waterfall-help

[reef fulcrum]

you'd need a plugin

[tidal musk]

which one

[reef fulcrum]

don't mention

idk

[tidal musk]

oops sorry

https://www.spigotmc.org/resources/bungeeportals.19/ would this work?

[tidal musk]

Is there a way to have a forge server with mods and paper servers with plugins connected to waterfall?

[royal hawk]

On 1.12.2 and older yes;
On anything newer no

That limitation exists because forge can’t be re-initialized after 1.12.2 during a connection

[tidal musk]

forge bad

use fabric 😴

[tropic ice]

I think not 😦

[reef fulcrum]

your proxy failed to connect to the server

I generally refuse to help with networking too much given that it's a basic expectation here

if youre hosting everything on the same server, use the loopback or, if ptero, the ptero interface to connect to the servers

[tribal shale]

Gotta love IntelliJ giving you the smallest button possible for reloading the pom.xml

Been looking for that thing for three hours now

[marble kelp]

I mean it's in the right click menu on the pom

as well as the module root in the project view

[tribal shale]

Guess I'm either blind or stupid

oh and sorry for my discord name btw, if it wasn't allowed

[deep fox]

Can't.... my brother wants the create mod on the server and we can only do that with forge

Due to the create mod only existing for forge

[marsh reef]

cough cough
https://github.com/Fabricators-of-Create/Create

[tidal musk]

cough cough

[reef fulcrum]

vs?

[tidal musk]

cough cough i care about my worlds

[tidal musk]

when someone do /server <servername> if the specified server is full, could i auto send them to a different server?

@tidal musk yep you can remove the /server command from the modules and create your own /server command

[marsh reef]

Backups?

[deep fox]

Also if I use the fabric create mod, how do I use the .zip file?

[pastel oyster]

Any good ticket plugins out there for BungeeCord/Waterfall? I've found two but I can't know if those devs are active, user-friendly and somewhat-like SimpleHelpTickets (add5star's fork) which is Spigot/Paper but has bungeecord support (says which server the ticket originates)

[marble kelp]

my server uses its own so if you are fine with basically zero documentation and the only support being for bugs via the issue tracker... xD

[tidal musk]

lol

[tidal musk]

what

[lost river]

you should expect people to get your servers ip in some way and make firewall rules/use bungeegard to protect the servers from attacks according to the knowledge that someone has the ip.

[tidal musk]

XD ehh

i have protected server

but i report here this problem only

so that other people don't have problems

however, it cannot be avoided so revealing IP results in attacks

[marble kelp]

ddos protection + firewalls pretty much prevent most issues like that

[dull wyvern]

im struggling real hard doing color codes

idk why

like hex color codes

are just not working

[royal hawk]

Depends on where you’re trying to use them
Some more context is needed

[royal hawk]

for example MOTD isn’t using the legacy codes

[hazy wave]

Error: Could not find or load main class 0
how do I solve this? It just works well 2 days ago

[reef fulcrum]

main class 0

that means that you fucked up your java flags

[nocturne trench]

Hey, what exactly does the option game_version do?

[ancient aurora]

i dont start up waterfall 1.18... please help me(latest.log)

Waterfall →PaperMC 1.18.2x2

[reef fulcrum]

It tells you

Server lobby (priority 0) is not defined

check your priorities list and point it to a server you've actually got defined

[ancient aurora]

config file

[tulip surge]

config.yml by @ancient aurora: https://paste.gg/46d8c1dc12fc435b9e5935f039309d78

[reef fulcrum]

Look at the file

lobby is indented improeprly

and not even in the correct place

[ancient aurora]

where is wrong?

[reef fulcrum]

It's literally not even under servers.

[ancient aurora]

please teach change here?

sorry,bad english (i`m japanese)

[reef fulcrum]

I've told you whats wrong

move the lobby server definition into the servers block

if you can't do that, I can't help you

[ancient aurora]

thanks! i got waterfall start up

[desert zephyr]

Hey so when I try running a bunge, on Linux terminal, it just says Killed, how do I fix that?

Never mind, got it working, my bad

[nocturne trench]

Hey, what exactly does the option game_version do?

[reef fulcrum]

it sets the version showed in the server list ping response

[nocturne trench]

Ahhh ok. Thank you

[tidal musk]

It is necessary to set the Log4j2 fix in the Waterfall flags.

If you're on the latest build, no

Thanks

[reef fulcrum]

I mean, if that's chat, that's 100% on the client itself

only way to disable that is to not show people in tab, basically

[old acorn]

I can't do the /server command even though I have sufficient permissions

[tidal musk]

you gave yourself permissions with the bungee config or?

[old acorn]

I haven't

[tidal musk]

How did you give yourself permissions then

[old acorn]

LuckPerms & I have op

[tidal musk]

Do you have luckperms on the waterfall?

OP doesn't affect proxy at all

[old acorn]

Yeah, the bungee version

[tidal musk]

Is your waterfall lp connected to spigot lp with mysql?

[old acorn]

Yes

[tidal musk]

go to editor through waterfall - lpb editor and check if you have the permissions you need

[old acorn]

Is it normal that there's nothing synced to the proxy bungee plugin?

[tidal musk]

you must have configured luckperms wrongly

[old acorn]

Don't /lp editor and /lpb editor bring up different editors?

[tidal musk]

After you apply the changes, the permissions and groups, etc. should be all the same

[reef fulcrum]

vpx, version?

[wise patio]

Yoo, i have velocity server and waterfall on same host but if i connect to velocity everything is good (pinging takes less than second), but if i connect to waterfall pinging takes so long, idk around 5 seconds

[tidal musk]

velocity

[vague quail]

anyone know how to create custom yml in waterfall?

[craggy trout]

Thats on dev Channel i guess

[open brook]

When the server is off, it shows this...
How can I turn it off ?

[reef fulcrum]

version?

[open brook]

1.18.2

[reef fulcrum]

I mean the version of waterfall

[open brook]

It's just unpleasant when it shows the IP

480

[reef fulcrum]

I'm asking for basic info as I thought that was fixed at some point

[open brook]

ver

M

[reef fulcrum]

...

[open brook]

So I told you what version

[reef fulcrum]

Fuck off

Don't mention

[open brook]

Waterfall 1.18 480

[reef fulcrum]

I read it

I just patched it

theres no need to ping me

theres no need to spam

[open brook]

So you've already fixed it? I do not need to adjust anything?

Roger that, thank you very much.
Excuse me...

I downloaded the latest build (482), the inscription disappeared. But in the console now gives the following error

[reef fulcrum]

Well

yes

it prints the exception

[open brook]

ок

[tame sigil]

hi, i've recently switched to a newer version of waterfall and there's problems of players getting kicked when viewing big items

overflow in packet detected! A string was longer than allowed. For more information, launch Waterfall with -Dwaterfall.packet-decode-logging=true```

[civic shore]

Hello, I am currently starting to use waterfall proxy for having better compatibility with forge and skinsrestorer (previously I used velocity)

I am having a problem where all the "Vanilla" servers work normally, but with forge server(mohist #313 1.12.2) I get disconnected immediately(read timed out)

it must not be a firewall problem as I have access without bungeecord and can also access the server with "velocity

[reef fulcrum]

timed out means that the thing didn't recieve any traffic in so long

the fact that you have Via in there saying that it basically just ignored the handshake, I mean, glhf

[civic shore]

well...
that's it

well

another one

The server you were previously on went down, you have been connected to a fallback server

[reef fulcrum]

mods and proxies are generally hit and miss to the degree that I generally don't provide much support for diagnosing such issues these days especially as most stuff often boils down to tryna work out whichever mod of the potentially 200 many servers have installed is borking up the process

generally if want mods the entire network has to be running modded without any proto hacks, etc

either that or modded players have to basically be auto redirected to the correct modded server

[civic shore]

it seems that enabling "disable_entity_metadata_rewrite" in waterfall.yml solved the problem

[tawny blade]

Hm

cant join Server after Modded it

[24.03 16:04:16] [Server] [ERROR] Caused by: java.lang.IllegalArgumentException: No provider for argument minecraft:cnpcs_list
[24.03 16:04:16] [Server] [ERROR]     at com.google.common.base.Preconditions.checkArgument(Preconditions.java:122)
[24.03 16:04:16] [Server] [ERROR]     at net.md_5.bungee.protocol.packet.Commands$ArgumentRegistry.read(Commands.java:603)
[24.03 16:04:16] [Server] [ERROR]     at net.md_5.bungee.protocol.packet.Commands$ArgumentRegistry.access$000(Commands.java:308)
[24.03 16:04:16] [Server] [ERROR]     at net.md_5.bungee.protocol.packet.Commands.read(Commands.java:85)
[24.03 16:04:16] [Server] [ERROR]     at net.md_5.bungee.protocol.DefinedPacket.read(DefinedPacket.java:266)
[24.03 16:04:16] [Server] [ERROR]     at net.md_5.bungee.protocol.MinecraftDecoder.decode(MinecraftDecoder.java:60)
[24.03 16:04:16] [Server] [ERROR]     at net.md_5.bungee.protocol.MinecraftDecoder.decode(MinecraftDecoder.java:12)
[24.03 16:04:16] [Server] [ERROR]     at io.netty.handler.codec.MessageToMessageDecoder.channelRead(MessageToMessageDecoder.java:88)```

[reef fulcrum]

modern forge isn't supported by any proxy

for fabric, use velocity + cross stitch

[tawny blade]

ah

rip

[craggy trout]

Well, there is a waterfall fork that implements it

But yeah, he asked support here

[royal hawk]

Besides that; the mod using a reserved namespace will break even that special fork

[stark acorn]

Does this look good for you?
I'm not sure about the java version. 😅

Okay. Java 11 is required 😄

[reef fulcrum]

waterfall requires 8

so, wrong channel?

[stark acorn]

wrong channel? <_<

[reef fulcrum]

either way, I don't exactly recommend the alpine images, but 🤷‍♂️

[stark acorn]

I know. I'm not using those

And I think, that I'm not using waterfall either.

