#waterfall-help
1 messages · Page 56 of 1
can you show some of the UUIDs?
bungeecord: true ?
I mean, I asked you multiple questions and you answered one
Like, Please actually answer the questions.
does not match
9f33db62-f827-32e9-a7ac-600745fd2b69 is the new one
4caf38e4-39d0-4b04-88e7-444a7794e3fd is the old one
first one is offline mode
was the offline mode the one in the whitelist file?
no
oh
makes sense
now
I'm an idiot
that's my bad
Ok, I just enabled it
it worked
thank you!
Ok I have the same problem again
bungeecord is true now
nvm
I found the problem
I have another problem
when I kick someone. I'd like for it to go to the lobby. Instead, it kicks from the whole server. I use litebans and enabled bungeecord in the config
you will need a plugin to do that
so you're saying I need a plugin so another plugin can kick them to the lobby?
ask Litebans support then?
Out of curiosity, does anybody here use jvm flags for their bungee / waterfall? And how much memory would you typically allocate
Look pinned messages
Is it possible to restart waterfall from the paper server
Waterfall has no in-built restart mechanism
I just mean stop
You could hook something like that up yourself, but ofc all connections would close too
No
I already have the restart script. Its just that everyone is timed out permanently for some reason and I cannot access the bungee server from my phone
Oof
I mean, if the connections are stalled, you won't have a connection to send the message over anyways
Damn looking at it it got an out of memory exception
am i able to replace bungeecord.jar for waterfall and have it run fine?
obv ill change the startup cmd but thats it?
generally, yes
wdym generally
99.99999% of the time it works fine
ok
any of y'all know why i started getting this error while loading Waterfall? https://paste.insprill.net/xagetudeku
happens on Java 11 and Java 16, and on builds 414, 415, 416, 417, and 418 (the ones i tested)
generally means that you're missing some other library
libxrender1 is generally the big missing one
beyond that, google that library it says it couldn't load and there is a good few answers out there
purging java and reinstalling seemed to do the trick. weird
actually no, when i start any java application from the servers terminal it can't find a bunch of classes, but if i start it though ssh it works fine 
it shows the same on both
does waterfall support forge 1.12?
Yes.
It is possible to use Aikar Flag's in Waterfall as well?
Yes, but, don't, see the pins
I saw them, but I wanted to affirm it before doing anything
ones in the pins are for waterfall, one on his site is for the server
Perfect, thank you (I don't want to mess up something xD)
Is it just drag and drop If I want to switch from bungeecord to waterfall? any steps I need to take before that?
Just replace bungeecord with waterfall and run it
thank you
Is there any way to make it we can type /hub to connect to hub server or for example /creative, /survival etc...?
ya, there are many plugins that do that, the /lobby or /hub part
yes found slashserver plugin!
thank you
the only problem the slashserver has is that it always sends a message "Connecting to Server" and if I remove that message from config, it will send 1 line of empty message to players
removed that line from config and this 1 line always shows up when switching servers
TELEPORTING: ''
ALREADY_TELEPORTING: '&cAlready teleporting you to a server'
Survival: 0
Creative: 0```
Any idea? anyone having this issue?
just add something for it to say
but I don't want it to say anything when switching servers XD
U prob gonna have to find another plugin or look in its config file to make it not send a message
does waterfall need restarts?
like does the performance degrade if i dont restart it periodically
Just like how there are aikar flags for a paper server, are there any flags I could use for a waterfall server to increase performance?
@tidal musk link me that plugin nvm I found it.
I'll ping you again soons.
@exotic remnant Check on the pins here
I see this message but it was in 4/13/2020, is it still an up-to-date startup command?
yes
My waterfall server is 1 GB
so do -Xmx1G
How much ram would you recommend allocating for the Xms and Xmx values?
well 1G host?
The server is on 1 GB ram yeah
oh then prob go with 512
Alright, thanks
gonna need to leave room for caches/container overhead
^
Gotcha
though WF wont be hitting FS cache though
is that bad?
How much ram usage would be considered "rough"?
You won't be able to see ram usage if you're using those flags your panel/whatever will just show the server as constantly using 512MB
oh
But if connecting to the proxy starts to feel "slower" than usual under stress it's a good sign you're actually using it all
Alright, would 2 GB be too much for a waterfall server? If not, what would you recommend as the Xms and Xmx values for it
Also, my waterfall server is currently on Java 11 and my paper servers are on Java 16
Should I switch the waterfall server to Java 16 as well?
What is waterfall ?
Like bungee cord?
Waterfall is a fork of bungeecord designed for better performance
@exotic remnant Unsure if Java 16 will "help" Waterfall but there's no reason not to try it.
Generally the amount of ram you want is kind of correlated to how many players you need
If your host let's you upgrade, try a little bit and see how it goes
If you want to push up the ram, do so
Alright, thanks so much for the help
I patched the plugin to not send a message if TELEPORTING is set to blank. https://github.com/DeJayDev/SlashServer-Bungee/releases/tag/41556cb
Pls let me know if its broken.
Would 1G be good for a 2 GB waterfall server?
hi
my server water fall was doing his job and pining normali
and suddenly it's says killed
any idea ?```
[06:10:29 INFO]: [/ip:54158] <-> InitialHandler has pinged
Killed```
but my vps have 2 other servers and OS decided to randomly kill the bungee ?
and what is dmesg ?
linux util which pulls in the kernel logs
generally means you ran out of memory and the proxy was just what the OS killed to have memory to continue
May 17 06:10:30 debian kernel: Total swap = 0kB```thanks zzzCat
That worked flawlessly. thank you very much. you did a work of a nation here ❤️
That's Swap... That's not RAM.
Also you haven't enabled swap either.... I suggest you do that too
i mean, just seeing that in there means that the OOM dump stuff happened
Anyone know a good tutorial for waterfall
use the bungee install guide
what
yeah so when I do bungee cord
since my server hosting locally I can put in the ips Cus I use Ngrok pls hel
you use localhost to connect to the servers if they're on the same machine
There is some level of expectation that you understand basic networking as we generally don't help with it
what’s localhost
google it
no
I just looked at a tutorial for bungee cord would that same thing apply for waterfall
Yes, that's why i told you to look for a bungee guide
so would the breakdown be a good tutorial
only tutorial we support + recommend is the bungee install guide on spigots site
oh okay
but could you help me with sharing my server with my friends without port forwarding
okay
Color codes seems to be broken on this version. but thats okay no big deal. I can remove the color codes. just wanted to let you know.
