#waterfall-help
1 messages · Page 2 of 1
Are you using Via plugins?
yes
Update to latest dev builds
i have
Well, try removing plugins then
done but dosnt helped
It is definitely Waterfall at fault
pls ping me if someone have an idea
Hello, so a chunk a corrupted at my spawn and playerdata files are part of the world. Is there anyway I can migrate my playerdata to a new world?
Hey does waterfall support Microsofts new chat reporting?
What is the best proxy to use? (With lowest ram usage)
What
If you have 200mb for example you can't run any proxy i mean
If you have 0 players yes
How can i enable chat verification or "fix" this?
As I know, proxies does not have that yet
But how do big networks do it? i saw it there? they definitely use bungeecprd/waterfall or something
offline mode?
I mean
I'd imagine that either they have the option to enforce chat signing set
or, are just not sending the server data packet or something
probs irrelevant
[20:26:04 INFO]: Using standard Java JCE cipher.
[20:26:04 INFO]: Using standard Java compressor.
[20:26:04 INFO]: Enabled Waterfall version git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:fc30291:498
[20:26:04 ERROR]: Exception in thread "main" unacceptable code point '' (0x0) special characters are not allowed
[20:26:04 ERROR]: in "'reader'", position 0
[20:26:04 ERROR]: at org.yaml.snakeyaml.reader.StreamReader.update(StreamReader.java:211)
...
[20:26:04 ERROR]: at org.yaml.snakeyaml.Yaml.load(Yaml.java:429)
[20:26:04 ERROR]: at net.md_5.bungee.module.ModuleManager.load(ModuleManager.java:63)
[20:26:04 ERROR]: at net.md_5.bungee.BungeeCord.start(BungeeCord.java:272)
[20:26:04 ERROR]: at net.md_5.bungee.BungeeCordLauncher.main(BungeeCordLauncher.java:67)
[20:26:04 ERROR]: at net.md_5.bungee.Bootstrap.main(Bootstrap.java:15)```
any idea what's up? this is a test server that was working just fine earlier this week...
build 500 is also doing this
Hi, I have problem with Waterfall. If the player takes damage, he kicks him out of the server (only on version 1.19+)
My version Waterfall:
1:20min
@unique linden
you can ignore this. Looks like it was file corruption due to my computer crashing earlier today
support ;-;
don't ping random staff
your config type is wrong
Seems your plugins issue try waterfall in vanilla
sorry..
@craggy troutHello, i've heard you are advanced in the proxies
I am dealing with one proxy issue: (I have been looking for this problem whole day)
(saw same issues on spigotmc forums but without solution)
Can you help me somehow please?
Look:
that's how java works
The entire point of the heap is that it's a block of managed memory
afaik, the server JRE doesn't really free memory back to the OS all that much
Before this problem started i was running bungee with -Xms128M -Xmx328M 24/7 without overload crash
I mean, what you're showing there is no an issue
When i set it to -Xmx328M now, it overload/crash very soon
crashes how?
out of memory
if heap memory, take a memory dump
My memory is going lower and lower without releasing
But when i restart the proxy - all that memory is going back
Because that's how java works
ok (so it's not issue then yea 😄 )
i will need to restart proxy like every 12hours 😄
depends on the activity of the players
it releasing memory back to the OS is literally not in the hands of the proxy, that's 100% on the JVM and the GC in play, which as said, the server JVM does not release memory back to the OS all that often
if you're running out of heap memory, there is a JVM flag which can be used to tell the thing to create a heap dump when it OOMs
-XX:+HeapDumpOnOutOfMemoryError
then you can load that up into eclipse MAT or whatever
so there is solution for this yea?
i am not the pro
The jvm is working as expected here
if you're actually suffering an OOM, you need to work out why, that's a seperate thing entirely to how the JVM works
But i saw exactly same issue on spigotmc forum:
if the JVM is running out of heap, either allocate more, or work out why it's running out
which is literally irrelevant to the fact that the server JVM does not release memory back to the OS as much as the client JVM does
every join adding 30MB+ to the allocated memory is entirely sus af and probs screams that something is being stupid somewhere
ok, so it's normal, that the proxy can use +1GB after like 10hours of a few players
but, most of the proxies memory usage if off heap, so it gets hard to say, especially as offheap memory usage is crazy
No
I literally never said that that was normal
last refresh
😦 soon crash
Thank you so much for help. but I don't know if I can solve it.
I can't help you
That will no solve your issue
I think that even you have more problems of stability using it with some patches that it haves.
yea you are right (with flamecord it's getting lower more quickly)
when i swap to hexagon (for a test)
it was getting lower more slowly 👍
ok, ok
ask geyser
chances are you've got geyser connecting to the wrong thing
i.e. it needs to go through the proxy, but, 🤷♂️
hello, please tell me how to get player count from another server
oh, yea, I forgot about that specific wording, mb; that option should fix it
Memory is stable now with WaterFall (maybe)
- is there some way how to support 1.7 clients too?, cuz i am doing PvP server.
and last question, i am using XCord's iptables but when i try to stresstest antibot it accept so much CPS compare to other protected servers
It looks like that IP tables are not doing their job or idk :/
First of all, we don't support forks and also offline mode is not supported tho
MCstorm can be handled by a proper firewall, machine and online mode to absorb the impact
Can it cause low CPU ?
What exactly?
With online mode, those "storm" ddos can't do anything
So yes, cpu usage will be barely afected by that dead traffic
Also as i said, if you have a proper firewall and good host will handle it cool
Ok thank you for help 👍 (cuz i've downgraded machine and CPU is pretty garbage there, but for players it's comfortable (the biggest problem was RAM - which seems to be resolved using WaterFall ❤️ )
i just thought that iptables aren't working 😄 (because I had to go through so much buggy stuff, so I wouldn't be surprised 😄 ..)
How do I let players with different server mod packs join the same hub?
one of the mods for one of the server packs is throwing a incomplete tags when joining the hub server
What do you think, why am I getting such an error when I enter the server with version 1.19.2? my lobby server is version 1.19 with plugins viaversion and viabackwards
on backend or proxy
im having this error:
A task raised an exception. Task: net.md_5.bungee.connection.InitialHandler$6$1@1341462a
java.lang.ClassCastException: java.lang.String cannot be cast to java.util.Collection
at net.md_5.bungee.conf.YamlConfig.getPermissions(YamlConfig.java:321)
at net.md_5.bungee.UserConnection.addGroups(UserConnection.java:550)
at net.md_5.bungee.UserConnection.init(UserConnection.java:162)
at net.md_5.bungee.connection.InitialHandler$6$1.run(InitialHandler.java:597)
at io.netty.util.concurrent.AbstractEventExecutor.runTask(AbstractEventExecutor.java:174)
at io.netty.util.concurrent.AbstractEventExecutor.safeExecute(AbstractEventExecutor.java:167)
at io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:470)
at io.netty.channel.epoll.EpollEventLoop.run(EpollEventLoop.java:391)
at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:995)
at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
at java.lang.Thread.run(Thread.java:748)
bad config
if i remove all plugins no works
but, your config is bad
wait is the permission to block the command /alert
how can I find out which plugin is to blame for this message?
by default only admins can use it
hmm is there a way to setup network chat to each other servers?
