMENT, not ask-a-bot

if you wanna develop a bot, feel free to ask about it

[topaz lynx]

How do I do YouTube notifs

[modest maple]

webhooks™️

[opal plank]

use youtube-api or scrape

or that

i think htey have webhooks

[amber quiver]

you can also doing it on websites

[opal plank]

surely youtube api cant be that bad

[topaz lynx]

I use bot designer

[amber quiver]

with fetch()

[modest maple]

yikes

[lusty quest]

have fun with the Google API, this thing is not really user friendly

[modest maple]

okay well gl with that

[opal plank]

yeah bot makers and stuff wont take you far at all

[modest maple]

Google api's are ass

for such a massive company ive never known docs to such so much

[quartz kindle]

in master or next?

[topaz lynx]

But, bot designer has 50m downloads

[slim heart]

master

[quartz kindle]

oh cool

[slim heart]

or maybe already in current version

[opal plank]

do you have the SMALLEST HINT of idea how much pain i went through to get my mail webhooks going? its SO FUCKING SHIT, you need to add pages and get verified and shit its HORRIBLE

[modest maple]

And that makes a difference because?

[lusty quest]

bot designer is making discord bots on training wheels while driving on Instapac

[topaz lynx]

Because?

[quartz kindle]

wait... where?

[opal plank]

googles api is absolutely nightmareish

either im too dumb or the docs are shit

[quartz kindle]

im looking at master commits and dont see anything

[slim heart]

oh nvrm

[lusty quest]

they are shit, had the same issue

[slim heart]

i woulda sworn i'd already seen it

[quartz kindle]

well you can do it through client.api

but thats always been there

[opal plank]

tim

bot still at 900MB

no more 4GB with d.js

[modest maple]

gasm

[opal plank]

legit tho

its idling at 900

[slim heart]

@quartz kindle https://github.com/discordjs/discord.js/blob/master/src/managers/MessageManager.js#L121

not nearly as far along as i'd thought but they added a few things

[modest maple]

im at 1.1GB atm

Just enough growing room

[opal plank]

how many guilds/members cached?

[modest maple]

10.5k Guilds, dont have members

cuz intents™️

[opal plank]

im caching everything currently, but it'll change soon

gonna make it dynamically cache it

[slim heart]

does detritus have custom prop stuff

like only caching certain props

[opal plank]

not sure

[slim heart]

should suggest

[opal plank]

you'd have to ask them

[slim heart]

ive done that for rose

[opal plank]

https://discord.gg/a89v9vSX

i sent you multiple times

just go there and suggest :p

[slim heart]

like here's an example for a shiro bot i made for rose https://github.com/jpbberry/shiro/blob/master/src/master.js#L12

it only caches channels and guilds, no props but id for guilds and channels just store nsfw props

its minute but it just add that little bit back

minute?

[opal plank]

i think it has BaseStructureData

[slim heart]

minoot?

idfk

[cinder patio]

That's a little overkill imo

[slim heart]

can never be too overkill when it comes to scaling

its overkill for tiny bots but when they get bigger it adds up

i did something similar with my custom lib for cb, and it dropped like 100MB off of ram

[modest maple]

Ram is pretty cheap though

wel..

[slim heart]

still

[modest maple]

incredibly cheap for most things

[slim heart]

i've no lack of ram

but regardless less resource usage is always poggers

[quartz kindle]

yeah i guess

[modest maple]

I gotta work with webRTC soon gonna not be too fun

im determined to learn it though

[quartz kindle]

https://detritusjs.com/interfaces/shardclientcacheoptions

[opal plank]

yeah i use that

[slim heart]

those choose what is to be cached

im talkin bout what props to be cached

within those individual caches

[opal plank]

you can select what you WANT and doesnt to be cached

[quartz kindle]

ah i guess not even detritus does that

[quartz kindle]

i had that in mind for quite a while

[opal plank]

this is the options per

[slim heart]

yeah that's not indiv props

[quartz kindle]

im gonna do that in my future testing stuff

[slim heart]

https://github.com/jpbberry/censorbot/blob/master/src/discord/events/CHANNEL_CREATE.js#L1 was inspired by my own implementation here

[onyx hare]

Error: 504 Gateway Time-out i got this on one of my bots, did i hit a rate limit?