This build worked great! no more empty 1 line message when switching servers. thank you again
Urg legacy colors
@tidal musk I can fix it in a few hours, I'll just ping you with the build.
Bungee uses md5 components
Which are most definitely not the components you use in Paper (Adventure)
tbh my solution to specifiednulls problem is just gonna be include adventure and adventure bungee
Hey! no worries! didn't want to bother you. just wanted to let you know. but thanks! appreciate it
What will be the recommend java version for WF when mc 1.17 is released?
I doubt anything will change there
Does travertine performance degrade if i dont restart it periodically?
if plugins are well developed there is no reason
i run waterfall for multiple weeks without any issue
hello I have a strange issue, my websites work fine but my Minecraft servers are unreachable all of the sudden.
People that are connected rn can still play but people cannot connect atm.
The webserver with the same domain and on the same machine work fine and the ports are open as checked with a port scan. This happen all of the sudden which is very strange.. I do not know what it might be :/ if I connect using just the IP it shows up in the server list though
Would probably be helpful to get a better idea on your set up. Like are you running every thing through another proxy like HAProxy, is the Webserver also running waterfall, or is everything on the same host. What do your logs say.
Alright so the issue is cloudflare from my tests but it's super odd.
Basically my domain is proxied through cloudflare and has been for many many many months and I use a SRV record for Minecraft which worked perfectly fine for months. Now it just doesn't for some reason and I do not know why. Connecting with just the IP works and disabling this proxy stuff works too but then I do not get SSL or other security stuff from cloudflare / cache savings from them. :/ ?
Probably because you do not use 127.0.0.1 for your servers (if they are on the same machine as your proxy) which causes high ping.
im already doing
Probably a priority or weight issue in cloudflare sounds like it's hitting your A or CNAME instead of the SRV.
And obviously only HTTP/HTTPS is proxied in cloudflare which means new connections will just fail.
Srv record for port 25565, worked perfect for many months, even new domains I added for a new server worked fine with proxy and mcsrvstat would show the real IP. Now all of the sudden it doesn’t which is strange asf
repositories {
maven { url = 'https://papermc.io/repo/repository/maven-public/' }
}
dependencies {
compileOnly 'io.github.waterfallmc:waterfall-api:1.16-R0.5-SNAPSHOT'
}
I don't know what i do wrong. He can't finde waterfall
Heres the fix:
Add a NEW subdomain, maybe name it connect
Turn off proxy for this subdomain only
Point the SRV record(s) to this subdomain. They will work just fine now.
example.com srv record points to connect.example.com and works fine when using example.com on minecraff
How do I make it so when the server goes offline, it sends them to the lobby server and not kick them from the server saying "Server Closed"?
use a plugin e.g. my KickInfo one
is there any performance buffs for using waterfall > bungee?
Some. But if you can give up BungeeCord plugins (and accept a smaller ecosystem) Velocity is the highest-performance proxy out there
I like that waterful uses Log4J logging system. much easier to track and see latest logs.
Hello, Will Bandwidth got limited cause "internal exception java.io.ioexception an existing connection was forcibly closed by the remote host" happen?
that means that a connection was closed externally in an unexpected manner
how about this " NativeIoException: readAddress(..) failed: Connection reset by peer"
same thing, closed in an unexpected manner
if i increase server bandwidth will improve the problem of this?
maybe?
ok thanks of answer
Is it worth using waterfall on java 16 on a public server or not yet?
There shouldn't be any ill effects if that's what you're asking...
I have mine running on Java 16 and have had no issues at all
Thank You
Is there a way to disable these messages?
[11:18:53 WARN]: [/192.168.4.178:52524] <-> InitialHandler - read timed out [11:19:23 WARN]: [/192.168.4.178:52526] <-> InitialHandler - read timed out [11:19:53 WARN]: [/192.168.4.178:52535] <-> InitialHandler - read timed out [11:23:54 WARN]: [/192.168.4.178:52565] <-> InitialHandler - read timed out [11:24:09 WARN]: [/192.168.4.178:52566] <-> InitialHandler - read timed out
or is there any way to solve these errors?
nvm
I see a way to disable it
in waterfall.yml
It's always funny when you're trying to find something and then you call or chat with someone to find out where it is and when they answer, you already found out where it was.
I still have the same problem actually
it says red stuff
Not very helpful... send a screenshot of the error please
Ok Wait
this
well, do as the message says
How
edit your bungeecord config to enable it
do i set to false our true
enable = true
ok
[12:35:13 WARN]: **** FAILED TO BIND TO PORT!
[12:35:13 WARN]: The exception was: java.net.BindException: Address already in use: bind
[12:35:13 WARN]: Perhaps a server is already running on that port?
Auora i connect on the bungeecord ip now it works tnx
How much RAM do you recomend for a server running 60-100 players on daily
(Waterfall)
1-2G?
What would be about the max amount of players you should put on a single waterfall instance?
it can handle a few hundred just fine but at that point you should generally be looking into load balancing anyways
Working fine at 1200 players
I cant understand wich tho
Nor can we, you'll have to start yanking plugins out basically
Just remove the last plugin you added for a test if its not him move one by one
I just rediscovered waterfall and I am running into some minor issues on bungee.
Which features does waterfall have that bungeecord does not?
Nvm. I'd already prefer the log file system lol. Also DoS mitigations, I've experienced DoS
But is there any article that I can read that is up to date which tells me features.
Specifically the api changes
Waterfall doesn't do DoS mitigations
Its just a proxy, get a host with a proper firewall
afaik waterfall does a little of dos mitigation
Hello, has anyone seen this error before when trying to use ProtocolLib in Waterfall?
As any developer here that looks at this should be able to tell you- Protocollib was made for the spigot api and does not run on bungeecord/waterfall. There are plugins that do run on bungeecord and provide functionality like Protocollib, but using something like that is still a very bad idea in my book
Not sure what to say then. Protocollib is a dependency for ServerListPlus which runs on the BC/Waterfall side.
“The latest release of ProtocolLib is required for most features to work on Bukkit/Spigot/Paper servers.”
No mention of Bungeecord
Again, not sure what to tell you. You left off the sentence just before the one you quoted and the author of Server List Plus said it is a dependency needed for gradient text to work. Which doesn't make sense to me since it is built into 1.16
You strike me as the individual who doesn’t read plugin description pages. That aside, if you refer to rgb as “gradient” support then you should know that bungeecord supports that natively without modifications.