None. Not supported on bungee/velocity yet
Also no, and I doubt bungee will ever get api for it
hmm i see
em
This is not the rigth channel, please go to #paper-help and don't try to tag all.
How do I make it so that if I turn off one of the servers on my network, the players connected to that mode are redirected to the lobby instead of disconnecting them?
I'd suggest using the search, the answer is pretty much gonna be "use a plugin", and theres been varying suggestions over the past god knows how long
wdym
is this normal for 8 days server using this much bandwidth
bcz players on my server do get ping spikes time to time... no ddos or anything
using ovh with 500mb up/down
How much players you have
1TB is not that much
Even 30 online "normaly" eats 1tb in one single month
20-25 at peak time
Hm, well in 8 days is not that normal but
but thats 2tb for 7 days
Probably you got some day ddosed
Any ways
OVH does not have traffic limit?
As far i know, yes even they're antiddos
OP has no effect on proxy servers
Iirc permission for it is Bungeecord.command.server
(also you need Luckperms bungee)
don't think so... no limit as far i know
No?
He can use bungeecord permission system tho (Adding imself as admin)
Well yes, but lets be honest, who doesn't use Luckperms nowadays
I mean, yes and no, there are servers that do not use permission plugins in a proxy, if in backends it seems to me something essential
Also I meant it in a way that he needs to have perms via proxy, many people assume that normal LP on backend will work with proxy perms
I see
Hi ^^ I don't know if you can help me here ^^ But i have very lot of ping in my waterfall console in localhost with different port i don't know why.
https://hastebin.de/ipenohofum.md
I have activated ping in config to show who ping me
Or it's probably a plugin to check for exemple player count or another ?
You can hide logs if are just anonnying you
I can yes, but if i see that i can see a ddos attack no ?
Is not
Yes i know
Yes it's spam all the time but i search what plugin spam for that every 3 secondes
No plugin is causing that
Do someone know what does it mean?
ip forward enabled in waterfall config.yml
bungeecord enabled in spigot.yml
Waterfall config
config.yml by @upbeat salmon: https://paste.gg/1b937f807f844d68bb72e02532e30265
spigot.yml
spigot.yml by @upbeat salmon: https://paste.gg/83c42a3d2a8149c4aad897312f6f5a09
For me seems a plugin
I dont have plugins on my servers
Wait, is that forge?+
Yes, magma
bruh
how to fix 1 slot problem in waterfall max player slots sets to 100 but it shows just 1 player slot
Do you have a MOTD plugin installed?
Yes but tried without motd plugin same thing without motd plugin
btw i am using waterfall fork (aegis) :kek:
-1
Then please ask in their support channel, if they have one.
@delicate phoenix
i changed to waterfall
same thing
player_limit: -1
prevent_proxy_connections: false
listeners:
- bind_local_address: true
force_default_server: true
forced_hosts:
pvp.md-5.net: pvp
host: 0.0.0.0:41694
max_players: 100
motd: '&1Another Bungee server'
ping_passthrough: false
priorities:- auth
proxy_protocol: false
query_enabled: false
query_port: 25577
tab_list: GLOBAL_PING
tab_size: 60
enforce_secure_profile: false
remote_ping_timeout: 5000
forge_support: false
remote_ping_cache: -1
inject_commands: false
server_connect_timeout: 5000
log_commands: false
log_pings: true
network_compression_threshold: 256
disabled_commands:
- auth
- disabledcommandhere
groups:
md_5:- admin
online_mode: false
- admin
heres config
Don't use magma
That's your first problem
I'm guessing you're trying to proxy a 1.16.5 modded server?
No, 1.18.2, but i found Arclight and it works fine
Yeah arclight works
The issue with lightfall though is that it requires client modification?
meaning you have to install their lightfall client mod iirc
been a while since i looked at that at all
Try without plugins, but other than that it should work. 👀
i'd just stay away from using magma / mohist if possible
Why?
their software is fairly unstable, and they do some shady checks on your plugins to replace them with their own modified versions, magma does some of the same but iirc during log4j they had a bit of a different mess to deal with. these are some of the reasons authors of plugins like essentialsx don't support these hybrids due to the changes they do
not sure if any of these things are ongoing still since i don't really use these kinds of software, but spongeforge is the way to go if you still want to run plugins. mohist and magma simply causes more harm than good
i think essentialsx still has their PSA up too with more information on why you should avoid them
So, is arclight ok?
arclight is a different story
arclight afaik is okay
it's just a waterfall fork that implements modern forge support, only downside with it that i know or have heard of is that it requires you to install their client mod to even connect to the arclight server
hybrid software is ass
It has poor performance, unsupported plugin developers, instability and bugs. Basically why not make everything separate, mods and plugins on separate servers
it also seems to me like most people value the ability to use spigot plugins over security
those kinds of software is something most people should stay away from, but i guess they fall into that category "but using plugins on a modern forge server is so cool"
waterfall current version is for all version of 1.19 right
Waterfall is a proxy multi-version from 1.8 to 1.19.2x
Hi if proxy is far away from the backend server the ping when playing on the backend server will be affected?
Hello, i need some help with running different minecraft server on the same machine, i need to edit the launch script to load the right version of jdk for that server version? And if that's the answer how can i do that?😅
https://gyazo.com/041521e57aa5e1453d0d0a238526d3c5
I'm getting when trying to join my pixelmon server 1.12.2 with the latest version of viaversion and all on my waterfall proxy
Can anyone help? Have 2 servers setup on waterfall and just added a third yet with the third server i receive this error trying to join 20:31:40 [SEVERE] [unknowndecay] -> UpstreamBridge - encountered exception io.netty.handler.codec.EncoderException: java.lang.IllegalArgumentException: Cannot get ID for packet class net.md_5.bungee.protocol.packet.Chat in phase GAME with direction TO_CLIENT at io.netty.handler.codec.MessageToByteEncoder.write(MessageToByteEncoder.java:125) at io.netty.channel.AbstractChannelHandlerContext.invokeWrite0(AbstractChannelHandlerContext.java:717) at
assuming you're using waterfall, that's an outdated version
yes
also, most issues like that seem to be down to some issue with plugins and potentially plugins + proto hacks; 🤷♂️
Defo latest one, from what I’ve seen online viaversion can be an issue
It’s just odd that it’s affected a new installation
o.O
What
I mean, pretty sure that the error message should be marginally different
So it’s outdated ?
No, that is apparently the latest version, at least, recent enough
but, that message appears to have been modified, so, no idea
if 1.19+ client, bad plugin or proto hack issue
Hmm ok, im going to start with removing plugins
this is something that I'm wondering because I don't know if this is a new BungeeCord behavior:
If I connect to BungeeCord, then move myself to another server, then leave the server, the Minecraft client shows the MOTD of the server that I was switched to 
So, let's suppose that my server (not BungeeCord!) MOTD is "hello world!!!"
if I switch to that server via BungeeCord, and leave, the MOTD in the list will be "hello world!!!", however, Minecraft automatically refreshes the server list so it is hard to see it, however if you end up getting disconnected in a way that doesn't cause a server list refresh (example: network issues), the server MOTD will persist
Which is weird, I didn't think that the client accepted server list packets after connecting to a server (in this case, Waterfall)
found out what is causing this: https://github.com/SpigotMC/BungeeCord/issues/3363 , so looks like the issue was on my end, because this is already fixed on the latest Waterfall version 
Bungeecord version 1656 Server version 1.19.1 Client version 1.19.1 Bungeecord plugins nothing The bug For some reason, when the sub server closes but the Bungeecord is still on, the bungeecord tak...