I may try velocity

or do you recommend waterfall?

[reef fulcrum]

I recommend velocity

[stark acorn]

website for velocity is verlocitypowered.com, right?

[reef fulcrum]

kiiinda

[stark acorn]

okay. Downloads are from papermc.io

[reef fulcrum]

much stuff is moving over to the paper org website itself

[stark acorn]

okay :/

[wide thistle]

What does t.online_mode do?

[craggy trout]

Even the name says it

[spring tulip]

Hey I'm using luckperms to give my self bungeecord permissions but I'm not able to execute bungeecord commands does someone know why?

[reef fulcrum]

/lpb

also, wasn't it like, bungee. ?

[spring tulip]

Does I need to run this command in the bungee console?

[reef fulcrum]

You probably wanna go read the LP wiki on how to setup the thing in a network properly

but, if you want access to /lpb then you're gonna need perms to run it on the proxy

[spring tulip]

Okay.

[barren hollow]

Does waterfall support 1.8-1.18.2?

[tidal musk]

Yes

[barren hollow]

How much does waterfall server use on 50 players?

[marsh reef]

Well, the way I do it is by having a sql server on my network, then configuring all LuckPerms servers/proxies to use it

[reef fulcrum]

which is like, the recommended way, as per the wiki which I directed them to

[tidal musk]

How I get the download link for the latest version of the waterfall everytime?

[reef fulcrum]

there is no "latest" link, you need to parse it from the API, basically

[tidal musk]

Okey, thx.

[stiff hollow]

An request Waterfall can "support" 90 players with 500Mb of memory?

[scenic pebble]

Yes

[stiff hollow]

90 players with 500Mb of memory (Ram)? 😳

[scenic pebble]

It's a proxy. It doesn't use much memory.

[spark path]

Hmm

is 1core 1gb ram enought for waterfall

Cause i am having issue hosting 2 waterfall server

i have 1 workijg

but the pther 2 not working

both with no error stated

[tulip surge]

latest.log by @spark path: https://paste.gg/40549cc7b6db4721bcabb54e09e2cf1a

[tulip surge]

latest_2.log by @spark path: https://paste.gg/0193e7bf5c81476e8b09fcd87f304033

[spark path]

@upper stirrup

Pls help

the both server is running on the same machine as this two

and this 2 (1core 1gb ram) is not loading and is stuck

for 12h

already

1.18 waterfall

bungeecord also show the same thing

[upper stirrup]

Please don't ping random people from the sidebar. Thanks! 🙂

[spark path]

Ok

[royal hawk]

Both geyser and ViaVersion require a lot more resources than that. What you said is fine for an installation without plugins but those two plugins are a lot more heavy

[nimble veldt]

Does waterfall support fabric? or velocity

[reef fulcrum]

waterfall doesn't go out of its way to support fabric

velocity is your best bet in the modern age

[nimble veldt]

should i use velocity in general instead of waterfall?

[reef fulcrum]

Yes

[surreal stump]

if you are fine with less public velocity plugins or have an developer, yes.

[reef fulcrum]

only real to stick with waterfall now is plugins

[nimble veldt]

I can develop my own plugins

[surreal stump]

then go for velocity

[nimble veldt]

ok

[tidal musk]

hi

my players randomly disconnect from the server

"The server you were previously on went down, you have been connected to a fallback server"

even tho the server isn't actually crashing

[marble kelp]

well that's a pretty clear error message

[nocturne fog]

You'd want to look at the backend server logs for the reason they were disconnected

[open garnet]

Using BungeeCord, I have servers setup on a machine...but the issue is that ....i am able to join the server...but when my members try it they get this error - disconnected with: ReadTimeoutException : null

[reef fulcrum]

all that says is that something didn't send a packet in so long

[open garnet]

but im guessing its an error cuz its happening to every player except ne

do you know any solutions that could fix this issue?

[reef fulcrum]

there is no magical solution

you need to work out why whatever printed that message didn't recieve anything in 30 seconds

[wise kite]

https://github.com/PaperMC/Paper/issues/895#issuecomment-372175216

[reef fulcrum]

is generally the "catch all" type thing for that kinda area, but, is far from conclusive

[open garnet]

ill check this out rn

[hidden light]

are there any good hosting providers for waterfall networks?

[spiral portal]

Does waterfall have the api that bungeecord has?

[surreal stream]

basically yes

[spiral portal]

ok thx

[plush root]

I have ViaVersion on backend server, does anyone know how to fix this issuee?

ping_passthrough: true

[open garnet]

hey umm....i did everything u listed

still the same error .....

did u ever solve this type of issue

maybe u could share how

[reef fulcrum]

client version?

[plush root]

There is a major memory leak

[plush root]

1.18.2

I can't update to latest version as it has memory leak

[reef fulcrum]

No idea what version you're using

[plush root]

456

[reef fulcrum]

that error message generally inclines that there was a protocol version mispatch, outdated client implies that the proxy thinks its newer

[plush root]

My proxy is version 456

[reef fulcrum]

which means that you don't have .2 release support

[plush root]

Do you know that there is a memory leak in the latest versions?

[reef fulcrum]

no

[plush root]

.

Froze after seconds

[reef fulcrum]

That shows nothing useful

[plush root]

So basically few days ago i updated to latest version

[reef fulcrum]

you're the only person complaining about it, and i'm not aware of any changes which could induce that

[plush root]

and then realized the memory goes up to 16gb in less then 24 hours (300 on)

[reef fulcrum]

I'm not playing into the speculative game, especially over 30 odd different changes

[plush root]

and doesn't clean itself up

But when typing sparkb health

All good

I'm not the only one experiencing it too

around 4 other large servers i know are experiuencing it too and decided to stick to 1.17.x

[reef fulcrum]

0 bug reports, 0 investigative reports into it, etc

[plush root]

The bug is basically the memory doesn't get actually cleaned

i have GC1

java -Xms128M -Xmx16810M -XX:+UseG1GC -XX:G1HeapRegionSize=4M -XX:+UnlockExperimentalVMOptions -XX:+ParallelRefProcEnabled -jar bungeecord.jar

But the server shows it's cleaned but task manager/the panel shows that it's not

I tried with Windows 10 too

[reef fulcrum]

Don't mention

[royal hawk]

You’re allowing the instance to have the full 16.81 gb of Ram. Don‘t assign what you wouldn’t want it to use

[reef fulcrum]

that provides 0 useful information

[plush root]

Yup.. i had maxmium of 6gb but it still crashed

[reef fulcrum]

Don't set Xmx == to the container limit

[plush root]

Actually before waterfall i was using flamecord which is fork of this, and i had it on 1gb

But then noticed memory = gas gas gas

[royal hawk]

A profile using spark or another heap analysis tool would be more useful

[plush root]

I have spark

[plush root]

.

Look, it says 16.44gb on the panel, spark says 3.9GB

[royal hawk]

Again, what the system shows isn’t the ram the instance is actually using, only what’s assigned

[plush root]

I know what is heap

[royal hawk]

If you assign Xmx an amount of ram you can expect it to take up that amount of ram. That‘s normal and perfectly fine. If there’s actually a memory problem you’ll get Out-of-memory errors or the instance crashes

[plush root]

In this picture the ram is maxmium 6gb and the heap is 6gb

[plush root]

I didn't get out of memory error, it just freezes

[royal hawk]

List your plugins please

[reef fulcrum]

It just freezes because you hit the containers memory limit

Which is somewhat expected especially as java seems more active towards reaching its allocation amount

reduce Xmx, waterfall uses a good chunk of off-heap memory

I'm not aware of any regressions here, and the image you're showing really doesn't give anything to go off, nor are we getting any reports; if there is a change, I'm not sure which of the past 30 builds could of introduced such an issue, and given you're so far back, and I can't reproduce it, you're gonna need to dig yourself

[plush root]

I have custom plugins but i ran the proxy without any plugin too and botted it with 200+ bots walking around for few hours but same results

[plush root]

I already have digged into it and found what causes it

Last few updates of waterfall*

My version issue is fixed, i just installed viaversion on proxy

[royal hawk]

Might have to add that ViaVersion wont actually help you support newer versions

[reef fulcrum]

1. stop mentioning

2. "last few updates" is literally nothing to go off

what do you consider the "last few updates"?

[royal hawk]

there are 27 builds between their build and the current one

I am not going out of my way to analyze what upstream and we did

but I am pretty certain we didn’t introduce a memory leak

[plush root]

Well i guess i can't do anything now but i'm pretty sure there is something wrong with either Waterfall/Bungee

[royal hawk]

I‘m guessing rather than that there’s a change that breaks at least one of your plugins but then again please do a heap analysis to find out what’s actually taking up your memory

[plush root]

I did a heap analysis and many around 4/day in the past two weeks i found nothing weird in the memory but the weird thing is that the the java application says it's using 3.9GB, Panel/Windows: 16.44GB

[desert zephyr]

Im getting this error encountered exception: net.md_5.bungee.util.QuietException: Unexpected packet received during server login process!

[reef fulcrum]

You'd need to work out why an unexpected packet was sent to the thing

server software/version?
plugins? (can you reproduce without?)

[royal hawk]

Is that bedrock edition? If yes you probably should be asking on the geyser discord

[desert zephyr]

Nvm it was a plugin causing the error, but thanks

[tidal musk]

What's waterfall anyway?

[serene tide]

Waterfall is a proxy based on BungeeCord

[tidal musk]

Is there a distinction with Paper and Velocity?

[tidal musk]

Wut

Isn't bungeecord paid or something?

[serene tide]

BungeeCord is not paid

[tidal musk]

Oh cool

So I don't need BungeeCord to get Waterfall do I?

[serene tide]

No, you do not.

[tidal musk]

noice

So how's it different from Paper and Velocity?

And what do you recommend?

[serene tide]

Velocity is not based on BungeeCord. Paper is a Minecraft server based on Spigot.

[tidal musk]

Oh ye there's hangar too

[serene tide]

For your servers, Paper, for proxies, Velocity.