[slim heart]

#site-status

[opal plank]

types for props sounds like a massive pain in the ass tbh

everything would have to be optional

[slim heart]

i mean look what i did

it's not that difficult to implement on rose at least

[opal plank]

for the user it still seems like a bitch to use

cuz everything would be optional

so it would highly increase complexity

[slim heart]

not really

[opal plank]

having to check if the data has the props you need

[slim heart]

you just pick out the props you actually use

[opal plank]

well yeah, but your compiler doesnt know that

[slim heart]

you rarely use more than 10% of props of any cache lmao

[analog tinsel]

random.js

index.js

how can i send (t) on index

im trying to get data from another js file

[lusty quest]

dont use var, var is deprecated and still there to not break the internet. Use Const or let instead

[opal plank]

^^

[cinder patio]

Why not just allow users handle all the caching themselves

[slim heart]

i mean you can

you can just do cache: false and it disables everything lol

[analog tinsel]

did you guys talking about me?

[cinder patio]

see but then why try to implement built-in caching. If I were to make a discord lib I would make it so users can easily build their own caching system that feels integrated into the lib, but in reality it's not

[slim heart]

its up to the user

you can customize built in caching and if u wanna do it urself just do it urself

[lucid prawn]

what this mean

(node:25) MaxListenersExceededWarning: Possible EventEmitter memory leak detected. 11 Hey @PokéHunt#0608, sunucuda link paylaşamazsın! listeners added to [CommandoClient]. Use emitter.setMaxListeners() to increase limit
(node:25) MaxListenersExceededWarning: Possible EventEmitter memory leak detected. 11 The bot is offline / restarting, please join our support server for updates: https://discord/cpYkJzd listeners added to [CommandoClient]. Use emitter.setMaxListeners() to increase limit
(node:25) MaxListenersExceededWarning: Possible EventEmitter memory leak detected. 11 **No i delete your message you can not share links here , @PokéHunt#0608** listeners added to [CommandoClient]. Use emitter.setMaxListeners() to increase limit

[solemn latch]

you have too many listeners

[slim heart]

Use emitter.setMaxListeners() to increase limit

tells you there

[solemn latch]

thats a bad solution

[lusty quest]

how many event listener did you have? like client.on(message)?

[slim heart]

not if its intended

[lucid prawn]

who is this

@PokéHunt#0608

[cinder patio]

You rarely need more than 1 event listener

[lucid prawn]

i have 6

[analog tinsel]

How can I pull the value(const/let) from another javascript file?

[slim heart]

especially with djs' message collector that's just not the case

[lucid prawn]

13

or 6

[lusty quest]

wtf?

[cinder patio]

You don't know?

[lusty quest]

how many client.on(message) etc did you got?

[lucid prawn]

yea 13

[lusty quest]

wtf?

why?

[lucid prawn]

afk and status and whitelist and my commands plus dm anti

and ect

[lusty quest]

but why?

[solemn latch]

you dont need a listener for each command

[lusty quest]

1 event or 2 would do it

[lucid prawn]

i dont

[lusty quest]

did you have duplicated event listeners like for ready, etc

[lucid prawn]

only 2

[lusty quest]

you have to got more, otherwise you wont get this error

[lucid prawn]

i will check brb

[lucid prawn]

client.on("guildCreate", guild => {
client.on('messageDelete', async message => {
client.on('guildMemberRemove', async member => {
client.on('guildBanAdd', async (guild, user) => {
client.on('message', ({channel, content, member}) => {
client.on('message', message => {
client.on(message, (message) => {
client.on('message', message => {
client.on('message', message => {
client.on('message', message => {
client.on('ready', () => {
client.on('message', (message) => {
client.on('ready', () => {
client.on("message", async message => {

[lusty quest]

why 8 message event listeners?