Next time just ask me directly my friend

Could the error be caused by bungee guard not being on the new server? Just remembered I had that installed
Hello
I have a problem
i can't use waterfall with mods
This server has mods that require Forge to be installed on the client. Contact your server admin for more details.
I have the mod in my mc client
and in the server
Is there anyway to ping a server from to see if it’s connectable
crtl+f
nope velocity doesnt have support either
Directly? no
So should I try reinstalling the proxy instead?
reinstalling the proxy isn't going to magically change the fact that something, somewhere, is sending bad packets
I mean, is this a 1.19 client?
Yes
so, you have something inducing a bad packet from somewhere
Why would it only affect one 1/3 servers tho
Then could be your backend?
It’s just odd as the server I’ve just added is identical to the other 2 yet receive that error
Well there is some conflict on that server, seeing what are you saying can't be identical must be some miss
Hmm
Ok just an update on how stupid i was being, i forgot to forward the port....
i am getting this issue when i tried to run waterfall
latest.log by @brave crescent: https://paste.gg/c6d7daf49cb34f0a91e4d66333fba467
ok i got it what's the issue
bad config
[05:23:44] [main/ERROR]: Exception in thread "main" java.lang.ClassCastException: class java.util.LinkedHashMap cannot be cast to class java.lang.String (java.util.LinkedHashMap and java.lang.String are in module java.base of loader 'bootstrap')
read it zd
Hey. I want to protect my server against ddos attacks. I use rn tcpshield but the problem is i cant add bedrock support. And now i want to find a way to protect my server without tcpshield. (hide my ip adresse or something like that) does anyone maybe knows how to fix that? I use waterfall and purpur. rn
You’ll end up with another server that you’ll somehow have to attach to your network. Good luck.
I mean, it's basic networking, you need somewhere for people to connect to
Like, TCPShield is the "easy" option, otherwise you get into "rent a cheapo VPS"
But, your DDoS protection there is gonna be much less vs the protection that TCPShield offers
UDP is much harder to protect
Which is why I don’t think there’s a service that offers it at all
Don't they have bedrock support in one of their plans?
Getting an extremely weird issue, where players randomly experiencing timeouts from our servers, and then when trying to re-connect, it's telling them they are already online.
Checking if they are online /find <username>claims they are online on the bungee level, but there not online spigot level.
Any solution for this?
They do.
#3382: Use the correct write method for ChatChain in ClientChat packet 
Hey! whats the best value for network_compression_threshold? I have a beefy amd ryzen 5900X and 2x1gbit connection. i want the best performance.. i know at the backend / gameservers i should do it -1 but whats about waterfall? is lower better?
Hello, I connected my servers but when I connect to WaterFall, I have this error, and no message in the console of my lobby : Could not connect to a default or fallback server, please try again later: io.netty.channel.ConnectTimeoutException
send the config in
Keep it on that length, lower isn't really better
Hi I want to make a plugin which make player with the specific permission could join server.
However,I listen PostLoginEvent to check permission and it makes log full of join and disconnect message.
Is there a better way to check a player's permission even if the player is not online?
<-> InitialHandler - encountered exception: net.md_5.bungee.util.QuietException: Unexpected packet received during login process!
I've disabled all plugins
See the logs from the server; all that says is that we got an unexpected packet from the server
generally, make sure that you setup ip forwarding properly, reproduce without plugins on the backend/using a vanilla client
I've turned on IP Forwarding in config.yml if that's what you mean
Can you try again with vanilla(no forge/fabric/quilt) and with no plugins on your backend server
I just woke up, did just that and the error still persists
figured it out lol
Lastest waterfall?
yup
Broken plugin messing with internals
Could it be ViaVersion?
No
How do i fix this?
Remove plugins till it stops, then you know the perpetrator
So... it is a plugin?
Why could it not be ViaVersion?
- It is a plugin doing something it shouldn’t be doing
- Via has their own implementation of these packets so will never access the platform implementation and is thus unable to generate this error
Ah okay.
I think it is maybe because I have LuckPerms Bukkit on my server?
(no clue how it got there)
I have LuckPerms bungee also...
Well it ain’t the Via plugins and not LuckPerms
That just leaves a few. Remove them till it stops
It only affects version 1.19.1/1.19.2
hence why I am saying ViaVersion since the server runs 1.17.1
so yeah
What I said isn’t changed by that fact
Aight Okay.
So the error is messaging?
By messaging what do you mean?
WaterFall --> Paper for example?
if so then its CommandForward
A plugin is trying to send a legacy chat message to the client
Those no longer exist in 1.19+
So the unsupported internal call leads to this error.
It is a plugin, I told you how to find it.
Ohhhhh
Okay.
I gotta wait till some people hop off the network.
One more really quick question, is it a proxy plugin issue or a paper plugin issue?
A proxy plugin, I said so twice now
How can I make it if my server goes down instead of disconnecting from the network they get kicked to hub?
Just checking. :3
Ping me with response pls
This is the RAM usage of my proxy server.
Recently, I deleted a plugin that caused the server to not allow 1.19+ clients to join(more info above)
This plugin was the Bukkit version of LuckPerms that was misplaced onto the proxy server.
Could that be the cause of the wave type RAM usage?
The proxy server should be using 2GB of ram. Those spikes are 6GB of RAM.
You can notice the spiking stop, I think that was the solution of the problem.
I need clarification. Is this possible?
you can notice the slow rise of RAM as well, for no reason.
use a plugin
What plugin does this?
search for them
please give me some updates on this
What is your XMX?
If it’s 6gb then you can expect it to use 6gb
Yeah, I can expect.
But those spikes are only 2 players online.
- I can't explain, but a 1GB of RAM getting used for ~15 minutes.
every now and then
those spikes do not resemble that though
they just resemble it jumping all over the place
its a plugin I use, again, I cant be bothered to type everything out, but it does use exactly one gig of ram for 15 minutes
Yell at your GC. I recommend you to set the XMS to the same as the XMX so you don’t overprovision your system memory
What that chart shows 90% sure isn’t the actual usage on the JVM but the JVM system memory usage
Yeah right. Okay.
What does the XMX value do?
ah nevermind they are the same
oh god this is embarrassing
i only just noticed...
my server is not running waterfall, but actually bungeecord 
oops... yeah I need to migrate to waterfall, I actually convinced myself it was running waterfall oops.
hence the performance issues probably
right. gonna migrate now.
No way bungeecord haves that several performance problems, seems your fault
What?
WaterFall is far more efficient.
It is the JVM using the RAM actually, pretty sure.