[tidal musk]

Wdym proxies?

[serene tide]

If you need a proxy, I'd recommend Velocity.

[tidal musk]

I'm kinda new to that whole thing lol

[tidal musk]

Coolio

[serene tide]

A proxy links multiple Minecraft servers together.

[tidal musk]

But it requires a host ye?

[serene tide]

Yes, you need to host the proxy if you want players to be able to connect to it.

[tidal musk]

Cool

[winged pilot]

how do you even begin to diagnose why everybody gets kicked with "Read Timed Out"

it's happening about twice a day now

no error, nothing, just player disconnected: read timed out in the waterfall console

[reef fulcrum]

depends, it's not a "trivial" issue to solve

read timed out just says that there hasn't been a packet recieved in 30 seconds, basically

and, that's really all way can say, from what end did it occur, was it a network issue, did the server stall out for a moment, etc, etc

[winged pilot]

nope no stall out, other than everyone being frozen in place as if you pulled the internet plug, until eventually reaching the timeout threshold

but, how would you determine what end it occured from, or whether or not it was the network

the backends dont have any errors or anything, and they're completely fine

could it be a crappy bungee plugin im using?

[reef fulcrum]

I mean, hard to say

that's where you potentially get into tools like mtr reports and such

there is no real guide out there for diagnosing this stuff afaik

[winged pilot]

ye ik i thought so

[viscid steppe]

Hello, would a setup like this work?

Forge client (1.12.2 modpack) -> waterfall -> lobby (1.18 paper, running via-rewind) -> the modpack server (running sponge forge)

[marble kelp]

depends on the modpack if it works, would be better to not go through the lobby with a modpack client

[viscid steppe]

Ok, so would the best way be to just have players use the /server command to directly connect?

[reef fulcrum]

server switching with mods is always shaky

some mods work fine with it, some don't

ideally, you just connect mods directly to the modpack server

[viscid steppe]

Ok, guess that makes more sense, might try anyways to see if the mods in use play nicely.

[open brook]

Hello, I configured TCPshield (protection from DDoS), it seems the server is displayed and everything works... but I can not let me in now
How can I fix it ?

[reef fulcrum]

probably bad setup

don't install their plugin on the actual servers

outside of that, consult them for support

[open brook]

I installed plugins on both servers, that must be the problem?

[reef fulcrum]

yes

[open brook]

Thanks, I just sometimes think that for many plugins to work properly - you need to install them on both servers

[dire ermine]

My players are having issues connecting to the server and this is the only message it’s giving in console.

com.mojang.authlib.GameProfile@7b7*****[id=92bd6844-89b5-4a7a-9967-ea80ea*****,name=Zepti****,properties={textures=[com.mojang.authlib.properties.Property@263c****]},legacy=false] (/*********) lost connection: Disconnected

I run a bungee network (Waterfall) and they only have this issue connecting to one of my three servers. They connect to the lobby and PvP server just fine but the vanilla server they can’t connect to and that’s the message that I get in console and it’s spammed because I have them connecting via ajQueue. I also have this message show up in console when I try and sent them to the server via bungee commands. Is there any reason why this would happen? It seems to have happened within the last week. (I updated the bungee/waterfall proxy and vanilla server to 1.18.2 this week, could that be related?).

[frail kayak]

Have client upload log to https://mclo.gs

Censors information that identifies them outside of their mc username.

[dire ermine]

Alright I’ll provide that as soon as the player can get it to me.

[reef fulcrum]

as the player can get it to you? what? it's your logs?

oh, client log, nvm

all that says is that the connection was disconnected, make sure that you're connecting to a server which has ip forwarding setup properly

i.e. not an actual vanilla server jar (unless patched)

[dire ermine]

It’s set up correctly and everything but players are randomly getting that message.

Not everybody is getting that message when they try and join but some people are, about 10 people within the past 3 days. Also it randomly fixes itself within a couple hours for individual players but not some other players.

I didn’t have this issue until I updated to 1.18.2 🤷‍♂️

[reef fulcrum]

nothing changed in 1.18.2 bar a few small trivial things

literally all that tells us is that the player was disconnected

[dire ermine]

Interesting, it started the same day I updated that one server but I’ve had no issues on the other server that’s also 1.18.2.

I’ll do some more digging and see if I can find anything

[sharp belfry]

just wondering anything big changed in 1.18.2 regarding bungee/waterfall? Switching servers/world felt slightly slower with more time in the loading screen

[reef fulcrum]

No

[nocturne fog]

the loading screen now has a hardcoded 2 second minimum though

[reef fulcrum]

Mojang made a change change which makes the terrain ^

[sharp belfry]

ah i see, is it just a client thing or does that happen with servers too?

the loading screen thing

[nocturne fog]

the client has the hardcoded value, and the client is used to join servers, so...

yes

see https://bugs.mojang.com/browse/MC-249059

theres a client-side mod you can use to fix it https://modrinth.com/mod/forcecloseworldloadingscreen

[sharp belfry]

I see, thanks for the info

[steep wharf]

My waterfall server and my papermc server are not on the same network. Then they occasionally lose connection in 5~10 minutes.
https://pastebin.com/vn2LyHZ 🙏

[royal hawk]

Network issue on your end. Can be anything really; usually is linked to some DDoS protections from one side
We can’t help with that sadly

[warped ferry]

I have two servers setup on my friend's vps. One is waterfall, the second one is paper. I can't seem to connect the waterfall server to the paper one.

I have dealt with bungee/waterfall servers for years now but I really can't seem to find the problem

It just looks like they can't communicate with each other. They are both on the same network, location and ip.

[royal hawk]

Are you using a hosting panel like pterodactyl? If not I assume its a firewall or rlimit problem

[warped ferry]

I am using pterodactyl

[royal hawk]

See pinned messages in #velocity-help

[tidal musk]

guys

why my server is not starting

in bungeecord

[marble kelp]

it is started

if you are refering to the info in the panel: contact your host/the panel support if self-hosting

[warped ferry]

Hi, I read everything and I did it but it still doesn't work...

Could you possibly come to a voice chat or something to guide me?

[marble kelp]

I suggest reading the docs of your panel for that

[warped ferry]

I use pterodactyl...

[marble kelp]

yes

[astral lynx]

hey i'm trying to start my waterfall proxy and this happens :
[15:26:50 ERROR]: Exception in thread "main" java.lang.IllegalArgumentException: Server betatest (priority 1) is not defined [15:26:50 ERROR]: at com.google.common.base.Preconditions.checkArgument(Preconditions.java:407) [15:26:50 ERROR]: at net.md_5.bungee.conf.Configuration.load(Configuration.java:160) [15:26:50 ERROR]: at io.github.waterfallmc.waterfall.conf.WaterfallConfiguration.load(WaterfallConfiguration.java:67) [15:26:50 ERROR]: at net.md_5.bungee.BungeeCord.start(BungeeCord.java:279) [15:26:50 ERROR]: at net.md_5.bungee.BungeeCordLauncher.main(BungeeCordLauncher.java:67) [15:26:50 ERROR]: at net.md_5.bungee.Bootstrap.main(Bootstrap.java:15)

[reef fulcrum]

Server betatest (priority 1) is not defined

[astral lynx]

So? how can i fix this ?

[reef fulcrum]

well, it tells you the issue

you have a server in your priorities list which isn't actually defined in your servers list

[astral lynx]

oh thx

[dire ermine]

Should bungee ip forwarding should be enabled?

[reef fulcrum]

generally, yes

[dire ermine]

Oh I already have it on, for some reason I was thinking it was off.

[rocky delta]

[20:58:39 ERROR]: Exception in thread "main" java.lang.ClassCastException: java.util.LinkedHashMap cannot be cast to java.util.Collection
[20:58:39 ERROR]:       at net.md_5.bungee.conf.YamlConfig.getListeners(YamlConfig.java:278)
[20:58:39 ERROR]:       at net.md_5.bungee.conf.Configuration.load(Configuration.java:92)
[20:58:39 ERROR]:       at io.github.waterfallmc.waterfall.conf.WaterfallConfiguration.load(WaterfallConfiguration.java:67)
[20:58:39 ERROR]:       at net.md_5.bungee.BungeeCord.start(BungeeCord.java:279)
[20:58:39 ERROR]:       at net.md_5.bungee.BungeeCordLauncher.main(BungeeCordLauncher.java:67)
[20:58:39 ERROR]:       at net.md_5.bungee.Bootstrap.main(Bootstrap.java:15)

Would this cause the proxy not to start?

[reef fulcrum]

yes

[rocky delta]

Whats the problem?

[reef fulcrum]

your config is bad

[rocky delta]

Does it tell me what line is bad?

[reef fulcrum]

no

I can tell you that you borked the listeners section, however

[rocky delta]

How can I get the server the player is on with waterfall api?

[rocky delta]

yea I screwed it up

It should be this...

[ruby badge]

https://pastebin.com/72xA4mcC
Getting this error every time i join

[blissful nebula]

Waterfall works on 1.12.2?

[surreal stump]

yes

1.8-1.18.2

[blissful nebula]

I do not work Waterfall. At first I could not log on to the server because it gave me an error and I could log directly to the server and not through the Waterfall after I set it to false online_mode and I could log on to the server with Waterfall but I can still log directly to the server

[dire ermine]

https://mclo.gs/1QQqWYC

Do you see any abnormalities in here?

[frail kayak]

Do two things

1. have them try no mods.

2. if that don't work. Verify mc server and client version

[dire ermine]

Alright I’ll do that

[frail kayak]

Looks like they get caught at the que system.

[dire ermine]

It’s completely random so it’s hard to replicate but I’ll update with the results here

[frail kayak]

Ah. Well imma DM you a link to my modpack I use, meant for vanilla things and has shaders and optimizations.

Throw that at them as it is built to run 16 chunk render distrance in single player off 2gb of ram.