[modest maple]

[lusty quest]

you only need 1

also you only need 1 ready event

[mellow kelp]

why 2 ready events

[lusty quest]

if you want to have the files cleaner, split the stuff into sub files and call them inside the event

[lucid prawn]

client.on('ready', () => {
console.log('Online');
client.user.setActivity('akhelp and watching 15 servers', { type: "Watching" })
.then(presence => console.log(Activity set to ${presence.activities[0].name}))
.catch(console.error);
client.user.setStatus('dnd')
.then(console.log)
client.user.setPresence({ activity: { name: akhelp || Watching ${client.guilds.cache.size} Servers and helping ${client.users.cache.size} Senpais || Watching Commands 81 || }, status: 'dnd' })
.then(console.log)
.catch(console.error);
})

client.on('ready', () => {
const antispam = require('better-discord-antispam');
antispam(client, {
limitUntilWarn: 3,
limitUntilMuted: 5,
interval: 2000,
warningMessage: "if you don't stop from spamming, I'm going to punish you!",
muteMessage: "was muted since we don't like too much advertisement type people!",
maxDuplicatesWarning: 7,
maxDuplicatesMute: 10,
ignoredRoles: [],
ignoredMembers: [],
mutedRole: "Muted",
timeMuted: 1000 * 600,
logChannel: "antispam-logs"
});
});

[lusty quest]

out them into one ready event

[lucid prawn]

ok

[lusty quest]

do the same with the message event, then your error will also go away

[lucid prawn]

i hope my code dont broke

when i do this

[lusty quest]

it will probably break, but this is the price for bad code

[lucid prawn]

bruh

it not bad

just not clean

[lusty quest]

you have 8 message event listeners, this is bad code

[lucid prawn]

it for differener stuff

[lusty quest]

yes but wont change the fact that its bad

[brittle fable]

Is anyone else having an issue with the webhook voting thingy?

[drifting wedge]

@placid iron can u help me setup a aiohttp webserver?

[brittle fable]

I'm getting a 500 error

[opal plank]

read #support

[brittle fable]

Oh that's what causing it

alr

[slender thistle]

I heard aiohttp

[drifting wedge]

yep

@slender thistle not a mod anymore

WOT

[pale vessel]

I heard requests

[slender thistle]

no flazepe shush

[restive furnace]

I heard http

[vestal rampart]

I need help

[drifting wedge]

basically

so

i need 2 things

1. a aiohttp server for dashboard to talk to bot

[opal plank]

@slender thistle no more mod? wot

[vestal rampart]

my bot is in a terrible server that I dont want it to be in due to what the server is about and they wont kick the bot, how would I make a command that would leave the server?

[modest maple]

👀

[drifting wedge]

and 2. a way to find what proccess/shard a guildID is in

[opal plank]

welcome to the blu chad squad shiv

[restive furnace]

guild.leave() should work on lamost every lib

[opal plank]

in which library?

[vestal rampart]

Discord.py 3.8.7

[drifting wedge]

ah

so

[opal plank]

SHIIIIIIIIIIIIIIIIIIIIIIIIIIIV

we got a py here

[modest maple]

get the guild

await guild.leave()

[drifting wedge]

await guildOBJECT.leave()

@opal plank i do py

[restive furnace]

i still bet its guild.leave()

[slender thistle]

get_guild, .leave()

[restive furnace]

yes

[drifting wedge]

dummy

[restive furnace]

i was righhtttt

[drifting wedge]

😦

[opal plank]

but i like pinging @slender thistle for no reason whatsoever

[drifting wedge]

saMEE

#pingshivclub

[vestal rampart]

I am so confused, can someone please dm me if you know how to do it ❤️

[opal plank]

both of them already told you how to use it

[slender thistle]

Use a cog for your routes ig

[vestal rampart]

But where do I put that? in the console? or like how do I put that into a command

[drifting wedge]

@slender thistle well yea

[slender thistle]

I'm sure you could figure that part out on your own

[drifting wedge]

i need a way to send it to the speicific cluster

[slender thistle]

As for the shard...