Not the actual usage.
like my proxy is using 4gb of ram just for 2 players
Yes, is more efficient, but that doesn't mean that bungee is that ass.
Is not using 4gb, they're keeped for that process if you are using aikar flags
Not using any flags except the default -XMS and -XMX
Well, that's exactly what i said
If you don't want to use 4gb, use 1 then or just run it without flags lol
Would one be enough to run 20 players max?
The recommended is 0.5mb per player if you don't have huge plugins, so yes.
ok cool
I have Plan, LuckPerms and that is it.
I do have a plugin though
It creates custom player servers.
Since I was so tired, it took me 1 hour just to notice.
That plugin uses Linux's Screen system, meaning that it isn't running in JVM, so I can set WaterFall to one gig, and that plugin can use 4-6GB server.
Ok cool, I get it now.
I am trying to connect to my waterfall server and I just get this
I set up this server like 15 mins ago and am new to waterfall (but not minecraft server hosting) so i would appreciate some help
generally means firewall or wrong IP, etc
what about the firewall since i know the IP is correct
would that a be a server side firewall or local on my pc
depends on your setup, etc
all I can tell you is that whatever address you tried to connect to, something there along the way rejected the connection
ok, it shouldn't be anything to do with the config tho, correct?
No, you've not even got the connection to waterfall established
depends on what "local" is
make sure that you're not using a public IP address to connect to the same machine as you're hosting from, or another machine on the network, not all routers like that
ok, well i am using private ip address to connect to a server on the same network as my pc
well, thank you for trying to help, i understand there isn't really much you can do for me here
this is interesting, if i open another minecraft server instance on the same server i am able to connect but then when i try to connect to it through waterfall, it just doesn't work
er
do note that waterfall uses a different port by default
25575 or something
oh, 25577
You probably wanna make sure that you either use that port, or, ideally, set waterfall to listen on 25565
oh
that is probably what i was missing
i appreciate it, thank you for the help even though it was such a stupid thing like a port number
When I try to log in with versions 1.19.1 and 1.19.2, I get such an error, all my players have such a "Waterfall#502" version installed, what am I doing wrong?
See logs, etc
Make sure that plugins are uodated
All we can see is that you where basicslly kicked because of some issue between the proxy and the backend
It gives such an error in the console, I just realized, do you know if it will be fixed when I add this, do you know that I will turn my server off and on again.
Placing bets on outdated plugins
also skinrestorer 💀
Update your plugins.
i also question the need for a two factor auth system on a server? but that's just me
I'm placing bets on the pack thing as the name sounds familiar
Otherwise, like, glhf
i mean, yeah the BungeePackFix's purpose is so that you don't have to redownload resource packs iirc?
now addresslimit is quite useless
I updated the bungeepackfix plugin and it's fixed, thanks for your help, I'll take your suggestions into account
❤️
don't use plugins on proxy 
There's a veeeery short list of things which is best to do on the proxy vs individual servers
Stuff on the proxy messing with specific things alsk generally tends to be stupendously more fragile
hello sir, i am attempting to link my pixelmon 1.12.2 server moistmc to my waterfall bungee but seem to fall into a error 25 bytes extra whilst etc.
on switch to that server or when you join it as the first server?
switch
some mods have issues with that, try joining the server as the first one or running the mod pack on your lobby too
would that not stop my spigot players from joining the lobby?
well yes, you would need to have them join a different one or make the mods optional (no idea if forge can easily do that, technically it's possible)
its tough, iv seen/been on a good few networks that have the combo just not sure how
could also have two lobbies, one for modded and one for normal players
also afaik it depends on the mods in the mod pack if switching works or not
and there are some tricks about resending stuff to the server/client
only 7 mods haha
You may not run into this issue with velocity
It’s a long shot
Il give it a shot. Its just the base 1.12.2 pixelmon network modpack. 7 mods which BOP is one
Hey
What can i do to prevent UUID Spoofing?
Im afraid of my server getting griefed 😦
outside of insecure backends, in which case, properly configured firewall or at least bungeeguard, literally hasn't been a thing for years
(outside of people who intentionally configure their servers in insecure manners)
I have bungeeguard
Then unless you have a compromised secret/key or plugin (or server in general) you should be good
uuid spoofing isnt real
It very much is if you don’t have protection measures against it
first of all, since you host a modded 1.12.2 server, (this applies to other modded servers too), avoid using mohistmc, magma or any other kind of bukkit / forge hybrid. they are quite unstable, and they do more harm than good. instead, use SpongeForge if plugins are an absolute must.
in regards to hosting a modded / vanilla combined hub for your players @sly sun what you can do is either run SpongeVanilla, or run Paper but add this to your startup arguments -Dpaper.disableChannelLimit=true this is because forge usually have a lot of custom channels which will eventually become a problem for you when it comes to joining vanilla servers.
try loading waterfall with the flag and give the error then
ah will do.
The flag didn't seem to work.
:(
Fixed it.
how?
disconnected with: Could not connect to a default or fallback server, please try again later:
help pls
Not enough info, see the full error
Why doesn't the documentation list permissions?
What's the permission for /send current?
Because we just refer people to the bungeecord documentation
The home of Spigot a high performance, no lag customized CraftBukkit Minecraft server API, and BungeeCord, the cloud server proxy.
yes
Seems like /send current is new? Or I just never knew about it before
Ugh, there are no subpermissions for /send, how stupid
the built in commands are pretty basic
part of what I think the hope is that people use plugins to replace the servers commands with competent replacements
I'm not familiar with any plugins that do /send but I haven't really looked either. Just not a fan of adding more plugins to accomplish things that should be built in
Obviously that's not Paper team's fault
Thing is that everything in bungee is basic
(Except for the world class backwards compatibility
)
Like, theres no sane permission system, you either have a string or you don't
I mean, the API literally just returns json strings everywhere
wdym
I mean that's not enough info for anybody to help with
i'm not a fan of the /send command, so i decided to make one myself tbh
Hello everyone, how to hide this flood in console?
hello, i hope that someone could help me :
io.netty.handler.codec.EncoderException: java.lang.NullPointerException
at io.netty.handler.codec.MessageToByteEncoder.write(MessageToByteEncoder.java:125) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.channel.AbstractChannelHandlerContext.invokeWrite0(AbstractChannelHandlerContext.java:717) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.channel.AbstractChannelHandlerContext.invokeWriteAndFlush(AbstractChannelHandlerContext.java:764) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.channel.AbstractChannelHandlerContext$WriteTask.run(AbstractChannelHandlerContext.java:1071) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.util.concurrent.AbstractEventExecutor.runTask(AbstractEventExecutor.java:174) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.util.concurrent.AbstractEventExecutor.safeExecute(AbstractEventExecutor.java:167) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:470) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.channel.epoll.EpollEventLoop.run(EpollEventLoop.java:391) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:995) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at java.lang.Thread.run(Thread.java:748) ~[?:1.8.0_312]
Caused by: java.lang.NullPointerException
at net.md_5.bungee.protocol.DefinedPacket.writeString(DefinedPacket.java:32) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at net.md_5.bungee.protocol.DefinedPacket.writeString(DefinedPacket.java:27) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at net.md_5.bungee.protocol.packet.Title.write(Title.java:70) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at net.md_5.bungee.protocol.MinecraftEncoder.encode(MinecraftEncoder.java:24) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at net.md_5.bungee.protocol.MinecraftEncoder.encode(MinecraftEncoder.java:9) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
at io.netty.handler.codec.MessageToByteEncoder.write(MessageToByteEncoder.java:107) ~[bungeecord.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:9c9bc55:502]
... 10 more
[05:12:40 INFO]: [Hemef] disconnected with: EncoderException : java.lang.NullPointerException @ io.netty.handler.codec.MessageToByteEncoder:125```
i cant connect on my server sometimes and this occur
Hellow anyone? 