[dire ermine]

Yeah they do and each time it tries to connect them to the target server, this pops up in the target server console. com.mojang.authlib.GameProfile@7b7*****[id=92bd6844-89b5-4a7a-9967-ea80ea*****,name=Zepti****,properties={textures=[com.mojang.authlib.properties.Property@263c****]},legacy=false] (/*********) lost connection: Disconnected

Alright

[hidden meadow]

Waterfall doesn't seem to be able to connect to my spigot/purpur servers, even though when I try to connect with bungeecord: false and online-mode: true it works

[reef fulcrum]

errors?

[hidden meadow]

``1`

[reef fulcrum]

99% of times are not setting up IP forwarding properly or still connecting directly to the servers and not the proxy

[hidden meadow]

[20:57:41 ERROR]: io.netty.channel.ConnectTimeoutException: connection timed out: /38.242.250.226:25566
[20:57:41 ERROR]:       at io.netty.channel.epoll.AbstractEpollChannel$AbstractEpollUnsafe$2.run(AbstractEpollChannel.java:613)
[20:57:41 ERROR]:       at io.netty.util.concurrent.PromiseTask.runTask(PromiseTask.java:98)
[20:57:41 ERROR]:       at io.netty.util.concurrent.ScheduledFutureTask.run(ScheduledFutureTask.java:170)
[20:57:41 ERROR]:       at io.netty.util.concurrent.AbstractEventExecutor.safeExecute(AbstractEventExecutor.java:164)
[20:57:41 ERROR]:       at io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:469)
[20:57:41 ERROR]:       at io.netty.channel.epoll.EpollEventLoop.run(EpollEventLoop.java:384)
[20:57:41 ERROR]:       at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:986)
[20:57:41 ERROR]:       at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
[20:57:41 ERROR]:       at java.base/java.lang.Thread.run(Thread.java:833)
[20:57:41 INFO]: [BoysWillBeBug] disconnected with: Could not connect to a default or fallback server. Incorrectly configured address/port/firewall? 

[reef fulcrum]

the connection to the server failed

[hidden meadow]

Is ip forwarding required?

[hidden meadow]

Servers online

[reef fulcrum]

Well, you generally wanna use it

don't mention

that's generally a network issue

if same server, use the loopback interface

[hidden meadow]

Ok ill try search the googles

Any chance you can direct me in the right direction?

[reef fulcrum]

I mean, depends on how you're hosting

if you're using a shared host you probably wanna speak to them

[hidden meadow]

I have it on the same vps on pterodactyl

[reef fulcrum]

so, your VPS probably doesn't allow you to use the public IP

[hidden meadow]

I set up the pterodactyl myself

[hidden meadow]

It does allow me to use the public pi

[reef fulcrum]

ptero has a guide on their site for how to configure networks up on their panel properly

[hidden meadow]

ip

[reef fulcrum]

don't mention

Literally not what I said

or, more, misinterpreted

[hidden meadow]

I'm sorry haha

[reef fulcrum]

not all hosts are configured to allow the VPS to ping itself using its own public IP

[hidden meadow]

do you test that using ping ip

[hidden meadow]

Tysm! It works now

[rocky delta]

Is it possible to use the Waterfall api in a spigot plugin? I just want to test what server the player is on

[surreal stream]

No.

[rocky delta]

So I have to write a separate plugin for each of my servers? I want to only be able to run a commands on one server.

[misty fern]

Hello is it possible to suspend with Waterfall simple multiproxy?

[ruby badge]

https://pastebin.com/N7JcrLfA

[reef fulcrum]

your database connection failed

[rocky delta]

What?

[marble kelp]

probably a plugin error

[jade egret]

I would like to ask if there are any tutorial to teach how to link the hub with the servers.
The hub and servers are all in one computer.
I would like to follow the tutorial to try to set up by myself.

[marble kelp]

not really much that is necessary, just make the servers listen on localhost and some port and add them to the listeners section in the config.ymk

*config.yml

[jade egret]

Each server use different port.
Then add them to the listeners section with same ip but different port.
Am I correct?

[reef fulcrum]

yes

[jade egret]

I will have a try.
Thanks!

[jade egret]

Is it possible to prevent players from joining the servers using the IP and port?
I want the players to join into the hub first.

[ruby badge]

How do i get it working so i can directly join a subserver?

[marble kelp]

@ruby badge what exactly is your goal?

@jade egret if they are on the same machine then run the servers on localhost, if not then use a firewall or vpn (or if you are on shared hosting BungeeGuard)

[ruby badge]

I have my lobby server and i want to go directly to that server so i dont need to go via waterfall

[marble kelp]

you can use forced hosts to join through a sub domain directly to specific servers

(forced hosts config section of the proxy)

[ruby badge]

Do i then join via a ip?

[marble kelp]

no, via the sub domain

[ruby badge]

This?
forced_hosts:
pvp.md-5.net: pvp

[marble kelp]

yes

[ruby badge]

But what is going to be there?

[reef fulcrum]

You can't connect to backend servers behind a proxy, that would introduce many dozen security risks and such

and so, the way to do it is to get a domain name, and use subdomains, e.g. as per the example there, pvp.md-5.net, to connect to the pvp server

[ruby badge]

Okey, But cant i do like connecting via the proxy but you just dont need to get via the lobby server but everyone other needs to do it

[reef fulcrum]

using a sub domain, yes

[ruby badge]

Okey

So like for the direct via the proxy to the lobby play. and skip that like survival.?

[tidal musk]

ok

uhh

so

What proxy should I choose:
Normal bungee
Waterfall
Velocity

[surreal stump]

Velocity >>>> waterfall > bungee

But for velocity aren't much plugins on the market, so you might code them by urself

[tidal musk]

so?

uhh

I would choose

uhh

waterfall

[tidal musk]

cause I have no idea how to code a plugin

but

what about hexacord?

or flamecord?

nvm hexacord only supports 1.7 - 1.11 and flamecord is a fork of waterfall, waterfall a fork of bungeecord and flamecord a fork of bungeecord water fall

[surreal stump]

Waterfall

[heavy hornet]

had no clue this was still maintained

swear it’s hella old

[marble kelp]

fix what exactly!

*?

[daring zodiac]

[16:16:29 WARN] [RedisBungee]: Player 0bed7147-fa4c-4e48-b608-839756136775 is on the proxy but not in Redis.
[16:17:29 WARN] [RedisBungee]: Player 0bed7147-fa4c-4e48-b608-839756136775 is on the proxy but not in Redis.
[16:33:29 WARN] [RedisBungee]: Player 0bed7147-fa4c-4e48-b608-839756136775 is on the proxy but not in Redis.
[16:34:29 WARN] [RedisBungee]: Player 0bed7147-fa4c-4e48-b608-839756136775 is on the proxy but not in Redis.
[16:36:29 WARN] [RedisBungee]: Player 0bed7147-fa4c-4e48-b608-839756136775 is on the proxy but not in Redis.
[16:52:29 WARN] [RedisBungee]: Player 0bed7147-fa4c-4e48-b608-839756136775 is on the proxy but not in Redis.```

[jade egret]

If they are on the same machine, I set the IP to be 127.0.0.1 in the server.properties of all paper servers except Waterfall.
No need to set up any firewall to prevent players directly connect to the paper servers.
If they are not on the same machine, install iptables in each machine which runs the paper server and block the port they are using.
Then, the players cannot connect to the paper servers directly as the port is blocked by the firewall.

Am I correct?

[prisma whale]

No need to set up any firewall to prevent players directly connect to the paper servers.
If your firewall is on and set to deny all by default and you dont have your backend servers open then no.

Regardless if your servers are on the same machine or not, they should be firewalled. The difference between same machine and different is on a different machine you would tell the firewall on the backend servers to accept connections from the proxy.

[tidal musk]

anyone help?

my server keeps showing this

https://paste.md-5.net/riharehezo.makefile

here is the config

https://images-ext-1.discordapp.net/external/wGmDTjm9fvMcF2S-s4qiy4R_kFa9gsKNh-PgKKK6_yY/%3Fwidth%3D1381%26height%3D586/https/media.discordapp.net/attachments/962036897532768327/962047164148772975/unknown.png?width=1273&height=540

and the error is here

[marble kelp]

@tidal musk nothing to do with the config. your network setup is simply wrong or you are connecting to the wrong ip/port

[alpine notch]

Hello, i'm trying to set up a waterfall server, and I have the servers set up, and (i think) properly configured, but when i go to connect to the main server, I get a "Time out" error

[icy cosmos]

This error occurring while im trying to connect to another server !

[real hollow]

I am getting an error

[23:37:18 INFO]: [SpicyTakis] disconnected with: IllegalStateException : Too many registered channels. This limit can be configured in the waterfall.yml @ com.google.common.base.Preconditions:502

Is there any drawbacks of increasing the limit in waterfall.yml?

[reef fulcrum]

well, back in the days people used to use that to run a server out of memory

So, yes, you can increase it but don't be stupid with it

[reef fulcrum]

waterfall requires you to do so to

you can't MITM auth

firewall or bungeeguard

latter, yes

[tulip surge]

https://docs.papermc.io/velocity/security

[nocturne fog]

see these docs as well. for velocity, but applies to waterfall as well

just ignore the bits about "Modern Forwarding" - that is velocity exclusive

oh and I guess that wouldn't talk about bungeeguard either. but everything else stands

[real hollow]

Alright, but what made it run out of channels though? Right now I am trying to join on a client with a heavily modded forge client. (upwards of 150 mods).

Ok so I dont get the error when I join through a vanilla client but it does show the error when joining through the forge client.

[reef fulcrum]

Because your forge mods apparently registered more channels than the proxies limit allows...

[real hollow]

hmm

What would a be a way to circumvent this? I currently have 2 vanilla servers and 1 forge server.

[reef fulcrum]

increase the limit?

[real hollow]

jeez I have it up to 2000 and it still throws the error 🥲

oh wait I think I have been changing the wrong setting

Yep I was changing the wrong one 🤣
I was changing plugin_channel_name_limit instead of registered_plugin_channels_limit

Anyways, thanks for the help!

[devout flume]

Is there a way to use Waterfall to host 2 bedrock-supporting servers on the same host and let players join them with different hostnames, no port necessary?