[modest maple]

gotta code it yourself

[drifting wedge]

well yea

[modest maple]

WS or just REST requests

[slender thistle]

yeah chill you can take over cuz I don't know shit there

[vestal rampart]

I am reaching out for help for a reason, because I couldnt figure it out

[modest maple]

plz no

[slender thistle]

LMFAO

[drifting wedge]

@modest maple i could set up a local server doe?

just localhost?

since will be on same server

dashboard and bot

[vestal rampart]

This is what I tried to do but it didnt work

    @commands.command()
    async def leaffs(self, ctx, guild_id: in ):
        guild_id = message
        guild = await self.client.get_guild(int(guild_id))
        channel = guild.system_channel
        await channel.send("Leaving this server due to misuse!")
        await guild.leave()```

[modest maple]

👀 You know local host is local right, so if its local what's the point of your dashboard

[drifting wedge]

We literally just told you

[drifting wedge]

the local thing is for dashboard to talk to bot

like dashboard backend

[slender thistle]

just do self, ctx, guild_id: int in your command arguments

[modest maple]

no get_x isnt a coroutine

[slender thistle]

can you do guild: discord.Guild?

[lucid prawn]

done

[placid iron]

@slender thistle yes

[lucid prawn]

and it didnt break

[slender thistle]

Even better

[modest maple]

only in master

or might still be a pr

[placid iron]

😩

[slender thistle]

Well, my previous option then

[slender thistle]

for this you'd then simply do self.client.get_guild(guild_id).leave()

[opal plank]

i dont think i would survive py

see those fancy ass names yall got?

self

ctx

[slender thistle]

[opal plank]

all that would be thrown in the garbage in my code lmao

client, payload, guildId

cuz fuck the rest

[drifting wedge]

@modest maple instead of setting up an aiohttp server, could i just set up the ipc on the clusters?

[modest maple]

Test shit out

[drifting wedge]

mhm

[modest maple]

this links back to the whole dont cluster stuff when you dont need to because it makes your shit 10000% more complicated

[drifting wedge]

@modest maple alr

can you explain why i cant just set up a instance just for this?

like ik i shouldnt

but im not sure WHY

[modest maple]

i mean you can

but then you login every time you make a worker

you eat up a large amount of ram per worker

you connect to the websocket every time

extra bandwidth used from events etc...

[drifting wedge]

wdym pet worker?

per*

[modest maple]

your webserver should run 2 * your_servers_core_count

e.g. 8 cores = 16 workers

so thats 16 bot instances

[drifting wedge]

the webserver is running 16 webservers?

wot

[modest maple]

no the webserver spawns child processes

[drifting wedge]

oh

alr so thats not a good idea

[modest maple]

typically gunicorn etc... spawn 2 * the core count

[drifting wedge]

mhm

so i could setup ipc for the clusters

[modest maple]

hence why i said... its a bad idea

[drifting wedge]

or a webserver for the clusters

like an aiohttp

[quartz kindle]

you guys wanna see how much js objects are bloated?