If you don’t believe me for whatever reason:
Here is the line of code for the dependencies of SLP on bukkit/spigot/paper:
https://github.com/Minecrell/ServerListPlus/blob/master/Bukkit/build.gradle.kts#L54
And here the line for bungeecord:
https://github.com/Minecrell/ServerListPlus/blob/master/Bungee/build.gradle.kts#L39
You strike me as an individual that is an arrogant asshat that just tries to belittle and put others down by trying to act like a know it all prick and you can GFY! I certainly did read it so you may not want to assume and try to act like a know it all prick. I am perfectly capable of reading just fine. I am telling you what the author told me when I asked about this which didn't make sense to me either. I am not some shit head kid who has no clue what they are doing. I have been running servers for many many years. So check your attitude at the door. Do you understand son?
Also, if you paid any damn attention to what I said about the native gradient in 1.16 in my last message. You would have noticed I even stated "Which doesn't make sense to me since it is built into 1.16".
I apologize because I missed that last part, though my point still stands. About the rest, I just couldn’t care less.
How can i download paper for 1.7.10
I will never understand why people still cling to that version
Does forcing default server make people go to that server no matter what server they logged off on?
or
yes
alright cool
I have a lot of people connected to a server but I want to force them onto a server im adding when they relog
You’ll probably need a custom plugin for that
force default just uses the priority list
hm I need to find a bungee whitelist plugin
bruh im gettin rekt for some reason
I added 127.0.0.1:serverport for the bungee server
online mode false; bungeecord true
and still cant connect to it
I've been getting a lot of overflow in packet detected! A string was longer than allowed. Since updating to the latest build, would anyone be able to help me please?
More info about your setup and client(s) maybe? That’s just too little information to say anything
The issue was only prevalent for geyser / bedrock users, rolling back to build #415 fixed it
@desert robin hi does this ring bells?

To note: there were stricter limits on packets imposed on waterfall not too long ago
Hello! Can't say it has been.
But I also haven't been able to be super active in support channels.
Is there a Waterfall debug flag we can use to determine which packet causes it? Or is there a stack trace attached to that error?
Actually maybe Floodgate is triggering it by adding to the handshake packet?
There is a launch flag -Dwaterfall.packet-decode-logging=true
Thanks.
Will try with Floodgate and the latest build.
Come to think of it... has Velocity recently implemented a similar change?
This was taken from velocity
I seeeeeeee
From: "Five (Xer)" <admin@fivepb.me>
Date: Sat, 30 Jan 2021 18:04:14 +0100
Subject: [PATCH] Additional DoS mitigations
Some stricter length checks and cached exceptions to withstand more illegitimate connections
Courtesy of Tux and the Velocity Contributors. See:
https://github.com/VelocityPowered/Velocity/commit/5ceac16a821ea35572ff11412ace8929fd06e278

Velocity 1.1.5 mysteriously made Floodgate 1.0 stop working, but since 2.0 worked fine and the Velocity implementation for 1.0 is hackier than any implementation of anything I'd ever seen, I left it be and just designated people to use 2.0.
spigot did recently added a length limit on the handshake
When I say 2.0 I mean Floodgate 2.0.
Ah i see
Waterfall 419 with Geyser and Floodgate and no forwarding and Paper 691 works fine for me.
Setting up IP forwarding and sending the Floodgate data to the backend also works for me.
Is there a setting enabled on fresh Waterfall instances and not old ones? @royal hawk since your name
hmmmmmmm
Well, this has been enlightening either way.
I will continue to investigate as I can.
Thank you both. Hopefully I will either be back or won't be back.
What's the max amount of ram a waterfall server should have?
Rather, the max it can handle
I would assume you could allocate all of the 64Bit java space, just around 16 Exabytes.
But in all seriousness, if you have plugins that use database things I would go with a max of 4gb. Anything above that is total overkill, and this already is kind of crazy.
Alright
So if it ever uses 4/4 GB ram, I just use something like redis bungee instead of getting even more ram?
I am back to say that Waterfall 421 does produce the error.
But apparently Floodgate 2.0 doesn't have that error so I think we're just going to push for that being available for 1.17.
I'm getting this on a dev server after updating to git:Waterfall-Bootstrap:1.16-R0.5-SNAPSHOT:e0e464b:421 yesterday. I'm not runner geyser/floodgate though and it's actively preventing me from accessing one of the servers on the network.
I'll be going through and removing each plugin to see if any of them are an issue but I wanted to chime in since it doesn't seem to be an issue just with me.
Could anyone help me convert my server to bungeecord?
Logs?
latests bungeecord commits added a lot of string length limits
i don't know if there is any trace printed when erroring
There is when you se the jvm flag to expose debug messages
ye good idea
Hello, I need a .json file to add to the pterodactyl panel. Where can I find it?
paper does not provide official pterodactyl eggs, but you can find one here: https://github.com/parkervcp/eggs/tree/master/minecraft/proxy/java/waterfall
Thanks ❤️
Im getting timed out from my servers, how can i fix it?
.895
23.05 12:52:50 [Server] INFO [_DJDan] disconnected with: The lobby server was not available, so you have been kicked from the network
23.05 12:52:50 [Server] INFO [/165.120.185.71:51445|_DJDan] -> UpstreamBridge has disconnected
that's not a waterfall message afaik
you'd need to check your logs as to why it's not available, there is no magical fix here
what does this do timeout: 30000
oop, sorry. Here they are. I just started the server, connected to the main server, tried to connect to the only other server on the network, disconnected, and then closed the Waterfall server
https://pastebin.com/5WBDbA38
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
launch Waterfall with -Dwaterfall.packet-decode-logging=true
On it
I've added it to the startup command so that it reads java -Xms128M -Xmx1024M -jar waterfall.jar -Dwaterfall.packet-decode-logging=true
and the error is exactly the same. I'm not seeing anything new mentioned in the log about a debug mode either?
ah, got it. Sorry
https://pastebin.com/pbvedQac
The error has changed to
[18:27:23 WARN]: [/172.18.0.1:60984] <-> InitialHandler - overflow in packet detected! Cannot receive string longer than 1020 (got 1333 bytes)
bungee supresses those exceptions, woo
But, tha can't be waterfall, nowhere do we have a limit check for 1020
Hmm, that's weird then. Thanks for the help though <3 I'll try removing plugins and see if any of them are causing the issue.
wa
Yeah, I'm lost too lol.