I want play.khavalon.com to go directly to our "Khavalon" server and play.asthonia.com to go directly to our "Asthonia" server. No port needed, and Waterfall would be completely transparent to the users (they wouldn't know it's a network - the servers would appear to be separate from users perspective).

Possible?

[reef fulcrum]

waterfall doesn't support bedrock itself

[devout flume]

I would need to use Geyser

[tidal musk]

tthen who do I fix it?

[cosmic cosmos]

Hey guys,
can you please help me.
I just made waterfall proxy with like 5 servers connected.
But the status blinking. Once online, once offline. (oneblock,event)
Both work, we can connect to them, but status blinking. My bungeecord config is almost default
.
Please ping me on answer
.

[tidal musk]

I am running a waterfall server with luckperm installed. I am trying to give ranks prefixes, but I have had no luck(no pun intended) doing so. How can I do this?

[rocky delta]

Your config is messed up

[icy cosmos]

Fixxed now !

Thanks for responding !

[mental lintel]

I got a small question:
When I use Waterfall on my server provider (gg servers), it works fine but when loading into the world it kinda chugs on encrypting (it stays on encrypting for a while )

Once it takes a bit you finally load in
This doesn't happen when I use bungeecord

Would anyone know what the cause of this may be?

[stark lagoon]

Hello 🙂 i have question about waterfall/bungecord 😄 i just go massive bot attack on my waterfall server, ram usage went to 100% and people from servers connected to it, disconnected 😄

what to do in such a situation when someone attacks waterfall server? and probably crash 🙂
people can't stay on other paper servers? waterfall server need to be always on? 🙂

i will increase server ram to 2gb and buy BotSentry to put on waterfall server so that he can block all these connections, for now maybe helps 🙂

I don't know what other networks are doing in this situation

[tulip vortex]

BotsEntry is one of the worst anti-bots out there, you could try to migrate to Velocity if you want to have a better management of the resources on your server and/or increase the ram available on your proxy, 1gb is a bit small to support bot attacks; also, the best thing you can do is to use online mode instead of offline mode

[reef fulcrum]

waterfall is a proxy, which, like

proxies.

if the proxy dies everything going through it dies

most servers often just use load balancing and proper firewall limits to try to cope with it

then theres mitigation services like TCPShield

[quaint pond]

or me :p, btw i updated the web as you requested

[plush parrot]

How can i set server-icon.png

[marble kelp]

simply putting a file named like that in to the proxies root should work if you disable ping-passthrough

otherwise set it on the server where you pass the ping through to

[round timber]

is waterfall a safe drop-in replacement for BungeeCord?

or would it require some extra configuration beyond just swapping the jars

[reef fulcrum]

generally safe drop in, only caveat is docker in which you'll wanna disable async DNS inside of the waterfall config

*docker hostnames

[round timber]

not currently running in docker so that's a non-issue

[icy cosmos]

While running /server , this error occurring & player can't connect to the server !

latest.log

[tulip surge]

latest.log by @icy cosmos: https://paste.gg/ff7387a563e84da19acea90b002e7050

[icy cosmos]

Here's the full log

From line 109

[royal hawk]

I don’t see where you’re having a problem
the proxy can’t reach that address. Did you configure your networking correctly?

[icy cosmos]

Yeah i did

Should I provide any ss ?

[royal hawk]

Then you have some kind of system issue. This error is only thrown when the system tells the proxy that it has no idea how to get to that address

If you’re running a hosting panel like pterodactyl you need to go to them for support

[icy cosmos]

Okhay !

Thanks for responding , appreciate it 💕

[tidal musk]

[22:35:09 ERROR]: [/93.123.16.47:60252] <-> InitialHandler - encountered exception: net.md_5.bungee.util.QuietException: Unexpected packet received during login process! 030000530100004f03003f47d7f7ba2c
[22:35:09 ERROR]: [/93.123.16.47:60255] <-> InitialHandler - encountered exception: net.md_5.bungee.util.QuietException: Unexpected packet received during login process! 03000069010000650303551ca7e47261

what is this

[quaint pond]

93.123.16.47 is sending weird packet to your bungee

[lofty wadi]

any idea why this happens?;

i have just reset the server and set it up the same way? :3

[lofty wadi]

i get the same. 🤔

[reef fulcrum]

something sent an unexpected packet to the server

well, i mean, the proxy

[lofty wadi]

i know is the proxy, but how to fix? 🤔

[reef fulcrum]

work out why it got invalid data

reproduce without plugins, make sure you're using a vanilla client, etc

[lofty wadi]

i'm using a vanilla client and is only vanilla jar installed? 🤔

^
that's server i wanna connect to

[reef fulcrum]

well, vanilla itself doesn't work behind a proxy server

you'd need to run without ip forwarding or find a way to patch the vanilla server jar

[lofty wadi]

okay, i'll see if find a work around.

[tulip surge]

message.txt by @crimson citrus: https://paste.gg/f0571f6e37d9419281e7664268e4e8ed

[tidal musk]

can someone tell me what is this and how to fix it

[royal hawk]

Some discord plugin you have is malfunctioning

[lucid sail]

I'm trying to set up a small server with waterfall, but I'm having issues with teleporting between servers. If the player is teleporting to a server where they are in a different dimension from the one they are currently in, they join and then immediately get kicked. The waterfall servers says:
UpstreamBridge - SocketException: Connection reset
[LMaster765] disconnected with: SocketException : Connection reset @ sun.nio.ch.SocketChannelImpl:394. Any idea why this is happening?

[flint swallow]

[23:32:05 ERROR]: io.netty.channel.ConnectTimeoutException: connection timed out: /65.108.245.87:25626
[23:32:05 ERROR]: at io.netty.channel.epoll.AbstractEpollChannel$AbstractEpollUnsafe$2.run(AbstractEpollChannel.java:613)
[23:32:05 ERROR]: at io.netty.util.concurrent.PromiseTask.runTask(PromiseTask.java:98)
[23:32:05 ERROR]: at io.netty.util.concurrent.ScheduledFutureTask.run(ScheduledFutureTask.java:170)
[23:32:05 ERROR]: at io.netty.util.concurrent.AbstractEventExecutor.safeExecute(AbstractEventExecutor.java:164)
[23:32:05 ERROR]: at io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:469)
[23:32:05 ERROR]: at io.netty.channel.epoll.EpollEventLoop.run(EpollEventLoop.java:384)
[23:32:05 ERROR]: at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:986)
[23:32:05 ERROR]: at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
[23:32:05 ERROR]: at java.base/java.lang.Thread.run(Thread.java:833)
[23:32:05 INFO]: [Volxy_] disconnected with: Could not connect to a default or fallback server. Incorrectly configured address/port/firewall?

[reef fulcrum]

UpstreamBridge

that's the connection between the client and the proxy, connection reset being that the connection was closed in an unexpected manner, check the client logs, etc

sour, network issue, if using a host, ask them for support; otherwise, if ptero, see the pins, otherwise, use the loopback

[tidal musk]

Hello, I am using waterfall but I am now discovering a weird issue where proxy crashes like every 1-5 hours. I couldn't figure out what is the main issue. I'm not 100% sure if this could be cause of memory limited reached?

[tulip surge]

latest.log by @topaz heron: https://paste.gg/291b10265c7c4bb6a0db2b20451707b0

[tidal musk]

Feel free to (@) notify me. Thanks!

[marsh reef]

In linux, you can use pmap to check memory usage, but you might have to get the pid for that (top)

[craggy trout]

@tidal musk I think i discovered the root issue, is TCPShield yeah that reverse proxy.

*19:32:29] [Netty Worker IO Thread #22/WARN]: Plugin listener net.tcpshield.tcpshield.bungee.handler.BungeeHandshakeHandler took 85ms to process event PlayerHandshakeEvent(connection=[/51.178.221.24:44923] <-> InitialHandler, handshake=Handshake(protocolVersion=758, host=havencraftmc.com///127.0.0.1:25565///1649849553///MGQCMH3sqUqlyqtcuEpuEbDIMYujJVh5Zvx1nqdMGXtq4/KaX02HBPlt761WePvRAep6zgIwCX0xkGY47wr4E+XoRQAB+Sotyl+ofHyt+0bezyF1ZiKqF/bRNxebMytmxtviKe9X, port=25565, requestedProtocol=1))!
Also, it pings a lot of times they're ip, then you just get out of memory by that.

See if some of your plugin is not working good with it.

[plucky briar]

I'm very new to bungee cord and waterfall so I don't really know much stuff but I got two world survival and lobby when I join the server it immediately bring me to survival how do I make it bring people to lobby first

[reef fulcrum]

in the config, you set the priorities to have lobby in there at the top

noting that if you don't have force default it will connect you to the server you logged off from by default

[plucky briar]

also is there a way to join the server without entering port or do i need to get a domain

Nvm ^^

[royal hawk]

I don’t recommend them but there are free services for getting a domain like freenom if you want one

Just that if the port isn’t 25565 you need either an SRV record or you need to enter the port

[reef fulcrum]

That's okly shown in the console/logs

[tidal musk]

[Kira_xDD] disconnected with: Exception Connecting:DecoderException : net.md_5.bungee.protocol.BadPacketException: Couldn't read all bytes from a packet. For more information, launch Waterfall with -Dwaterfall.packet-decode-logging=true @ io.netty.handler.codec.MessageToMessageDecoder:98

[reef zenith]

Could not connect to a default or fallback server. Incorrectly configured address/port/firewall? my promblem while connecting to my waterfall proxy

[surreal stump]

In the "servers" section of the waterfall.yml is the address/port incorrect or the backend server is offline

[plucky briar]

So I found something really confusing, I want my server to be online mode but when i set online mode to true (Before this i set it to false) and run the server but when i tried connecting to the server it say "exception connecting quiet exception server is online mode" and i cant seem to find a way to fix it

I have 2 server, lobby and survival i set both online mode to true i also set The Waterfall config online mode to true

[reef fulcrum]