[lyric mountain]

I JUST F****N IMPLEMENTED HOTSWAP HELL YEAH

[drifting wedge]

ig so

[near plover]

js is kinda bloated in a lot of areas tho

[misty sigil]

yes

[opal plank]

hence why u use Typescript

and dont you dare tell me ts is js

shh

stop writing

[misty sigil]

what if I don’t wanna use ts

no

[opal plank]

then frick u

[misty sigil]

where n when

[quartz kindle]

enjoy

[misty sigil]

thanks tim

did enjoy

[opal plank]

imagine not converting to human redable

[summer torrent]

so c++ best

[opal plank]

shame on u tim

[quartz kindle]

who cares about human readable

[opal plank]

frick u

[quartz kindle]

we are machines

[misty sigil]

beep boop

[opal plank]

ha, pass this captcha then

[misty sigil]

I cannot.

[quartz kindle]

btw you see that FTSet there? thats my lib that works on strings only. which means, if you convert the entire data to string, it uses 3x less ram

[summer torrent]

tim can convert bytes to human readable

[misty sigil]

so

tim pog basically

[summer torrent]

link please

[flat pelican]

Okay, I think it's the first time I need help in #development

[misty sigil]

nohello smh

[quartz kindle]

npm ftset

[modest maple]

dont ask to ask

🤣

[solemn latch]

welcome eri

[misty sigil]

it’s more a nohello tbf

[flat pelican]

My bot is going crazy rn, it has 20 shards and it's currently restarting again and again and I can't figure out why https://i.imgur.com/EUbRMyF.png

it returns this weird websocket error

but nothing else

I'm using d.py

[modest maple]

what error

also what guild count

[flat pelican]

21k

[misty sigil]

weebsocket is closin

[modest maple]

on one instance?

[flat pelican]

yeah, I mean 20k guilds in a single instance should be fine tho as long as you use shard it correctly?

[modest maple]

erm well

not really lol

[misty sigil]

twinty shards

[flat pelican]

I'm using an autosharded client rn

[modest maple]

do you have members / presence

[flat pelican]

only have members

[modest maple]

whats your server specs?

[flat pelican]

I have members chunking disabled

2cores and 3GB RAM

the bot is using 1,2GB

[modest maple]

do you have the logging Info at hand?

[near plover]

anyone happen to have a suggestion for an ORM for java?

[modest maple]

in-all likelyhood your at the max you can support on a single instance

[opal plank]

@flat pelican can you eval stuff?

[flat pelican]

Where am I supposed to get the logging info, except my console?

[opal plank]

check os cpu and threads if they are getting clogged

[modest maple]

logging.basiconfig(level=logging.INFO) put that at the top of your main file

[flat pelican]

eval seems to be working for 1/5 of times

[opal plank]

does the vps have a montiro for that?

[modest maple]

can you get your cpu usage btw

[opal plank]

im under the suspicion it might be clogging some threads and disconnecting

[eternal osprey]

hey so i have actually put an image in top.gg as background

[modest maple]

20k is pretty high for a single instance to support

[opal plank]

css

[eternal osprey]

however, it is not showing the image when i submit it? Only works in preview

[opal plank]

yeah dont trust preview

its borked

[eternal osprey]

excuse me?

[opal plank]

change the style with console

yeah preview is borked

broken

not working

[eternal osprey]

ah okay.