I literally just made it print the stack trace when one is passed to the InitialHandler, nothing there will print rando new lines
Hmm, let me make sure that my server is working ok.
Ok, sorry, I'm incredibly dumb and didn't make sure that the file was named correctly. How having the W capitalized and not lowercase would result in endless blank lines is beyond me though...
re, wa
apparently you're not running with the flag from before?
So, that exception is irrelevant because of the fact that you're apparently missing that flag
I noticed that, I removed it while debbuging. I've added it and just generated a log. Sory.
And here it is https://pastebin.com/5WLtMFCi
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
I have a little problem, forge 1.16.5 won't connect to waterfall, ping connexion wont even made.
Uh, you two might be about to kill me.
It looks like I had my proxy pointing at my proxy...
Only found out because I was doing some port juggling and then got an error saying the proxy couldn't connect to the one port I hadn't changed. Sorry for wasting your time, <3
Forge 1.13 and newer is currently not supported
Well that explains a lot
Anyone?
most of the memory for waterfall is gonna be off heap
hard rules like that just doesn't work
generally the limit is on the CPU and stuff installed on the proxy rather than the ram
"so much" is subjective
Never seen it use "so much" here, but, it can boil down to many factors, CPU usage being a bad metric in part, down to things like plugins and even the OS you use
Is there a good way to limit the max speed of each connection? Several players on my server are using too much network, causing delays for everyone. I have used iptables before, but it is not easy to use.
“Too much network” it seems like you’re trying to limit what players can do- if you limit it on the system level the players will time out/ be disconnected. The traffic is there for a reason and if you feel like you must limit it then please do so on the server level
Using waterfall it should reduce the amount of bandwidth due to packet aggregation
If you don't notice that you can try a plugin I have made
whats the best way to prevent easymc and alts?
Unlike versions before 1.7.2 netty packets are framed and compressed individually so there is little to none saving to be had.
What may help is lowering the compression threshold to 64
You can do a round-robin channel flushing but that just makes it equally worse for all players
I have measurements made, and low speed connections improve a bit
normal connections get like 1ms extra delay
I can’t imagine anything besides these two that don’t negatively impact gameplay
the amount of bw saved is about 4/5
80%
I understand
doen't seem feasable
the concept is packet agregation
Minecraft generates dozens of packets by second
each packet has a tcp footprint
about 56 bytes, I think
to send 11 bytes it spends almost 70
If you’re aggregating chunk or lighting packets I can understand that you’d save a bit.
But yes indeed this impacts gameplay
it is not a bit
do you have mesurements on your server
average by amount of players?
mb/s
I’m not saying it’s stupid, I’m sure it works, but it breaks some small nuances, how does the rendering fair and how does entity movement check out?
not bad
it is a bit more complex than this
but it constantly measures the relation by time and amount of data been packed
to minimize the lag vs optimize the bw
of id a big amount of data is been sent to the client, no delay is made
I’m sure there are some who can really use a clever packing to save bandwidth, but I don’t think those are many ¯_(ツ)_/¯
I’m not sure how much the split chunk packets in newer versions impact this
You can still aggregate that to a full chunk packet
And save some bits that way
let me grab some prints I have made in my study
But I’m not sure
I’m still kind of salty that the new JoinGame packet is uncompressed at least 500kb
seems i have deleted it when I reinstalled my server
yes the initial packets are quite heavy
much more then 500kb
when a player joins
The 1.17 impl has uncompressed around 2.4mb
I bet MC could compress it a lot
That’s a lot in burst packets
Yes I mean- Network NBT is horrible
And json over string is even worse
When I first saw the dimension and biome registry I was like WTH IS THAT
on the left with my plugin
150ms window vs 20ms window
both windows full capture have the same amount of time, the time i referred above are the times visible
at the right you can see that to send 6 bytes, it was used a pack with total 50 bytes on the wire
There is a minimum size to packets, so it needs to basically pad the size of the packet to meet rhat
donno about the minecraft protocol itself, but we see alot of packets with 11 data bytes when the bytes on wire (with tcp headers) make it around 70 bytes
and it is quite network costy
lets imagine this, to send a 1GB file to your friend, you rather send it with 11 byte packets or 4 to 8kB? 😉
now with a realtime game, ofcourse the lag is very important
that is why it is automatically balanced in realtime
to give best performance possible at all levels
MC has no framing, however
1 packet = 1 packet
framing is also a nightmare if you wanna try and keep stuff fast
minecrafts protocol is known to be blazingly inefficent, but, tbqh, so is gaming in general as you don't need to send some huge packet to say "i interacted with this block"
ideally you'd frame stuff, but, that means adding an extra encapsulation layer, it also means tryna consilidate flushing so that you're not keeping stuff held back and tryna avoid sending stuff "too happily" would which negate the gains
yep
the hard part is the fine tunning
I have submitted the project so others can try it out and give they feedback https://www.curseforge.com/minecraft/bukkit-plugins/proxytel
it is unther acceptance
just send the github link 👀
haven't created yet
Randomly a couple of hours after a restart this happened and would not allow players to rejoin until i restarted it again when i woke up
[12:33:36] [Netty Worker IO Thread #7/WARN]: Event PreLoginEvent(cancelled=false, cancelReasonComponents=null, connection=[/**.**.**.**:59373|Nexus_Spider] <-> InitialHandler) took 3,076ms to process! [12:33:36] [Netty Worker IO Thread #7/WARN]: [/**.**.**.**:60999] <-> InitialHandler - NativeIoException: readAddress(..) failed: Connection reset by peer
reset by peer means that something external killed the connection
What could do that. It was like this for like 10 hours until i woke up to reset it
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
This was all the code related to it
The servers just disconnected from the waterfall
Umm, any ideas on this? I am not using a cracked account nor have I ever used an alt gen.
18:32:14 [SEVERE] [Zumux] <-> ServerConnector [Hub] - encountered exception: net.md_5.bungee.util.QuietException: Server is online mode!
18:32:14 [INFO] [Zumux] disconnected with: Exception Connecting:QuietException : Server is online mode!