You can't MITM auth

and so the backend servers need to be in online mode, and you let the proxy deal with the authentication

[glass palm]

is there any reason why Waterfall sits on Encrypting... for 5 seconds and BungeeCord doesn't? its not a huge issue, just feels slow logging in. we don't use Fastlogin or anything similar

[reef fulcrum]

Basically, it's a long standing issue, disabling netty dns in waterfall.yml can help with some aspects of the issue, but, generally, who knows; some people experience it with one or the other, some experience it with both, etc, etc

[glass palm]

yeah its weird. i made a clone of our waterfall setup and pasted it onto a different node and it worked perfectly fine

setting use_netty_dns_resolver to false completely solved the issue, ty

[tidal musk]

anyone know how to coding this bot for minecraft server status? if anyone know sned me a inbox massege

[royal hawk]

That’s the wrong channel for this

[tidal musk]

what channel @royal hawk

[reef fulcrum]

I mean, that's more #general given that it's nothing to do with waterfall or paper

wiki.vg has lots of info on stuff, but, I mean, theres probs a library or some prewritten function you can copy/paste for pinging servers in whatever programming language you opt to use

[wise kite]

https://docs.papermc.io/paper/anti-xray

[icy cosmos]

latest.log

[tulip surge]

latest.log by @icy cosmos: https://paste.gg/09a113b9730d4029953b41887b39c55f

[wise kite]

In 1.17+ Mojang removed the level-seed by default from server.properties. But you can still set a custom seed! Just manually add level-seed back to server.properties. It will still apply to freshly generated worlds

[blazing ocean]

You need ViaVersion if you want ViaBackwards to work

[icy cosmos]

I've all of them 😢

[reef fulcrum]

I mean

according to the log, nope

You'd need to work out why it didn't load viaversion

[icy cosmos]

Wait what it's not what I wanted to send !

Im sorry !

proxy.log.0

[tulip surge]

proxy.log.0 by @icy cosmos: https://paste.gg/2b3e717dd62345d4b77ee376c8428ae4

[reef fulcrum]

that's not waterfall

[icy cosmos]

Wait what ?

It should be !

Lemme check

[reef fulcrum]

Waterfall doesn't do proxy.log.0, etc

so, it's not

and viaversion is throwing exceptions, in which you'd wanna speak to them, not us

[craggy trout]

You are using default BungeeCord.

[icy cosmos]

But i need Aegis as anti bot too 😖

[reef fulcrum]

Then ask them for support

You're not using our software

[craggy trout]

That's a very bad fork, and not even of waterfall.
You should use Velocity with firewall or waterfall with firewall.

[brisk fox]

Does anyone knows about any kind of issue regarding players being able to move, others seeing them moving, console receiving commands, but player not receiving any commands result, and finally getting kicked due to keepalive timeout?

[craggy trout]

Must be a plugin

[brisk fox]

Well, maybe thats the issue. But i have been using the same plugins for years, none of them seem to cause such error or issues. But, upon setup of a Waterfall proxy with papremc servers, this started to happen.

Any hint of how must i try to debug it?

As an example, i can get into server, and spawn automatically on server A
I can move, others can see me move, and i can write /server B but without any effects/messages.
Then after a few seconds i'll be disconnected.

Upon reconnecting again, im on server B, as the command were successfully processed.

[icy cosmos]

I don't know how to use waterfall with firewall !

Is there any info or docs about it ?

[craggy trout]

I mean iptables or linux, or some reverse proxy idk

[hardy spade]

Hello, is there a startup parameter proporsal from PaperMC for Bungeecord?
Which Java Version should I use for Waterfall Latest?

[surreal stump]

java flags -> see pins
java 17 or 18

[paper meadow]

Hello. I'm currently trying to set up a new proxy and everything works fine excepting binding the port.

[16:08:26 WARN]: Could not bind to host /0.0.0.0:201
io.netty.channel.unix.Errors$NativeIoException: bind(..) failed: Permission denied

I know that it states "Permission denied", but I don't know how to grant the user more permission as he is the owner of the directory

[reef fulcrum]

bind

nothing to do with the directory perms

You're tryna bind on a reserved port number

[paper meadow]

okay, so I should choose a different port and it should work?

[reef fulcrum]

use port 1024 or higher, need root by default for anything below that

[paper meadow]

okay, good to know. Thank you :)

[tidal musk]

what is the version of minecraft waterfall is based off of (yes I've looked on Google and nothing)

[reef fulcrum]

None, it's a proxy

it supports 1.8-1.18.2

[tidal musk]

awesome so no Via Rewind

yay

[reef fulcrum]

well, it's a proxy

it doesn't do any protocol translation, you need a 1.18 client to connect to a 1.18 server, etc

[tidal musk]

ah so players can't connect with newer versions of Minecraft?

ah

okay

and I expect most of our players will be running 1.18 clients

[tough vale]

you can run viaver/via rewind on the backend servers (unless viaver has a bungee version, don't think so though)

[tidal musk]

also just curious I wanted to run GeyserMC so bedrock players can connect is that possible or would I have to give them the Ip's to the servers themselves

and just run Geyser on our paper based servers

[tough vale]

uh I think geyser has versions for proxies, if not you can run standalone geyser pointing to the proxy

[tidal musk]

true

and I can look

yep it's right on their jenkins

they have a version for bungeecord

[sand prism]

same as you but only with some player

[brisk fox]

It does happen with all of us, even me. On my case, the second timw we login, everything works fine.

Are you also on waterfall/papermc?

[surreal stream]

papermc?

im confused

[wide thistle]

Is there a way to use bungee commands in game?

[reef fulcrum]

depends on what you mean

if you have perms to run a bungee command, you just run them like literally any other command

if you mean, make the server itself run a command, i.e. a plugin, a command block, no, you'd need a plugin to deal with sending the command over or something which uses the messaging channel to do stuff

[wide thistle]

So how to give perms to use bungee commands?

[reef fulcrum]

in the config file, or, ideally, using a perm plugin like LuckPerms on the proxy

[wide thistle]

What's the perm's name?

[reef fulcrum]

don't mention

[tulip surge]

BungeeCord Commands | SpigotMC - High Performance Minecraft

https://www.spigotmc.org/wiki/bungeecord-commands/

Jun 1, 2021 ... Permission node: bungeecord.command.alert. Issues an alert to all users connected to BungeeCord. This is useful for announcements that need ...

About 1,100,000 results (0.26 seconds)

[sand prism]

write me , so i can tell you the situation

[reef fulcrum]

It's been reported over the years in bungee and waterfall in varying manners, sometimes both hit it, sometimes only one of them, etc

Nobody can reproduce it in a useful manner for looking into it, etc

[brisk fox]

Sorry, Are you talking about our issue?

[reef fulcrum]

yes...

[sand prism]

the thing is that there no error on proxy, when the timout happend there just ''player has disconnect'' like he click on disconnect

[reef fulcrum]

I know

The connections just oddly seem to stall

sometimes it seems to recover in some manner

sometimes all you can do is reconnect

[sand prism]

the nice thing is that the problem suddenly appeared

[serene shadow]

is there a way to connect the waterfall server with fort forge 1.18 1?

[reef fulcrum]

basically, no

there was a waterfall fork but that needed a client mod to work

[serene shadow]

wanted to add this server create live 3 to my network but the connect with the mod doesn't work, is there any alternative to make it work

[reef fulcrum]

no

[serene shadow]

ok

[strong mantle]

Did you end up figuring this out? I'm bashing my head against the wall, can't resolve or understand what's going on...

[brisk fox]

At all..

Whats your value in config.yml for waterfall?

network_compression_threshold: 256

Is this value the same in your subservers?

[surreal stump]

the networks goes over the proxy - waterfall.
my value is default

[brisk fox]

Not sure if default means 256, or 512.

[surreal stream]

default is 256 iirc

should be

[royal hawk]

The second one is the correct one; Also this only applies for paper- Bungeecord doesn’t have a gui

[brisk fox]

@sand prism @strong mantle @surreal stump
AS far as i have discovered, this only occurs on a proxy level, and has nothing to to with subserver configuration. Those, in fact, recognise the players connection.

Its waterfall who fails on handle the information, and throw no errors at all.

Using following plugins:

advancedban, AdvancedNmotd, ALBungeeBridge (Advanced Login) BungeeControlRed (ChatControl), BungeeGuard, BungeeJoinMessages, LuckPerms, PlHidePro, PremiumVanish, Protocolize, SkinRestorer, TAB

Among with following modules:

version: 2
modules:
- jenkins://cmd_alert
- jenkins://cmd_find
- jenkins://cmd_list
- jenkins://cmd_send
- jenkins://cmd_server
- jenkins://reconnect_yaml

[sand prism]

thanks for the info! , so its there a way to solve?

[strong mantle]

Happens on a standalone paper server for me

Maybe the issue is there

[brisk fox]

No, Now you shall post your info, and we may find common points.

[sand prism]

i have only 4 pl as yours

[brisk fox]

I dont think so, it may be another issue, as on our side, commands are properly executed on server, but proxy fails to recognise the updated connection info.

[brisk fox]

Do you remember any other kind of information it could suits me as hint?

[tidal musk]

if i have
priorities:

• bungee1
• bungee2
  and bungee1 is offline will i join bungee2? (ping me)

[reef fulcrum]

I've heard the odd mention that disabling compression on the proxy will magically resolve it

but, that's basically it

[agile lotus]

Hi guy,
When I start my bungeecord server he spaming during 30 minutes (too many open files)

Like that :
[07:05:13] [Netty Boss IO Thread #0/WARN] [io.netty.channel.DefaultChannelPipeline]: An exceptionCaught() event was fired, and it reached at the tail of the pipeline. It usually means the last handler in the pipeline did not handle the exception.
io.netty.channel.unix.Errors$NativeIoException: accept(..) failed: Too many open files
[07:05:13] [Netty Worker IO Thread #17/INFO]: [/127.0.0.1:51924] <-> InitialHandler has pinged
[07:05:13] [Netty Worker IO Thread #17/ERROR]: java.io.FileNotFoundException: plugins/StelyMotd2/config.yml (Too many open files)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at java.base/java.io.FileInputStream.open0(Native Method)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at java.base/java.io.FileInputStream.open(FileInputStream.java:219)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at java.base/java.io.FileInputStream.<init>(FileInputStream.java:157)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at net.md_5.bungee.config.YamlConfiguration.load(YamlConfiguration.java:77)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at net.md_5.bungee.config.YamlConfiguration.load(YamlConfiguration.java:71)

I saw that : https://mtyurt.net/post/docker-how-to-increase-number-of-open-files-limit.html
Do you think that might be the solution?