<style>
  body {
  background-image: url(https://wallpaperaccess.com/full/1204217.jpg);
  background-color:#FF0000;
}
</style>
```this is what i've done. Doesn't seem to work.

But it worked before.

So i've got no clue what's up.

[opal plank]

im fairly ceertain you need to tag a specific class

.class {css}

[modest maple]

body should work

[opal plank]

though i could be VERY mistaken

[modest maple]

but you probably need to force override it with !important

[flat pelican]

According to netdata when I attempt to start the bot and htop: https://i.imgur.com/fsYSfaW.png https://i.imgur.com/1XvRPMP.png

[eternal osprey]

body should work fine. I did it before.

[opal plank]

dunno then

[eternal osprey]

It just stopped working after a few months.

[opal plank]

so i assume its not cpus

[modest maple]

looks pretty reasonable that you're just maxing out at the most your server can support per instance guilds wise

just looking at those spikes

[solemn latch]

those spikes look like less than 60%

[modest maple]

yeah you'd expect that

[eternal osprey]

is it maybe because i am forcing dark page?

[modest maple]

you shouldnt see a python process using 100% ever because it cant realistically

[flat pelican]

60-66% for the spikes

[solemn latch]

oh its python 👀

[flat pelican]

yeah

[tranquil geyser]

[opal plank]

thats why i havent touched anything further, i can only assist with stuff outside of coding language here

[modest maple]

can you enable logging output quickly to double check

[flat pelican]

Sure sure

give me a sec

[flat pelican]

https://i.imgur.com/8Tm6Mhj.png

so shards are connecting

[modest maple]

so all those WS error happen before any shards identify?

or is it docker just repeating logs

[flat pelican]

I don't use docker

but https://i.imgur.com/Vhh5LHC.png

[modest maple]

ah thats what im looking for

[flat pelican]

Wait

[modest maple]

i mean this is concerning but shouldnt be blocking that much

[flat pelican]

is it really blocking because requests fails to post stats?

[modest maple]

it shouldnt be for that long

it'll have a affect but it shouldnt be causing those timeouts

[flat pelican]

Lemme try to remove it, just in case

[modest maple]

mhmm

[placid iron]

why are you using requests

[drifting wedge]

termius ftw

@flat pelican do you have termius premium?

u get it with github edu

[flat pelican]

why not

[flat pelican]

ye, got it with ghub edu

[modest maple]

i mean how often is that called?

[drifting wedge]

nice

@flat pelican u should be using aiohttp

requests is sync

[flat pelican]

@modest maple every 10 minutes

[modest maple]

ehhh i wouldnt really care that much then

[drifting wedge]

@flat pelican still should use async

sync stuff can really fuck stuff up

[modest maple]

litterally doesnt matter that much

[drifting wedge]

i mean theyres no reason not to

[modest maple]

i doubt that request is take several seconds to send a request

[drifting wedge]

@modest maple true

@flat pelican happens to me a lot

also

"websockets are closing"

[placid iron]

with aiohttp you can use a session which is poggers

[modest maple]

thats concerning considering your bot is small

[drifting wedge]

its literally just a waiting thing

[modest maple]

requests have that aswell

[modest maple]

erm, it really isnt

[drifting wedge]

for me it was

:/

[flat pelican]

shards are waking up, let's see

[drifting wedge]

then what is it?

[modest maple]

Then your code is causing some really serious issues

[drifting wedge]

if u know what it is u can help me lol

[modest maple]

1. your internet is so bad it doesnt let the bot chunk in time blocking heartbeats
2. your server is so week that your single instance cant cope with the influx of events
3. your code is blocking in some way causing hearbeats to be skipped

which reminds me

[flat pelican]

I got my on_ready event

[modest maple]

@flat pelican do you chunk manually or default?

[flat pelican]

manually, yeeted the default because it was literally taking 100 MINUTES to chunk everything

[modest maple]

yeah do be like that

[tardy hornet]

case 'lb':
  var money = db.fetch(`money_${message.guild.id}`)