18:32:14 [INFO] [Zumux] -> UpstreamBridge has disconnected
18:32:14 [INFO] [Zumux] <-> ServerConnector [Hub] has disconnected```
I have had 2 other people test it too and the same this occurs
Waterfall/Bungecord requires the server to be in offline mode
but you can set bungeecord: true in spigot.yml to allow the proxy to send the real uuid and ip to the game server
It is set as true
you can set online-mode to false in server.properties then
as long as your proxy is in online mode and your game server inaccessible from the internet, no cracked client will be able to join
Ok so I did that and that worked. So what exactly do I need to make sure of so no cracked clients can join? Also, why wouldn't it work the other way?
if your proxy and your server are on the same machine, you can simply set server-ip to 127.0.0.1 in server.properties
that way the server will only listen on the local interface
if they aren't on the same machine, you can either use firewall rules or plugins for this purpose
and I'm not 100% sure why the server has to be in offline mode, but it's probably because the proxy can't easily spoof a real minecraft account
ok thank you
Here's the startup command for my waterfall server
java -Xms1024M -Xmx1024M -XX:+UseG1GC -XX:G1HeapRegionSize=4M -XX:+UnlockExperimentalVMOptions -XX:+ParallelRefProcEnabled -XX:+AlwaysPreTouch -jar KolakZCore.jar
But, the ram usage is at 1.41 GB / 1.98 GB
Is that normal?
Yes
You're specifically telling it to allocate itself all of the ram
It'll reserve 1024M (1.0GB) and then a tidge bit more for the container itself.
Can you change the message that gets sent to everyone when the proxy is restarting?
Alright
You'll have to change that in the code. To compile Waterfall, you need JDK8, git, bash, maven, and an internet connection.
https://github.com/PaperMC/Waterfall
oop. Well thank you for that. Just got a differnt way though haha
Glad to hear that
Any reason why my server icon wont show? It is called server-icon.png and is 64*64 pixels
It should show you a reason why it's upset in your logs then
There's no logs about it
Is it in the same folder as the waterfall.jar and the config.yml?
Is the version you have this issue with 1.16.x?
Yes, and yes
I never had the chance to check another version though
1.16 has a bug that prevents it form displaying server-icons randomly. There is no known fix. Try it on 1.14
If it displays there then its that bug again
Oh wait. If the bug is happening for me, would it happen for someone else too? I got my friend to check it aswell on 1.16 and he didn't see it either
Unfortunately I cant get to my computer right now, I'd have to check another version tomorrow
It might or it might not. There is no clear way to see the behavior on 1.16
Test your server with https://mcsrvstat.us
if the icon displays there then its your client
Nope, isn't there
It is not
There might be an issue with the color-profile of your icon then
Out of curiosity... Is there a way to change "1.8.x, 1.9.x, 1.10.x, 1.11.x, 1.12.x, 1.13.x, 1.14.x, 1.15.x, 1.16.x (Waterfall)"... To at least something like "1.8.x-1.16.x (BungeeCord)"..
you can do that in waterfall.yml, don't know what it is for bungeecord
That's for changing the versions supported... What I'm after is shortening it...
can change the versions, we don't support replacing our attribution, however
Ah ok
it lets you set the version string, so you'd just put 1.8.x-1.16.x in there
Oh ok, so I can literally just set it to 1.8.x-1.16.x.
I mean.. I really only want people playing with the latest.. So just setting it to 1.16.* would also make me happy...
it's a string that you set it to whatever you want
it has no bearing on the versions that the proxy supports, however, you'd need a plugin if you wanted the proxy to say that it only supports specific versions of the protocol itself
Ah that makes more sense then
Yeah the servers will reject an outdated client so that is good
Can you tie together a sever on a host and a different server on an unrelated vps with wtaerfall?
Yes you can... I hope they're pretty close together though
Have done it before with all bar one server local and then the other in another country 😛
Looking good.. Cheers mate.
It's at 1.8 GB / 1.98 GB now, I'm getting a little worried
https://dev.bukkit.org/projects/proxytel you have ways to see it 😉
Hi.
>.... [20:49:36 WARN]: [/0:0:0:0:0:0:0:1%0:49571] <-> InitialHandler - overflow in packet detected! A string was longer than allowed. For more information, launch Waterfall with -Dwaterfall.packet-decode-logging=true```
geyser?
bungee added some limits to the size of some packets and apparently broke geyser in some cases
then again, some person who was having issues also said that they had the proxy configured to connect to itself, so, woo
lolz
Nothing like creating a proxy loop...
Any idea why my server icon wont show? It's in the right place, is 64*64 pixels, is called server-icon.png. There's no logs or errors about it. I think it's an issue with my proxy, it shows on the hubs linked to it fine.
does anyone know a good maintenance plugin for bungeecord? I tried one, but it was spitting out errors like mad
make sure that you reload or at least restart the proxy for server-icon, but, there is a client bug which causes the logo to sometimes not show
Where can i download waterfall? The Download server is offline or something
git clone https://github.com/PaperMC/Waterfall.git && ./waterfall b
What does the parameter b do? 🤔
My fault, I didn't thought it was git, I've read wget 😅
Huh... First time I've heard about keybase.. Seems like a pretty awesome idea.
you missed out on a couple hundred bucks then 👀
You what?
They gave out free cash
yeah
Welp
They added a wallet to keybase and basically did drops of cash, which where worth practically fuck all at the time, but, now, I got like £450 sitting there
apparently the 1k XLM that I got were worth 800€ at some point lol
Damn... That's pretty good
Yea, it's dropped a bit over the past while, but, I dont got ID and shit, so, is just sitting there
using any of the exchanges or whatever

I think I don't even have real information on those 👀
you only really need ID once you try to sell for real money, as long as you just trade between cryptos it wont matter
and why would you sell bitcoin anyways? it will only go up ¯_(ツ)_/¯
also there are ways to get real money without ID cough cash cough gift cards
Don't come to Australia then lol... You need ID regardless... Oh and the tax man will try and get you if you go between crypto too...
That's your fault
Well, your families fault
Don't do the crime if you can't do the time
runs
oh god, one of my old mates over there took his car to get some list rust spots welded up
his car became the centre of a murder investigation
I'd have assumed the murder victims would be the centre of any investigation 🤔
Well, it was in the middle of it all, apparently some guy just flipped and ended up deading somebody who was working on his car
"Two bodies, both stabbed"
"nah nah nah, tell me about the Skoda"
lol
Man that derailed spectacularly... Haha
Signed file sharing platform to murder investigations...
OH, SQUIRREL
Anyone?
.heap
-Xmx controls the amount of heap memory assigned to the JVM, this does not include memory used for java itself and the native memory used by java itself, as well as other libraries such as netty (for networking) and SQLite.
But it's almost at the max so should I get more ram?
reduce Xmx if needed
Most of waterfalls memory usage, outside of plugins, is off heap
It's at 1G, would 512M be good?