[reef fulcrum]

I mean, if you're using docker, "yes"

[brisk fox]

This did not work on my case.

[reef fulcrum]

I mean, idk if 1024 is an accurate default there, that just seems wrong, but, you'd need to verify

[brisk fox]

All i can do is testing velocity, i guess.

[agile lotus]

I don't know what "docker" is

[reef fulcrum]

then that article is probably irrelevant given that it's literally for docker.

but, generally, you'll either need to work out why you're hitting the configured file limit, or, increase said limit

[agile lotus]

But I did the "ulimit -n" command in my Putty and the displayed result is 1024

[reef fulcrum]

then you'll need to increase the ulimit, basically

[agile lotus]

If i make "$ ulimit -n 2048" it's good for a test ?

[reef fulcrum]

I mean, sure?

but, I mean, you probably wanna actually set that option in your system so that it sticks

[agile lotus]

I don't know much about it...
I just wish my bungeecord would stop spamming me every morning when I restart "too many open files"

This is the only answer I could find on the discord while searching so...

[reef fulcrum]

your issue is a linux config thing, not a waterfall specific thing

you probably wanna work out how you're hitting that many open files

but, increasin the limit will at least stave it off

https://askubuntu.com/questions/1049058/how-to-increase-max-open-files-limit-on-ubuntu-18-04

[agile lotus]

But how can I increase this limit in Linux?
Since this is a Linux problem according to you...

[agile lotus]

Hmm where I can past ?

[tulip surge]

limits.conf by @agile lotus: https://paste.gg/74cf5f8a5ead4db99cbce85e6b36bac1

[reef fulcrum]

don't mention

[agile lotus]

Okey:x

[reef fulcrum]

@AlixAxel : This answer worked for me, in addition to * - nofile 16384 ( I was only attempting 16384) in /etc/security/limits.conf. –
Doug Smythies
Jun 23, 2018 at 14:45

in the comments of the first answer

[agile lotus]

I speak about " added the following lines to /etc/security/limits.conf:"

[reef fulcrum]

Yes.

[agile lotus]

I see the file where you have to add the lines but I don't know where to add them

[reef fulcrum]

at the bottom?

It's like a single line

See the last answer

or the comment

[agile lotus]

It's not a single line I have to add

• soft nofile 65536
• hard nofile 65536
  alix soft nofile 65536
  alix hard nofile 65536
  root soft nofile 65536
  root hard nofile 65536

No ?

[reef fulcrum]

• is a wildcard.

which, covers all.

Which

is literally covered in the last answer

or, the comment I referred you to

[agile lotus]

I'm French and I don't understand it at all so if someone tells me to go to a site in another language where there are hundreds of lines where I don't understand anything, that doesn't help me.

[reef fulcrum]

there are multiple answers on that page

the last one, as I've said multiple times, eappears to be the easy "it just works" one

if you can't read that answer, then you can't read what I'm saying and are thus wasting my time

[agile lotus]

So in :
/etc/systemd/system.conf
Actually I have >> #DefaultLimitNOFILE=1024:524288
I have to change to >> DefaultLimitNOFILE=65536

In :
/etc/systemd/user.conf
Actually I have >> #DefaultLimitNOFILE=
I have to change to >> DefaultLimitNOFILE=65536

And in :
/etc/security/limits.conf
I have to add to the end >> * - nofile 65536

[reef fulcrum]

I'm not sure that you need to edit the systemd configs

but, as per the thing I've told you to read multiple times now, yes.

[agile lotus]

Do you think I can just try doing this for test?
And in :
/etc/security/limits.conf
I have to add to the end >> * - nofile 65536

[reef fulcrum]

yes.

[agile lotus]

With 65536 the max file with "ulimit -n" is 65536 or they have a convertion ?

Because with 4096 I should have had enough I think nn?

It's 4 times more anyway

[reef fulcrum]

I mean, 1024 is generally enough for most servers

double it, quadruple it, idk what all is best for you

as I said, you probably wanna work out why you're using so many files

But, I seem to be stuck going around in circles and would rather not spend the next hour or so going around in circles

[agile lotus]

I'm not dev i can't found the problem

I can only try to fix as I can

So like this ? :x

[agile lotus]

😑

[reef fulcrum]

I've said yes several dozen times now

I'm not going to keep repeating yes for the next 24 hours until you finally get the message

[agile lotus]

Thk bro

[dapper quail]

Hey. I have a question. I know thats the purpose of the proxy but it gives to everyone the same ip right? What happens if I use something that needs to know someon'es ip on the server. For example an ip ban

is it possible to disable that or is there a workaround?

[tidal musk]

that's what ip forwarding is for

[dapper quail]

is it possible to change it?

[tidal musk]

ip_forward in bungeecord.yml iirc

and set bungeecord to true in spigot.yml

[dapper quail]

thanks

let me try

[dapper quail]

Thanks so much 😄

[shut musk]

Is waterfall faster than velocity? In terms of memory and CPU usage?

[young rampart]

doubt it

[shut musk]

Cus I'm kinda being bottlenecked by velocity

[young rampart]

Waterfall is just bungeecord with a few more patches on top, Velocity is completely rewritten from the ground up to be a good proxy

I doubt that too, but I guess try asking for help in #velocity-help

[dapper quail]

Any help?

[royal hawk]

It’s just that. Fix your connection from proxy to server

[dapper quail]

I am sure I have done everything correctly

I don't know

[royal hawk]

Are you using a hosting panel like pterodactyl

[dapper quail]

Yes

I host it on a vps

Where I installed ptero

[royal hawk]

See pinned messages in #velocity-help for pterodactyl (also applies there)

[agile lotus]

I did that and this morning my limit is still at 1024...

So if someone can take a few minutes to explain to me which file I have to modify to change this limit thank you.

[craggy trout]

Why waterfall does not backport the commit that added this.
Is very bad when lot of connections try to join an unreachable server the console some times just crashes.
This is the commit: https://github.com/SpigotMC/BungeeCord/commit/eae9d45c8a4142541dfda7f6322784d148d4b027

[royal hawk]

@craggy trout see this
That’s not how waterfall works and this code is very much inside Waterfall

[reef fulcrum]

That commit was partially backported because

1. sending ips of servers it can't connect to players is bad
2. the 'fix' for that was literally to suppress errors so you can't see why normies couldn't connect
3. fix your setup, we either have 0 info when shit goes sideways, or, we have info, shamefully, we need info to help support people.

[tidal musk]

Is there any way I can run a proxy command on a backend server through console?

I have a bungeecord made by my server host that starts a server and I would like it so that when you click on an NPC it starts that server

[reef fulcrum]

You'd need a plugin on the proxy which the server can send commands to

[tidal musk]

Any plugins that do that?

[reef fulcrum]

Yes, but, idk what it was

[tidal musk]

Aww you don’t remember a name?

I really need something like that

If anyone knows of a plugin that does that please dm me

[craggy trout]

https://github.com/netty/netty/blob/4.1/transport/src/main/java/io/netty/channel/nio/AbstractNioChannel.java#L262

[reef fulcrum]

yes? and?

[craggy trout]

I mean, in my opinion is just useless on some point.

[reef fulcrum]

I mean, I fail to understand what you're even getting at

that patch exists in waterfall

[craggy trout]

not really

[reef fulcrum]

waterfall reverted that because, that line you linked

with spigots patch there

Literally sends the IP address to the client.

They later limited it to admins only, however now supress all info in the logs, etc

r. Incorrectly configured address/port/firewall?

You'll note that that message shows up in the client when they're booted

said change was literally done in the commit you linked.

[craggy trout]

i see

[reef fulcrum]

annnd?

Yes, I'mma look into cutting down what is logged to the console

[craggy trout]

In my humble opinion, this is a bit ugly in some cases.

[reef fulcrum]

But

Once again

The commit you're linking

Literally, sends that info

to the client.

Including the IP address.

[craggy trout]

Yes, my bad just i was going to mean this-

Imagine a large server, which for some reason crashes the spigot, its console would be a blood party.

[reef fulcrum]

Yes.

I know.

as I've said it's on my list of stuff to deal with.

[craggy trout]

Thanks, I would pr it but I don't know if it would be the right way.

[tidal musk]

Does anyone know of any plugins that do that?

[marble kelp]

@tidal musk https://github.com/Phoenix616/ConnectorPlugin 👀

[tidal musk]

What are the benefits of waterfall/downsides completed to bungee cord?

[surreal stump]

you can read the whole commit history but yeah. its better

[zealous grail]

the command /alert works in game
but using in the bungee console alert does nothing

no message appears in game

any idea why?

actually no command works in console

it doesnt even return unknown

I thought it might be the ptero egg, but it seems fine

hhmm

[blazing ocean]

What's the recommended way to run this with? Just redis? I have a bunch of paper servers behind TCPShield, this might be a decent way for running commands on all of them

[agile lotus]

Hi,
I made this but my problem still.
Someone have an idea to resolve my request ?
All day when my bungeecord restarting his spaming with "too many files" while 30 min...
Like that :
[07:05:13] [Netty Boss IO Thread #0/WARN] [io.netty.channel.DefaultChannelPipeline]: An exceptionCaught() event was fired, and it reached at the tail of the pipeline. It usually means the last handler in the pipeline did not handle the exception.
io.netty.channel.unix.Errors$NativeIoException: accept(..) failed: Too many open files
[07:05:13] [Netty Worker IO Thread #17/INFO]: [/127.0.0.1:51924] <-> InitialHandler has pinged
[07:05:13] [Netty Worker IO Thread #17/ERROR]: java.io.FileNotFoundException: plugins/StelyMotd2/config.yml (Too many open files)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at java.base/java.io.FileInputStream.open0(Native Method)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at java.base/java.io.FileInputStream.open(FileInputStream.java:219)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at java.base/java.io.FileInputStream.<init>(FileInputStream.java:157)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at net.md_5.bungee.config.YamlConfiguration.load(YamlConfiguration.java:77)
[07:05:13] [Netty Worker IO Thread #17/ERROR]: at net.md_5.bungee.config.YamlConfiguration.load(YamlConfiguration.java:71)

Thank's for answers.