250000018c0000ff06136d696e656372```
Removing all plugins and initializing user data results in this error
make sure that you reproduced without plugins on both the proxy and the server
see the server logs, all we can see is that waterfall got a packet it wasn't expecting
I don't support offline mode
and, that doesn't give anything useful, outside of the fact that that ain't without plugins
I tested it when there was no plugin.
Then shows logs without plugins
would need full logs
latest.log by @fair dagger: https://paste.gg/c0a1a186c87d494d8551c6d43cbd49ea
/127.0.0.1:56283 lost connection: If you wish to use IP forwarding, please enable it in your BungeeCord config as well!
fixed
Well, it's fixed because I touched it many times.
Sorry
on the proxy ?
yes
I noticed that it occur for 1.18 1.19 but not for lower versions
Because the code in which that is being thrown only runs for 1.17+
something is causing a bad Title packet to be sent
That thing can literally only be something running on the proxy
okok ty I'll check and let u know
Would 1.5.2 player be actually able to join the proxy? I would like to have some modpack servers running alongside
No, Waterfall doesn't support 1.5.2.
Hello, I have this error every time a player that is on 1.19 joins the server
18:55:00 [WARNING] [/ip:port] <-> InitialHandler - could not decode packet! java.lang.IndexOutOfBoundsException: readerIndex(7) + length(1) exceeds writerIndex(7): PooledSlicedByteBuf(ridx: 7, widx: 7, cap: 7/7, unwrapped: PooledUnsafeDirectByteBuf(ridx: 239, widx: 239, cap: 2048))
Ive searched this error in the past on waterfall help to see if someone had the same problem, unfortunately that was true and the solution was to check the plugins, in this case I disabled every single plugin on my server and the issue still persists. I even enabled the debug flag -Dwaterfall.packet-decode-logging=true and still I couldnt get anything to know about the issue, can someone help me? It would be appreciated. Thanks
All we can see is that something sent a packet which the proxy wasn't happy with
reproduce without plugins on the proxy and the server
make sure that the proxy is updated
The proxy has been updated now I get the following error:
https://pastebin.com/8dxjA78U
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.
What placeholder can I use to get players count on each server? I want it to use in server motd
%bungee_your-specific-server% for a specific server %bungee_total% for globally
it's not working in motd from bungee
i think that depends more on the plugin
please can someone help me with this? is necessary
We have similar errors for past 2 days but we are getting hit by bots in waves. That what ovh says but they do nothing in this matter...
For some time now we getting this
[20:31:33 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
Generally its weird cuz we didn't change anything past week, vps or bungee side and those timeouts are really bad for our network :/
Hi how can i solve this?
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.
Team HFLVZ_Z111 already exists in this scoreboard
Check your plugins on the non-proxy servers. It looks like one is trying to register a scoreboard team twice.
hey so I wanted to know how can I make it so when a player joins it gets put in the lobby / authentication server even tho they where in another server?
server op has nothing to do with proxy permissions
so what do i do?
You have to use a permission plugin on proxy
i have luckperms
Are you using something like LuckPerms on your server ?
yes
ok so install and configure it on your proxy as well
yes i have it there
Ok so nothing more to do
Just be aware that the op thing is not a proxy thing
It only applies to server things
Hi, is the latest version of Waterfall compatible with Minecraft 1.18.2?
I get the error message Kicked whilst connecting to Survival: This server has mods that require Forge to be installed on the client. Contact your server admin for more details and I have Forge on both the server and the client and the option of Forge_support option set to true. Can you help me? Thanks
Modern forge is not supported currently without hacky tricks
Waterfall supports from 1.8 to 1.19.2x
And what hacky tricks are those?
use a fork of waterfall wich depends on a mod to work
Which fork? Sorry but I don't know much about it
ligthfall
thank you very much for your help
Where i can set to connect players to specific server when players server go down?
Use a plugin like RedirectPlus, so when a server crashes it sends the player to the specified server like lobby for example.
What is Waterfall?
a fork of BungeeCord proxy
Ok thanks
Hi there, i recently updated my Waterfull too waterfall-1.19-500.jar and i have the config setup correctly. I can provide info with retracted details if required. Connection is from Waterfall (docker java 8, 17 was tested) -> backend(ptero connection) -> 1.8.8 (docker java 11) server (i know it's not supported for support) and i believe it is possible that Waterfall is the fault. I don't have proof but it's a possibility.
java.lang.RuntimeException: Unable to access address of buffer
at io.netty.channel.epoll.Native.read(Native Method) ~[patched.jar:git-PaperSpigot-"4c7641d"]
at io.netty.channel.epoll.EpollSocketChannel$EpollSocketUnsafe.doReadBytes(EpollSocketChannel.java:678) [patched.jar:git-PaperSpigot-"4c7641d"]
at io.netty.channel.epoll.EpollSocketChannel$EpollSocketUnsafe.epollInReady(EpollSocketChannel.java:714) [patched.jar:git-PaperSpigot-"4c7641d"]
at io.netty.channel.epoll.EpollSocketChannel$EpollSocketUnsafe$3.run(EpollSocketChannel.java:755) [patched.jar:git-PaperSpigot-"4c7641d"]
at io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:380) [patched.jar:git-PaperSpigot-"4c7641d"]
at io.netty.channel.epoll.EpollEventLoop.run(EpollEventLoop.java:268) [patched.jar:git-PaperSpigot-"4c7641d"]
at io.netty.util.concurrent.SingleThreadEventExecutor$2.run(SingleThreadEventExecutor.java:116) [patched.jar:git-PaperSpigot-"4c7641d"]
at java.base/java.lang.Thread.run(Thread.java:829) [?:?]
``` and this is the error that spams on my 1.8.8 server.
FIXED: go into server.properties and turn ``use-native-transport=true`` to false
Disable native epoll on the 1.8 server, should be in server.properties iirc
use-native-transport=true -> false?
you guys are legends! i would never had suspected this!
I wouldn't do that
Performance goes really down
Why you just don't use a fork that implements new netty on the serializer foir new java support or just use diferent java versions?
can also downgrade to java eight yeah
turning off native epoll is just a lot simpler to do :P
(and most people probably won't notice the performance difference because they're not at a large scale)
It's been fixed. please read the ENTIRE message
It's not a real fix, it's just a mitigation which comes with a performance hit
That's not a fix, for me is called a tricky hack
Hello everyone, does anyone know if there are any plugins or configuration that improve Waterfall performance? Sometimes I get stuck on the loading screen, other times it seems like my connection is corrupted and the map doesn't load and it's only resolved when I reconnect to the server
Can anyone help?