  let content8 = "";

  for (let i = 0; i < money.length; i++) {
    let user8 = bot.users.get(money[i].ID.split('_')[2]).username

    content8 += `${i+1}. ${user8} ~ ${money[i].data}$\n`
  }
    const leaderboardembed = new Discord.MessageEmbed()
    .setAuthor(`${message.guild.name} - Leaderboard!`, message.guild.iconURL)
    .setDescription(content8)
    .setColor(bsc)

  
message.channel.send(leaderboardembed)
break;

it should work as a server leaderboard, but it send an error that its 2000+ in length

[modest maple]

Overall looking at this i would highly advise you start clustering

how you managed to get to 20k without this before amazes me tbh

[drifting wedge]

@modest maple my on_message event is really fucking up my bot

like its causing it to never get "ready"

[flat pelican]

I don't think clustering is needed on 20k tbh

[modest maple]

That doesnt make any sense unless you block the event loop for periods of time

[modest maple]

20k is pretty high for most servers

[flat pelican]

Uhm

I think requests was causing that

[modest maple]

bearing in mind your only a single instance and most servers have a rather low clock speed

presence accounts for alot of load but still pretty high

[flat pelican]

wtf https://i.imgur.com/10qJeMp.png

What's the ratelimit for webhooks?

[modest maple]

standard message ratelimit

[flat pelican]

like, I just use it to post and log error messages

when a command errors out

[modest maple]

you got multiple webhooks on that channel?

[flat pelican]

only one

it just posts an embed with a small traceback

i'Ll check that later I guess

[modest maple]

how often is it doing it though

[flat pelican]

when a command errors out... so it's not supposed to happen every seconds lol

i'd say 1 time every minute, I get some errors for unchunked guilds

I'll check that later

[modest maple]

you sure you're not hitting global then with all the shards etc... if this is on startup?

[flat pelican]

but btw, the root cause of this whole shitstorm was requests

[flat pelican]

for the requests issue? or webhooks?

[vernal moth]

how do I make my bot page have server count?

[quartz kindle]

https://docs.top.gg

[vernal moth]

oh

never used Top.gg API before

so i need to npm i something?

[misty sigil]

no

[modest maple]

webhooks

[flat pelican]

probably yeah

[modest maple]

logins are normally really quite heavy on your ip ratelimit

[vernal moth]

hmm

i don't see token anywhere on my profile ;-;

[modest maple]

bot -> edit -> webhooks

[flat pelican]

I think it was hitting the ratelimits today because of this issue, it probably attempted to post too much stuff in there

[vernal moth]

tyyy

now i gotta update

[flat pelican]

Btw, thanks @modest maple and @opal plank for the help ❤️

[modest maple]

👍

[tardy hornet]

can someone who understand quick.db help me please?

[vernal moth]

so I put autoposter in my code

uh oh

is this my problem or the server is down?

[pale vessel]

just wait

[vernal moth]

okie

[tardy hornet]

?

[vernal moth]

eyyyy it worked ty :3

[tardy hornet]

case 'lb':
  var money = db.fetch(`money_${message.guild.id}`)
  let content8 = "";

  for (let i = 0; i < money.length; i++) {
    let user8 = bot.users.get(money[i].ID.split('_')[2]).username

    content8 += `${i+1}. ${user8} ~ ${money[i].data}$\n`
  }
    const leaderboardembed = new Discord.MessageEmbed()
    .setAuthor(`${message.guild.name} - Leaderboard!`, message.guild.iconURL)
    .setDescription(content8)
    .setColor(bsc)

  
message.channel.send(leaderboardembed)
break;

it should work as a server leaderboard, but it send an error that its 2000+ in length, i want it to show the top 10 in the server that the command was sent.

[uncut swallow]

how?? teach me to do that because I have a mistake

[vernal moth]

oh

you use discord.js?

[uncut swallow]

yes

[vernal moth]

ok

so

[mellow kelp]

he's basically top.gg's official stackoverflow

[vernal moth]

"npm i top-gg/sdk"
"npm i topgg-autoposter"

[uncut swallow]

i need to install 2?

[vernal moth]

const ap = AutoPoster('insert token here', client)

ap.on('posted', () => {