Yes
Also
I was told that I shouldn't put for than 4 GB ram on a waterfall server
If it gets a lot of players
and it's almost at 4 GB
What should I do?
Because redisbungee is outdated
waterfall itself doesn't use that much memory
plugins on the other hand vary
Can't really comment on specific cases
Just in general, lets say the server gets like 5 thousand players
Wouldn't I need multiple bungee instances to support that?
at that point you should be scaling instances horizontally
Is there an alternate for redisbungee to have multiple bungeecord instances?
Not that I'm aware of, most networks either don't bother or roll their own
So how would that work?
you run multiple proxies and use a load balancing technique, either an actual load balancer like haproxy or RR DNS
Okay, thanks
Specifically Floodgate (1.0). I changed some stuff in Floodgate 2.0 which (unintentionally) made the handshake data shorter, because the issue hasn't been reported (and couldn't be confirmed) with Floodgate 2.0.
Forgot to turn off the ping ^
Hi guys, can you send me newest waterfall .jar file? Looks like something is wrong with a download site - https://papermc.io/downloads
pins, and, #announcements
This is why I don't like gremlins
i want the download for travertine
the downloads are down for the moment, you would have to build it from here: https://github.com/PaperMC/Travertine, just clone and run ./travertine b
or actually I think I have it here 1 second
no I don't, nevermind
I've added it to the pins
Hello I come to ask for help my server loses the authentication here are the logs https://pastebin.com/jkELXt4c I think this is a ddos?
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
the log doesn't say much
can you send your timmings
and maybe full operation log
full log from start until you get the first error
and anonymize the connected ips
looks like it
cause you are getting multiple IPs per second
Is there any way to remove Waterfall <-?
By editing the source-code directly: yes
Otherwise: Maybe but it’s gonna be difficult and require a plugin that digs into the internals of the proxy.
Paper doesn’t support removing that as it’s the only little piece credit they get, and that’s the case for many other projects too
what java version is needed to build waterfall by maven?
f3name plugin
Last I looked it was JDK8 but due recent changes by Mojang it could be changed to stay on the same version. So could be either 11 or 16.
But just look into the pom.xml
Thanks
irc bridge
Oh
How could I change some of the server messages? For example, the proxy restart message & /alert message
some of them are configurable between the .yml files
others, you'd need a plugin to hijack
oh, wait, wrong
So, yea, some are .yml
others you can change by extracting the messages file from the jar and placing it in the folder where the proxy is
some you may need a plugin
Okay
What is the best way to have flags for the latest waterfall?
I am currently using java -Xms128M -Xmx7074M -jar server.jar
see pins
Thank You
How much generally could a 4 GB waterfall server with 10 plugins hold?
most memory is offheap
512MB initial + 128MB per 100 players or something dumb was the "recommended" thing
and would 2G be good for Xmx and Xms values?
But, you'll bottle the CPU before you run out of ram
So I should worry more about cpu threads than the ram?
CPU, actual bandwidth, etc
Also, anymore than 4 GB ram is overkill for a waterfall server. Is there any cpu thread amount that would be overkill as well?
depends on far too many factors
Like, "ideal" setup is that you deploy waterfall on a few VPSes across a few servers and RR DNS them or something
ofc, that's extra costs, latency, and technical implications
sleep?
😆
how come I get this when trying to switch servers?
The server you are connecting to has online mode enabled.
Hello, I have a proxy in offline-mode and I would like to know if it was possible to change the ip forward to paper. So, if a premium player connects, I make a request to the mojang api to get the premium uuid and then set it everywhere on the infrastructure like that by doing a player.getUniqueId(), I get the premium uuid if the player is premium and the generated one if the player is crack
not without a plugin, but you simply using online mode is the better solution anyways
mixed UUIDs will can tons of issues
but the problem is that the admin team wants the server to allow cracks :/ and we really wouldn't mind making a plugin, in fact we can do it but we are trying to figure out how to make it
well people around here tend to not support piracy so ¯_(ツ)_/¯
oh it's piracy ? alright, didn't know, sorry about that
not buying the game is of course piracy, lol
My problem
I have tablist from deluxehub in lobby and waterfall copying him to survival, but survival has titlemanager and other tablist, waterfall kick me automatically
How can i fix it
Known issue on the tracker
Not all too much of a fix available as it's a long standing bungee cord flaw
just use BungeeTabListPlus...
With bungeetablistplus can i use custom tablist from titlemanager plugin?
or i cant
I won't remove TitleManager
if TitleManager does it right, sure
oh wait, I thought TitleManager does Title stuff, not tab list stuff lol
who even thought of that name -.-
no idea then, simply using a proper BTLP config instead of some random plugin would go a long way
Title manager made some changes which exposes a huge gaping flaw in bungee
Fixing that flaw implies changing API behavior potentially, which would likely break some plugins
what flaw?
So, there is no magical fix as title manager for some reason made dumb changes
also why even use a tablist plugin on bukkit if you run a proxy? 
Old scoreboard list on the proxy is cleared too late in the switch process
So, if server a and b have the same scoreboard name/teams, it blows up
https://pastebin.com/RArjQ0eV could that be the switch from waterfall version 418 to 422?
No
you are using invalid YAML in your PartyAndFriends config
wait actually why does it claim to tbe the metrics loader? 
yamllint says all is fine, and that three plugins wont work idk
check whatever file contains those strings that it mentions in the error
That screams that you fucked with the metrics yaml config
ah i could know the problem
waterfall.yml, try disabling tab list rewriting
best guess is make sure that you set up ip forwarding and bungee mode properly
as that screams that you didn't
Beyond that, no idea, something in your setup is bork
Please don't ping
Read the installation instructions for bungee
You need to enable online mode on the proxy, and ip_forward, and bungee mode in the spigot.yml config
beyond that, no idea
thats weird, I removed the plugin with the folder - same issue. But there stand 100% the same but the wrong line, 12 as line 6 so idk
just fix the config file lol
well find the one that has the issue
in the issue stand something with playtime, I removed the plugin with the folder, restarted server multiple times but same error.
the metrics config is in the plugins folder
it's not in the playtime folder
it's the metrics config, as we've said
which is in the metrics folder
Note that you have several plugins complaining with the exact same error
where is the metrics folder?
you could just grep for it lol
I am overwhelmed
Normally a plugin creates a folder for config etc. in my case the BungeePlaytime Plugin the BungeePlaytime folder. I removed both but same issue. But never heard of a metrics folder
metrics folder is in the plugins folder with bstats iirc
ahhh, that make sense, thanks 😄
ah, yea
Bayer: The command will be blue on all servers where the command is available
If you have that folder and its been touched quite recently, you got a damn old plugin lol
i want it to be blue all time
why would you want it to be blue if it isn't executable?