When I try to connect to my proxy, It just says the world is loading, And then I eventually get kicked, Meaning I cannot connect to my server D:
Then either code it yourself or find something/something else to figure it out. as i gotno clue about actually fixing it from the jar.
ReadTimeoutException: Null
I keep getting this/
I said him how to
Think is though, Featherboard doesn't work with Java 8 anymore. it's Java 11 or higher now.
at least that's what the console said
[FeatherBoard] You are using Java 16.0.2 on a Spigot version below 1.16.5. Either update to Spigot 1.16.5 or use Java 15.
You can use my plugin if you want
Is free, open source, helly better and optimizied than fb
why can't I join with 1.19.2 even when I downloaded the newest version?
what error does it give you
ok never mind i was stupid and forgot to add viaversion to the server
is there a way to only allow going to one server with the /server command, ex. the lobby.
"yes"
if the server is set to restricted in the config, then you need the perm for jumping to it with /server
nvm i understand but then you could go to any server from that one you have perms on?
If you set every other server to restricted, the only server you'd be able to jump to with the server command, is the one you didn't restrict, unless you gave them perms for those servers
Plugins will be able to still let people jump between servers
If you want anything more convoluted, you'd need plugins
oh i see
Hi, What is the difference between waterfall and velocity? And which one would you recommend I use?
Waterfall is basically BungeeCord but with some patches like optimizations and other cool things. Velocity is a completely standalone and has nothing to do with Waterfall/BungeeCord. Me and any other would recommend Velocity.
read that to understand what's better for you
Placeholder for Alive players??
hey am having trouble java.lang.NoSuchMethodError: 'java.lang.String[][] net.md_5.bungee.protocol.packet.PlayerListItem$Item.getProperties()' at me.neznamy.tab.platforms.bungeecord.BungeePacketBuilder.readPlayerInfo(BungeePacketBuilder.java:120) ~[?:?] at me.neznamy.tab.shared.FeatureManagerImpl.onPacketPlayOutPlayerInfo(FeatureManagerImpl.java:93) ~[?:?] at me.neznamy.tab.platforms.bungeecord.BungeePipelineInjector$BungeeChannelDuplexHandler.write(BungeePipelineInjector.java:69) ~[?:?] at io.netty.channel.AbstractChannelHandlerContext.invokeWrite0(AbstractChannelHandlerContext.java:717) ~[server.jar:git:Waterfall-Bootstrap:1.19-R0.1-SNAPSHOT:5c1e956:504]
message.txt by @hybrid bay: https://paste.gg/d5a443318deb4063badf0ea7c06dfeb1
[upd: solved ✅ ]
Hello guys!
I'm running a simple waterfall-proxy on the DigitalOcean droplet, and it suddenly just stop working. The problem is that I do not know why and can't find any logs.
I'm running waterfall.jar through the nohup, and also have written a script in /etc/init/ to launch my start_script on system load (after reboot).
Could someone guide me or give any advice on how to study the problem and find any solution or answers why the hell the app is just "stopping".
I have a full access to ubuntu machine, so can try to search for any logs you may suggest.
The only thing that is known, the "monitoring charts" that provides DO: https://ibb.co/RDC4fhb
So we see there a spikes at 9:30: memory goes down - I guess it's where the waterfall app is crashed, and the cpu load goes up - I guess it is also somehow related to that crash 🙂
[upd: solved ✅ ]
upd: there are nothing in the waterfall's latest.log – just the normal log, some connections, and that's it. No errors, no any crashes that points that the waterfall is going down.
It looks like there are the application process is just been killed, so the waterfall stops not "normally" and as a result there are no logs.
[upd: solved ✅ ]
prove me if I wrong, but I guess this is it, right?
https://pastebin.com/8nZ6LBb6
Out of memory: Killed process 921 (java) total-vm:2427700kB, anon-rss:204452kB, file-rss:0kB, shmem-rss:0kB, UID:0 pgtables:684kB oom_score_adj:0
[upd: __un__solved ❌ ]
Ok, I guess I've fixed that.
But there is one thing:
**What is recommended memory allocation for the waterfall? **
Like, in the docs there are -Xms512M -Xmx512M setup in the example. But is there any suggestions on how can I calculate the right amount for me?
I have a "main" waterfall server hosted on my own serve machine, with 512Mb or 1Gb ram allocated. Not sure. And there are no problems with it.
But also, I have a "proxy" waterfall, to allow to connect players from different geo-region (just take that fact): so I've hosted my "proxy"-waterfall on the DigitalOcean droplet with only 512mb memory. Where I set up the -Xms256M -Xmx256M. Is that enough?
|| the case described above, when I have process killed because out of memory, there was -Xms256M -Xmx320M and in the console of my droplet there are only 466mb RAM available, so I guess that's why the app was killed ||
tab
I think you just need to update it because the method has been updated @hybrid bay
Same as skins plugins and that stuff, generaly i don't even recommend the usage of those type of plugins on the proxy.
I think running this "geo" proxy on a 512MB droplet might not be feasible. And, in many cases, running a proxy in a different region just to improve latency, doesn't actually increase latency, as you're still connecting players to the same backend serbur.
It works if let's say you're trying to bypass some short of firewall, but otherwise not worth the hassle in my opinion.
Hi, im not the dev on my team but my dev isnt here to help me with this issue, so i am trying to fix it for him
Does anyone know how to attach my restart script to bungee?
i start the server using ./start.sh but it crashes when i leave the screen on our sever (using tmux)
@ me when you respond please!
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.
that is the error
proxies often dont even need a restart, that error looks like hardware issue
The general idea is that you should start the process in the screen, and then attach/deattach to/from it when needed.
the screen isnt able to run commands, im just gonna wait for my dev
You have an issue with your machine
was a config issue with the MOTD
invalid characters was causing it to crash
haha
Hey guys, I don't know if it is possible, I've seen some issues with the package handling on waterfall servers with Forge on 1.16.5. Just as a last resort....
Does anyone know if there is a way????
Either use lightfall and their client mod or try velocity with ambassador
Thanks will do!!
Hello everyone
Is there a way to bypass "You are already connected to this Server"?
I'm trying to force "/server lobby" even though the player is already connected to it
I'm doing that to fake player leave/join events without having to kick the player off the server
I know this is possible by just temporarily sending the player to a different server, but It won't be ideal
without editing the software, no
that's before you get into the potential issues that that would induce due to the order that connections occur
i.e. you connect to the new server before you dc the old one
actually my initial idea was writing a plugin or a middleware to ensure that the player won't actually disconnect, but I realized that this already exist (bungeecord, waterfall). unfortunately I've faced that issue and I'm exploring solutions
so basically switching servers is done by connecting the player to the new server before disconnecting him from the old one?
my end goal is forcing leave/join events without having to kick the player off the server
sorry
I'm plying with UUID switching, what I'm trying to achieve is the ability of having multiple player profiles per account within a server. Switching a player UUID can only be done on player join event.