  console.log('SYSTEM> Posted stats to top.gg')
})```

[uncut swallow]

thanks

[vernal moth]

the 2nd one is preferred

[uncut swallow]

ok

[vernal moth]

but you can install both

do you know how to get your token?

[uncut swallow]

yes

[vernal moth]

it's not the same as your bot's token

it's your top.gg token

[uncut swallow]

yes i understand

[vernal moth]

okie :3

you are welcome :))

[uncut swallow]

Can I put the code anywhere?

[vernal moth]

i put it at the beginning but you can put it anywhere except inside events

[uncut swallow]

oh thanks

[vernal moth]

like this

[uncut swallow]

ohh epic

[mellow kelp]

shouldn't you be using new though?

[glossy spoke]

ya

[vernal moth]

wat

[mellow kelp]

new AutoPoster('thing')

[glossy spoke]

add a new before that AutoPoster

[glossy spoke]

this

[sturdy dock]

would this work using internal sharding?

or would i get rid of the "servercount" since shard id and shardcount are already there

[vernal moth]

oh

you wanna put it in my on ready function?

[mellow kelp]

i think you have to use broadcastEval to get the guild count

[sturdy dock]

oh the topgg api doesn't add up the shard servers?

[mellow kelp]

not sure

[sturdy dock]

i thought each shard reported servers seperately to the api based on id

[mellow kelp]

hmm

[sturdy dock]

i'm probably wrong

[mellow kelp]

no idea, haven't used shards yet

you could use topgg-autoposter

i guess it'll handle shards correctly

[vernal moth]

i use that

[sturdy dock]

probably

but i'm using internal sharding

[mellow kelp]

o

[sturdy dock]

yeah i'll just go with that, thanks

[mellow kelp]

aight, np

[vernal moth]

so my code is wrong?

[uncut swallow]

what is shards?

[earnest phoenix]

https://discord.com/developers/docs/topics/gateway#sharding

[tardy hornet]

cry, you understand a bit of quick.db and arrays? @earnest phoenix

[earnest phoenix]

i never used quickdb

so im basically useless here

[tardy hornet]

k nvm then, ty tho

[uncut swallow]

the server count is fine in the principal page ,but why dont show ??

[cosmic phoenix]

the server count only shows on the bot's page, search/list and widget

not on the user profile

[uncut swallow]

oh thanks

for the information

why the command for the server count in 5 minutes say's : time out

only solution?

[lyric mountain]

am i supposed to be excited for this?

[uncut swallow]

TopGGAPIError [Top.GG API Error]: 504 Gateway Time-out
at Api._request (D:\DATA.C\Documents\bot_sandwich\node_modules\topgg-autoposter\node_modules@top-gg\sdk\dist\structs\Api.js:74:19)
at runMicrotasks (<anonymous>)
at processTicksAndRejections (internal/process/task_queues.js:97:5)
at async Api.postStats (D:\DATA.C\Documents\bot_sandwich\node_modules\topgg-autoposter\node_modules@top-gg\sdk\dist\structs\Api.js:94:9) {
response:

what is it?

is a error from autoposter

[mellow kelp]

regular top.gg stuff

[uncut swallow]

xd

[mellow kelp]

or maybe it's your internet connection

[uncut swallow]

uhmmm probabbly

[snow urchin]

https://i.callumdev.pw/2ga2x.png
why is this not working? trying to send an image through form-data

the error is coming from data.append

[wise umbra]

embed.attachFiles([{ attachment: canvas.toBuffer(), name: 'sample.png'}])

Not sure what's happening, but i'm not being able to set the image onto the Embed and i tried .setImage not sure whether i did it right or wrong.

[lyric mountain]

you use attachments://name.extension on embeds

then u send files with the message

[opal plank]

no preblomo

[lyric mountain]

preblomo

postblomo

[drifting wedge]

NameError: name 'client' is not defined```

when im starting ipc

and also, if i have 2 clusters, how will the ipc like choose which one to connect to?

@modest maple sorry for the ping

but any ideas?

[modest maple]

it doesnt

[drifting wedge]

like if i have 2 clusters how will i connect to both of them?

[modest maple]

you cant use it for multiple clusters like that

[drifting wedge]

so just running another instance without caching anything would be a good option?

like it wouldnt use resources

and it wouldnt respond to commands?

[modest maple]

no, it is never a good option