I think I was too late yesterday and put this in the wrong folder. 🥲
would never have thought of it and now i have learned that the bstats folder is the metrics folder
phoenix see his post in #paper-help and previous convo
they are using some shitty alias plugin thats not registering commands correctly
phoenix616 because when i do /menu and the MENU is red it is like bleeh
use a non-shitty alias plugin
on any server i go i can do /dffkhgisdyhsdguhsdhuds and it is still blue
why
so where is problem
that sounds dumb af
i will cry really
you should be doing stuff that actually helps players play on your server, not makes their life more difficult
i am trying to figure out for 2 hours
ok so at least how i can make certain commands blue for example /menu?
by registering it with the server
any plugin that provides commands properly will register them
and then they will show up in blue as a known command
well i use deluxemenu and i have there alias for help as menu but the menu still red
so idk man
contact the plugin's author then to tell them to register aliases properly
or just use the commands.yml to specify aliases
I get a feeling its only blue when you refuse to send any commands to the client maybe?
(ok, well, there is no commands.yml on the proxy)
But, like, don't fuck your players over 🤷♂️
so i added it to commands yml but still red lol i can go kill myself
wooooow
now it does work 🙂
https://hastebin.com/ijubamajoz.less what does it cause?
the plugin is doing stuff java no longer allows
try updating the plugin, failing that, downgrade java or find a replacement
uh okay
Hi! It appears that the plugin "WarpSystem" is fully broken with the new Vanilla block limits
Elaborate? Logs? Description of setup?
No logs / errors
The plugin is unable to detect that BungeeCord is set up at all. Using the global warps (per-server-portal) command, it asks me to install the plugin on Bungee
& if I try to use a global portal it says the server is not online
It's like that update literally just blinded the plugin from the proxy
There is no way this is caused by the size limits imposed by waterfall as triggering those would disconnect you
After the waterfall update that specifically changed the limits, the plugin broke
I have tried every version and even forks (FlameCord) that utilize that update
I do not believe Bungeecord itself is the issue, but I can compile & test
That’d be appreciated. You don’t need to compile it, you can use https://ci.md-5.net/job/BungeeCord/lastSuccessfulBuild/artifact/bootstrap/target/BungeeCord.jar
Downloading & testing rn
But even if Waterfall is the cause, the plugin is not open-source so we can’t really diagnose the issue
It's waterfall
BungeeCord fully works
Let me record what should happen & what does happen
What should happen ^
what update exactly?
I use FlameCord so I'll have to find the commit they merged
But I was able to trace the problem to Waterfall
I really don’t think it’s caused by anything that’s supported
It looks like this was the patch in FlameCord forked from Waterfall but let me find the Waterfall commit
https://github.com/2lstudios-mc/FlameCord/blob/master/Waterfall-Proxy-Patches/0032-Apply-buffer-limits.patch
that's not anything to do with waterfall
Erm that was never in waterfall. I want you to find the waterfall commit that broke it
^
yea I'm trying to find the version in waterfall thats causing the issue
I know that the proxy I use was updated may 18th
It was before then I'm crawling thru commits
Unless you can point to a waterfall version that it worked in and then didn't, we can't really do much to look at it
Let me try
It’s probably this one https://github.com/PaperMC/Waterfall/commit/c19c4771deb96ad88e01781674d7d716dd10d3c0
nothing should be sending empty packets though?
Wait selected the wrong one

Hmmm
If that’s not it then it’s probably this one https://github.com/PaperMC/Waterfall/commit/2c5c9541c95282ad60a68b14d6f2ac40babef64d
that would produce errors in the logs
Yea as I said
hol up rq
But custom plugins tend to suppress those
I’ve seen that
Thousands of exceptions thrown, caught and not handled
starting to want to die
And then they wonder why it’s laggy
gimme a sec
ok I unironically want minecraft to burn
running bungee made it work
Moving to waterfall it now works
BUT going back to how I had it set up originally NOW it doesn't work

i need to do more testing
Sounds more to me like the plugins folder is borked
Windows has a nasty issue sometimes that makes the spigot classloader only load the plugin main file
I’m not sure I ever found out what causes that
anyone are having issues with older versions of minecraft? i can't see players on my server
Are you using a mod like optifine?
nope i'm not using optifine 😦
hey is there a way to force players when they login the server, to join the server lobby on bungee, instead of sending them to the last server they were on? I am using Waterfall
force default in config.yml
I never knew what that pvp.md-5.net was
zzzCat I see force_default_server so I can set it to true, but where do I specify default server?
uses the priorities list
so I just need to switch force_default_server: to true and nothing else?
I am pretty sure I did that in the past and never worked
and what is: forced_hosts:
pvp.md-5.net: pvp ?
it's a hostname > server map
you can make it so that the proxy will connect you to a server based on the hostname
so my server is play.gp-mc.net and my geyser: bedrock.gp-mc.net
what should I write there?
so then does this look OK to you? basically all my servers are 1.16.5, I will add a 1.8 server and lobby will have the via plugin to allow 1.8 to connect to it and from there they can join the 1.8 server. But if they were on 1.16.5 survival and try to join with 1.8 they would get kicked for wrong version, so I guess forcing to lobby that allows 1.8 will let them connect always with 1.8 client or 1.16 client
am I right?
the proxy works by tryna connect to them in succession
if they fail to connect to the 1st they'll try the 2nd
thx it works love you zzzCat
How can I include dots in my motd using server list plus? It wont recognize it as a valid yaml
You'd have to ask the devs of ServerListPlus. I have fullstop's in my waterafall's MOTD without any issues i.e. [Now running 1.16.5]
Are you using slp?
Nope
i think you could use escape characters
You have read their Wiki, yes? https://github.com/Minecrell/ServerListPlus/wiki/Configuration... Also looking at this plugin I am kinda tempted to add it just for the Player Hover Message feature... Hmmm
Yeah it does but you shouldn't need to escape things like fullstop... Comment was for Zoomy about reading the wiki..