[surreal stump]

your issue is a linux config thing, you have to increase your max open files

[reef fulcrum]

I mean

either you failed to up the amount, have you checked?

or, you have something spewing resources

see lsof

[agile lotus]

That's what I'm trying to do but I can't my number of files in ulimit -Hn / ulimit -Sn is 4096 and 1024 I can't increase it.
If you know how to do it would help me a lot 🙂

I did about 5 tests with different files but I can't increase this limit...

[marble kelp]

@blazing ocean redis is the way I use it but any of the others should work (although plugin messages come with the obvious limitation of requiring a player online)

[blazing ocean]

Just set it up with redis, works like a dream 🙂 Thanks!

[grizzled zealot]

Hey all, I'm having some issues after a big server failiure. We ran out of storage and it caused a lot of issues, we've got the servers up but are now running into more issues. People are joining and being assigned new Mojang UUIDs and their IPs are showing up as Localhost

I've got no idea where to start in fixing it, all the server configs should be the same as before the issues, they're all on offline mode with bugneecord enabled

I restored a backup of the Bungeecord setup to one that is known to work and it's on the latest version, though the servers are still 1.16.5

[reef fulcrum]

mojang UUIDs come from, well, mojang

offline mode UUIDs are calculated automagically from the name

so, stuff like that would either imply some setting got flipped back to default due to the config file being lost somewhere, or, you have a janky plugin

[grizzled zealot]

Ahh I see

Okay, let me have another look

[tidal musk]

I cant find the download for the bungee version of the plugin? Where is it?

[sand prism]

i try with velocity, same problem

[brisk fox]

Thanks for reporting back, that was useful, but unexpected.

[lost river]

There is just one version that works on bukkit, bungee, and velocity.

[sand prism]

i think is something else, related to the server in question that block connection in some way, what pl do you have on your server that give you this problem?

[brisk fox]

AdvancedLogin*, BungeeGuard, ChatControlRed, Citizens, CitizensGUI, CoreProtect, CustomCommands, DailyBonus, DiscordSRV, DropEdit2, EliteMobs, Essentials, EssentialsChat, EssentialsSpawn, GadgetsMenu, GCore, HeadsPlus, HolographicDisplays, HuskHomes, InventoryRollbackPlus, ItemFrameShops, LagAssist, Lands, LibsDisguises, LuckPerms, LWC, mcMMO, Multiverse-Core, Multiverse-NetherPortals, PlaceholderAPI, PremiumVanish, ProtocolLib, Shopkeepers, SimplePortals, SkinsRestorer, SuperChairs, TAB-Bridge, Vault, ViaBackwards, ViaVersion, WGEF-Reborn, WorldBorder, WorldEdit, WorldEditSUI, WorldGuar

[sand prism]

you do recent change when the problem start?

[brisk fox]

At all, whats you plugin list?

[sand prism]

citizen, hp , libs and mcmmo are the more relevant, the other are different or not connected to this problem

[fathom jetty]

offline server?

[brisk fox]

Subservers are always ofline, as far as i know. Are you pointing out AdvancedLogin?

[brisk fox]

An actual complete list would be helpful.

[fathom jetty]

the login and skinrestorer, yes 🙂 why else would you need them if the proxy is online and forwarding is properly set up?

[brisk fox]

Because of shared accounts in first instance, and people aiming to change skins easily. 💁‍♂️

Back to the point, those plugins removal meant no change at the issue.

[open brook]

Hello, how can I hide players in spectator mode ? For example 10 players online, two of them in spectator, but will still show 10... Waterfall.

[surreal stump]

maybe with a plugin that count the spectator and take them from the count

[open brook]

ok

thx

[gritty fossil]

is it okay to start server with this flag... allowing 1.8.9 players to connect to server
i know i know, 1.8.9 is old but players wanted it 😛

[reef fulcrum]

that flag doesn't "allow" anything

it just tells waterfall to not use cached exceptions so that the actual error can be displayed and you can see what was wrong

something sent an oversized string

[gritty fossil]

😅 resourcepack and itemsadder

so ya.. thats an issue

[tidal musk]

hi, does velocity protect itself from attacks in any way? if not, which plugins should I use?

[royal hawk]

Define protection
Velocity is hardened against DoS attacks moreso than Waterfall
But that’s roughly where the story ends.
It’s still just software and that means it can be overwhelmed somehow.
Usually worst-case scenario I’ve seen is a bit of lag for the existing players and new players being unable to connect temporarily due to a soft-api ban by mojang following an attack

[tidal musk]

I care about protection against bots and exploits

[craggy trout]

Velocity handles better than waterfall those attacks, but is not built into that you will need to use an largest solution like an Layer firewall or an reverse proxy.

[reef fulcrum]

bot attacks, velocity is better with resources but doesn't cater specifically to tryna solve attacks, that's a whole set of extra effort which is best focused on another piece of software rather than tryna marry the two together, marketplace competition and all...

exploits, velocity is generally hardened against that type of stuff better in the first place, and generally has a much better and cleaner system making it easier to often apply fixes where viable

But, ultimatly, it's the amount of resources and the size of the pipe that you've got

hence why there are many services willing to sell you a bigger pipe and a filter mesh for a small fee, e.g. TCPShield, etc

[craggy trout]

https://github.com/realDragonium/Ultraviolet

Or even haproxy with Cloudfare tunned

[reef fulcrum]

not a bigger pipe

[craggy trout]

Uh, well TCPShield is free only for 25 players.

[reef fulcrum]

Well, yes.

Because they offer a bigger pipe.

[craggy trout]

Also, the ping increases a lot

Even if you're near to OVH

[reef fulcrum]

What you're talking about there is putting in a small app which will deal with some aspects before they hit the proxy, but, it's not at the level of what services like TCPShield do

[craggy trout]

I tried that reverse proxy with an VPS of OVH and all the trafic got filtered.-

Well yes, tcpshield is a higher level.

[reef fulcrum]

Which is literally what I'm saying.

Yes, ligherweight "solutions" exist which will ward off some specific type of attacks

[craggy trout]

But any server could host its own reverse proxy on a machine that has good BW, and easily mitigate large attacks plus adding iptables firewall.

[reef fulcrum]

a proxy which needs to deal with stuff authenticating connections is gonna eat more resources than something which just attempts to act as a passthrough

[craggy trout]

[craggy trout]

ye xd

[reef fulcrum]

which is why stuff like UV will work for a limited amount of stuff

OVHs DDoS protection will mitigate some other stuff

but, it's basically the cheese slice model where you're only really mitigating the cheaper attacks for the most part

thankfully, most people don't care to invest too much in, but, those solutions are basically weak against anything which really cares to attack you

[craggy trout]

However, most attacks are very bad, and even an "antibot" can easily mitigate them and dead proxies.

https://javapipe.com/blog/iptables-ddos-protection/ I spent time searching and trying and this blocked most of trash crashers.

[open brook]

I could not solve this problem, what is your advice ?

[reef fulcrum]

at the end of the day the solutin is gonna be "find a plugin which does that"

Waterfall doesn't really care to track what gamemode your players are in

[craggy trout]

[acoustic jewel]

Is it also possible in Waterfall to monitor plugin performance like /timings in paper.

[reef fulcrum]

using a profiler, yes

theres no concept of a main thread or a timed tick loop in waterfall, and so timings doesn't really git

[agile lotus]

Since no one can help me increase my "max open file" limit, I give up.
Goodbye

[reef fulcrum]

This channel is for support with waterfall, not linux

I tried providing support and spent an hour basically repeating myself over and over

Might I suggest hiring a system administrator or something, because I'm not gonna spend another few hours trying to provide support for something is

1. Nothing to do with this channel
2. generally a waste of time because as I first suggested, you should probably work out what is eating so many file handles with lsof

[agile lotus]

Yes but you are a big developer for you this kind of thing is done in 5min because it is your field unlike me who does not work at all in that

[reef fulcrum]

I'm not your system administrator

I've already told you what to do, even linked you an SO article which went over even more steps if that doesn't work

I'm not gonna walk you through something of which I've already linked you step by step instructions for, nor am I going to provide support when the time I linked you to said step by step articles, I ended up spending the next hour repeating and answering information from said tutorial

Go hire a system admin, am sure you can pick one up for basically nothing

[agile lotus]

Tes

[sly hill]

hello people ^^

ok, i have PaperMC runs from about 8 years. The server is stable, in 1.18.2 whith plugins. Never crashed ^^ . So my curiosity says to me " what is WaterFall? ", and i try it. I have installed it, configured, and it works perfectly to connect to it whith a forge client. some plugins works in multiplayer like minimap ... Great! But, when i install a plugin like 'Mekanism' , it works in solo, but not in multiplayer. ( objects are in game but disapear instantly when put on the ground ). If i whant to put it in the server/mod , Waterfall says [09:26:34 ERROR]: Could not load 'plugins/Mekanism-1.18.2-10.2.0.459.jar' in folder 'plugins' org.bukkit.plugin.InvalidDescriptionException: Invalid plugin.yml , and dont load it. Is anyone have a tip to make plugins works in multiplayer from WaterFall linked to Paper ? tx.

sorry for my english from france.

[young rampart]

Waterfall doesn't have Forge/Fabric support. Client mods that only affect your client, like a Minimap, will still work, but anything that is server level will not work