Well, if you wanna change UUID you'll either need to make the server do that, or reconnect
You can switch UUIDs in play using the paper API iirc
You can't just fake the UUID of a player in-game, you'll need to "re-render" that player and this impossible without bugs
Does paper introduce a way to "refresh" player entities?
paper lets you set a players game profile in play
it will cause the player to respawn
that's interesting, I'll try that
afaik it should support switching the UUID, you will however need to deal with plugins
my concern is mainly about how plugins would handle that
or, wait, maybe not, I don't think it deals with the UUID internally
there is a UUID switcher that works with reflection
as I said before, it only happens when the player joins the server
I think I'll make a packet manipulator plugin that intercept "player kick" packets with the reason "reconnect" and prevent the player from receiving these packets and simulating (or replying) "player join" packets while showing him a loading screen
Ok, I literally spent the day configuring a forge server with Lightfall, pretty interesting stuff, learned a lot. I'm not kidding I spent over 8 hours just tinkering and playing with the configs. I only got it to work without any mods installed but the Lightfall client. It might be one of the mods, gotta review them individually, but it's a PIA since it's mostly Better Minecraft (>150 mods + their dependencies ect.)
I can't find Ambassador anywhere, since it's such a common word, no relevant results have popped in my searches, any recommendations.
BTW the issue Lightfall it's giving me is that it kicks me out of the server saying that it shut down and it can't connect with the fallback server...
It's been over 8 hours......
for ambassador, search in the velocity channels, there should be a link in there somewhere, maybe more so the dev one
outside of that, generally a huge level of "glhf", mods and proxies, especially bungee based proxies, are generally asking for some issues due to how they work, there are sometimes ways to marginally improve the situation but I cant comment too much on forks
Any error in the log? And yea as cat said, search for ambassador in the velocity help or dev channel
I’m assuming this bug exists in lightfall because brigadier isn’t capsulated
actually i setuped a bungeecord network
i have one proxy one hub and survival
so when i am joining the server its showing
this
showing this
and the ip is everything correct
For me seems a missconfig
Like ip or port/firewall or just your backend is unaccesible
So wht i have to do
Dealing with it?
I said you the possible stacktraces, deal with them.
idk how to
thats why i came in in help section
I mean, i told you the stacktraces
stacktraces?
Look at your log file
K
It shows the samething bro
it generally shows that and the actual error
if theres nothing more there I can't help you
@royal hawk @reef fulcrum
Thanks for all the help, still playing with the configs. As of now, I'm able to set it up without any mods. At least it connects to 1 server using the Lightfall port. It seems that it's not "seeing" the second server. It should be a matter of "config playing" and log file digging...
I'll keep you posted. It's been a fun project so far!!!
You should seriously try velocity with ambassador. The lightfall project is questionable even when you’re not biased as I am
Yeah, that's my project for the day. I'm literally at work with my server open on my left monitor and my home computer open on my right one running minecraft and testing configs (both remotely XD). Just a tiny amount of work done on the central monitor XDXDXD
Is there a different command to move from server to server usually in bungee its just /server but its not giving me anything. Maybe Ive got it set up and im not using the right command...
that would be so funny :/
BTW ITS WORKING!!!!!!!
Now starts the gruesome task of testing Mod by mod until it works...
I've got 2 Forge 1.16.5 (running Lightfall) connected to an Arclight Server and I'm able to switch between servers with /server. This has been quite the journey
Thanks everyone!
https://mclo.gs/M2zcUOU
is this a waterfall specific attack?
only included a small section of the log but there were probably a couple million of those
And I claimed victory waay too soon...
Where can I download Waterfall 1.19-R0.1-SNAPSHOT?
which one supports 1.19.1?
use .2
Is their a way I can get ride of this? Its in the plugin folder but get an error when trying to use it
I use Advancedban but that doesnt seem to be taking over
Gives me this error no matter what I do!
Hellow,
Is it possible to hide the number of players of a specific server in the multiplayer menu?
I would like this because I have a dev server connected to my bungee network and the players see that there is a player that they do not see :/
In fact it's weird, my admin account is hidden but when I connect a double account with player permission it is visible, but I don't know what is the action to do to hide it too (I suppose a permission?)
don't use ping passthrough if that's what you're using
otherwise, as with anything, "use a plugin"
What is "passthrough" ?
ping passthrough
it is in false
if you want more control over what shows up in there you'd need a plugin
But it doesn't really explain why my Admin account is properly hidden from him when I don't seem to have anything special with it x)
There is no hiding of that stuff outside of the clients setting
waterfall either passes the ping request to a server, if ping passthrough is enabled, or answers the ping response itself using info from the proxy
if you wanna adjust that stuff, you'd need a plugin
You need a plugin that would control the ping request, the motd thing is part of that
One message removed from a suspended account.
yes
One message removed from a suspended account.
which java ver should i use for my proxy server
I mean, waterfall supports 8+, I recommend using the latest lts however
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
You can use any other port, also you can set-up a DNS record to avoid having to use IP:port
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
There should be priority section in main config, just set lobby as first there
And also remove reconnect module or how is it called
Otherwise it will be sending players to their last server
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
Good evening.
I have a small problem with anti-cheat in the kernel.
I would like to ask professionals how to cut it out or turn it off completely, since I don’t really understand Java
(Version 1.12.2 or 1.17.1)
Добрый вечер.
У меня возникла небольшая проблема, связанная с античитом в ядре.
Я хотел бы узнать у профессионалов как его можно вырезать или полностью отключить, так как не особо разбираюсь в Java
(Версия 1.12.2 или 1.17.1)
- 1.12 and 1.17 are no longer supported by us, assuming you mean the actual server software
- if you're talking about the server, wrong channel (not that you'll get support anyways)
Then it's being caused by one of your plugins/reconnect module
Is there any option in bungeecord or waterfall config to kick players to lobby if spigot instance get down instead of full kick from proxy?
There are plugins that do this, maybe the fallback servers but I use plugins so idk
One message removed from a suspended account.
Hey,
I'm running a proxy on minehut which redirects players on a server hosted somewhere else. I've set the online-mode for the proxy to true while for the spigot server to false. I've also added BungeeGuard. However, I noticed cracked accounts are able to join which I don't like. I can see that they were joining via the proxy directly which shouldn't be possible as I enabled online-mode. They always got a dot infront of their name (so it's like ".NAME123"). I tried to reproduce it with TLauncher but it doesn't work for me so I'm wondering how they're doing it.
That's geyser
What do you mean?
Those are bedrock users joining through Geyser's floodgate https://wiki.geysermc.org/floodgate/
Oh, is it enabled by default? Cause I haven't set it up or installed anything.
https://minehut.zendesk.com/hc/en-us/articles/4637018652179-Can-I-play-on-Bedrock- looks like it is, yeah
I see, guess it's the best to contact them directly and ask if it's possible to disable this feature. Thanks for your help! 🙂
really they force geyser?