Yes, and I'm confused
Ok what does it look like currently
Hello, I get this Error and I don´t know why and where I have to ask...
https://pastebin.com/v3v8D5ny
Thanks for help ^^
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
Ok thx
easiest fix: use Velocity
1 word: google
Ok sry
The Javadocs for Waterfall seem kinda broken
Because when I f.e. want to access ProxyServer.getName(), which the search bar does show, do I get a 404
https://papermc.io/javadocs/waterfall/undefined/net/md_5/bungee/api/ProxyServer.html#getName()
It probs is the undefined that for some reason gets added
minus undefined because we are defined 
So perhaps a Javadoc misconfiguration
Also, I assume getName() is safe to use for distinguishing different Proxy types/forks?
Like to find out if I run base BungeeCord or Waterfall.
Yup
Same
Is there a waterfall/proxy compatible with the 1.17 pre-releases?
There is /send module
You need to add the correct JVM-flag to unlock it but yes, current version should support that
Does changing network-compression-threshold in server.propertise have any effect since we use bungee/waterfal?
the one in servers affects proxy <=> server
one in waterfall config.yml covers client <=> proxy
so basically since proxy and server both in localhost, I could set network-compression-threshold to -1 to disable?
I wasn't sure to set it to -1 or 99999 which would actually disable
will this do in server.propertise? network-compression-threshold=-1
iirc yes, otherwise see the wiki or whatever
thank you
:o this is useful information
Does anyone know how I can include a fullstop in my motd using ServerListPlus plugin? Once I add one it's not recognized as a valid yaml and I get errors
https://pastebin.com/FsWNBxKL
just use its utf8 encoding string however yaml accepts those
also pretty sure strings surrounded by proper quotes accept more symbols than the fancy multiline stuff
How?
no idea how yaml does it, google should easily tell you that
I disabled network_compression_threshold on all paper servers and only enabled it on waterfal and I saw a significant improve on load time and login. it was very noticable.
now on waterfal, setting compression threshold to 128-256 was the sweetspot. anything lower or higher than that causes slight delay on loadtimes and switching serevrs.
Hey, I'm a little confused on how I would setup a second bungeecotd server for load balancing
Would i connect the second bungeecord server to the same paper servers as the first one or?
But wouldn't it block it from connecting since there is already another bungeecord server connected to it?
no, the proxy isn't connected to the servers, it creates a normal player connectionf or each new player
Ah okay
and would I need to set a motd for the second bungeecord server?
Or does it only show the primary one on the ip
ideally you should set it up the same way as your main one, maybe even with some syncing
Alright
Sorry if these sound like dumb questions, this is my first time setting up multiple
and assumign you load balance correctly it would show one or the other in the ping response in the server list based on which server the load balancer selects
I have 2 Proxy servers on 2 different servers and having issuing connecting to them via TCP Load balancer
I can connect direct to each proxy but not when using the Load Balancer IP
you need to enable proxy-protocol iirc, and make sure you properly configured your LB
yeah i have Proxy-protocol enabled on both
what is your LB ?
yeah thats all setup right
you added both waterfall as backend of your lb ?
yeah
what is the error you get when trying to join
Can't connect to server
then i think your configuration is not well applied/loaded
Would it work if I just used a dev build of ViaVersion which would allow 1.17 pre releases
Like would waterfall support that?
pretty sure there is a config option for snapshot protocols in bungee now or something?
using viaversion on the proxy might also work
i don't think it is updated yet
Alright
Yeah but they don't recommend putting it on the proxy
I can't find that anywhere
https://github.com/SpigotMC/BungeeCord/commit/f0908b663f868ef5fe1f3ac8c81a29827cd6abe1 but I don't think that got updated yet
there is a fork of waterfall iirc which does update see the PRs
Ah okay
well, the Waterfall PRs were a by-product of Velocity updates
To add onto that, you just have to change the version ID from 21w19a to the current prerelease
Which is a few-line patch that can be merged into waterfall already so it’s ready for release
755 is the release network version
well for some resaon ? it seems to be listening on IPv6 and not IPv4
waterfall binds to 0.0.0.0 by default
some OSes are weird and you may need to tell the jvm to prefer ipv4, but, should bind to v4 by default
on ubuntu, using the software CloudNet, it sets it to the host ip
if you have rando software messing with the config there is little we can advise about that
Speak to the software dev
from what i can tell its not an CloudNet issue
waterfall just listens on the IP it's told to bind on
yeah
if your panel tells it to bind on something else, that's on the software
how can I know the actual pre-release ID?
Sorry for being a noob
wiki.vg generally posts it otherwise pretty sure there is a command to dump all that info from the jar
How do I add it?
I'm using net.md_5.bungee.protocol.snapshot but I really do not know how to add it to my JVM
Do I add it after the waterfall.jar?
Thanks a lot!
we don't provide support for forks and not even the right channel
Can a suggestion be made for a good waterfall queue plugin?
I just wrote my own 👀
Hello there, guys!
I'm new to bungeecord (and as a result of the Waterfall), so my question might be a little stupid as I actually am, so please be patient 🙂
I have a pet project - a self-hosted Minecraft server, it is a private server just for friends so nothing commercial.
And I have bought a white IP and running on it using domain.name:port link to access the game server.
I've installed Waterfall and set it up by some guide, enable IP-forwarding and already tested it on two servers: lobby and server. All is fine.
||Should I note that I use Windows 10 as a host machine? kappa||
But the question is might not be regarding the Waterfall at all (I'm not sure):
How could I made the link not domain.name:port but play.domain.name? So just to make the link nicer, easy to understand, and without having to remember the port.
I have a dedicated clear white IP with bought and linked domain name for it
what?
if your IP is not shared, bind the proxy on 25565 and then you don't need to specify the port
then it's just a typical IP address, you'd create a record for play. to point to the server
bind the proxy on 25565
I guess that part is a summary of my question
How to do that?
change the listener in config.yml and set the port there
the host is already set to my ip:port
so the main link to access the game server is ip:port and domain.name:port also works
What the.. next step?
|| Should I double-note that I'd like to get rid of port inside my access link? ||
Ok, done. What's next?
that's it from the waterfall side
just create an A record for play which points to your IP
Where should I create it?
I'm not sure I'm clear enough in my wishes..
I'd like to make such a structure:
play.domain.name >> ip:25565
creative.domain.name >> ip:25566
pvp.domain.name >> ip:25567
etc.
** Using BungeeCord/Waterfall is not necessary at all I just hear somewhere that Bungee Cord is a proxy that could allow me to do that. **