Dumb question, can waterfall forward to a mc server on our domain or does it need a static IP?
pretty sure dns works
Do I need to configure it a special way or can I just put billis.tk as the IP for one of the Minecraft servers? Also are there any issues running the proxy on a separate box from client/server?
no to both
you can put a domain as long as it resolves to the correct IP iirc
and yes you can run a proxy separately
just make sure you secure the backends and configure your firewalls properly
I want to join the server from one Launcher but it does not join the server and it gives warning in console: [15:02:42 WARN]: [/] <-> InitialHandler - overflow in packet detected! A string was longer than allowed. For more information, launch Waterfall with -Dwaterfall.packet-decode-logging=true
What should I do?
Login is provided from TLauncher, but when I enter from another Launcher, it gives a warning on the console
piracy is not supported
One question, can I connect a FORGE server to waterfall?
Same answer as with velocity.
xd moment, mojang want's to keep offline mode
yes
Definitly yes, they can remove it when they want.
If they don't remove it is because it is profitable, many people start in offline mode and end up buying the game. Pure and hard marketing that is not supported by people who create external software is totally different
thats like saying Apple doesnt release the same iPhones every year. Yes they do.
they have their own reason. One of the main one is to keep support for older versions where its not possible to authenticate with Mojang servers.
Second reason (tho im not sure they consider this as a reason) is server support. Proxies like waterfall/velocity still need servers to be in offline mode to work
will waterfall work with forge or should I move to velocity?
help pls
delete the modules folder and try again, otherwise generally work out what's borkin them
the same
Using waterfall?
Cool
Oh, I see
Update waterfall
yes
i do
You're literally using a 2 year old build of waterfall
this is a new waterfall setup/network, not sure if this is waterfall issue or paper/server issue, but i can connect to the default/lobby server fine but when i try to connect to the other 2 servers iv setup, i get this error message
99.999% of issues like that is often down to plugins, many times on the proxy, mangling packets
Viaversion?
message.txt by @merry root: https://paste.gg/4455168b622a420bb0fbbb9f1f7e6d65
I live in Turkey, minecraft in my country 500 currency in my country, no one can buy original due to economic problems
Bots are all germany location how can i close the logins from germany?
.
Not all type of exploits are fixed so still not the better option. I understand that people here doesn't support """PIRACITY"""
ofc
But, like
You can't open the gates yourself and magically expect people will care
because you didnt go through minecraft authenithecation servers?
This is competly stupid as my view, Mojang doesn't want to remove offline servers.
i remember notch allowed you to "pirate" the game since it usually lead to people purchasing the game
Is it possible to setup a waterfall network as a server within velocity, im trying to do sharding things.
"yes"
- such a setup is generally dumb
- Waterfall, and bungee, don't support player info forwarding on incoming connections, so you'll need to deal with IP/UUID stuff yourself
Why not two Waterfalls or velocity?
Does that work
Yes, you can use RedisBungee for example to split your proxy into multiple instances
I'm hacking together a sketchy sharding solution
Whst is vanilla+?
I'm getting an error like this, what do you suggest I do?
hi, i have a problem with my proxy. sometimes waterfall start using 2000% of cpu with no reason
https://spark.lucko.me/reXnT70k87
Use Waterfall
DDoS attack?
I feel that a ddos related to he's minecraft layer can reach that without protection or offline mode
How can I optimize my waterfall server to handle more than 50 players in 2 cores
Cuz it's lagging
there isn't any real "optimisation" options available
You can disable things like entity metadata rewriting, but, for the most part, it's just copying data around
when I create a server and want to connect to the server, the following error comes up and I get timed out. only happens when I start the server for the first time. Does anyone know what could be the reason? or is it due to waterfall at all? https://pastebin.com/raw/XcBGUVci
error s like that are generally down to things screwing with the protocol, etc
with which protocol?
the minecraft protocol
so it's not because of waterfall?
Basically, you entered an illegal state to the proxy
as many times as I've seen that, I've never been able to reproduce it in waterfall, or work out how somebody could trigger it
do you have any ideas how to fix it?
generally it's induced by plugins messing with internals
outside of "reproduce without plugins", no
okay i'll take a look at this. thanks
nothing to do with waterfall
and it tells you what plugin is borked
(make sure that WG is installed, otherwise, remove it and find something that's maintained or whatever)
Its java, it uses as much ram as you tell it to use, and then its own memory for java itself, etc
aha ok thank you
How do i set the default server?
By default already one is seted i think, just replace it?
Hello! When setting up the waterfall, it is only used for the bungee server folder while the other servers use regular paper jars right?
If this isn't the case then I may have messed something up with EssentialsX UUID because of that or some other waterfall issue
Make sure that you read the bungee install guide, primarily, making sure that you setup ip forwarding
Will check rn
I checked and as far as I saw there is only one Bungeecord config file. IP-forward is set to true for that. also:
- Connection-throttle for all bukkit.yml across server is set to -1
- Bungeecord is true for all spigot backends
Not sure if there is something else I'm somehow missing
Which file would that be referring to? I'm using the waterfall proxy and dont see that in waterfall.yml
in config.yml
As far as I see, it's all setup
The config is only in the bungee folder which is the one that has the waterfall.jar
the other folders have paper.jar's
You'd need to define what's wrong
How fix this is? I want to go to the server with mods from the vanilla lobby. I installed forge and the desired mod on my client. But it still doesn't connect me...
I can go with mods on a vanilla server. So why can't I switch from vanilla to mods? What kind of nonsense? On the client, I need to eat everything.
Let's take a plugin as an example ViaVersion. If I put it on my lobby, then this server can be accessed from any version, even 1.8. But, for example, I will make a survival mode, where there will be an entrance only from version 1.19.2, then those who logged in from 1.8 will not be able to enter it. That is, for this they will need to install version 1.19.2 and only then will they be able to log in.
And with mods, I assumed a similar principle. That is, those who installed Forge and mods on the client will be able to enter, and those who do not, respectively, will not be able to.
1.13 forge made changes which are not trivial to support for proxies, especially if you wanna be able to switch servers
Is this problem on all proxy cores? Or maybe somehow fixed for example on an velocity?
Do I understand correctly, if I use a server with mods on version 1.13 and below on an waterfall, will the connection between servers work?
you want modded servers below 1.13, not on 1.13
also tbh just don't try to make them switch between servers (or at least not between servers that don't have the mod pack and have it) and you should be fine-ish
Yes, it's a problem on all proxies, forge made changes which brokes with how proxies work
theres a few forks of the mainline proxies which have client mods which try to hack back in support
Use ligthfall
It depends on a mod to work, but that's all if you want to use modern forge
It should be latest
That's not latest
hey, i'm having some issues with waterfall
- it seems to be defaulting to survival even if hub is above it in the priority list
- i get a cryptic error message when trying to manually connect to hub https://hastebin.com/bepacudefi.properties
Hastebin is a free web-based pastebin service for storing and sharing text and code snippets with anyone. Get started now.
message.txt by @tribal totem: https://paste.gg/57a2e9f48cb24b168f8c71a96f82c0d9
thanks... already did that tho
Something induced an oversizded handshake packet
most cases are basically doing things like connecting the proxy to itself, or hostnames which are too long for mc
check your config, etc
also, as for 1, that's probs config too; The proxy remembers the last server you where on, you can disable that thing by setting to force the default server in the config file
Is this normal for waterfall server? The bandwidth usage is very high
how many online?
Like 20-ish most of the time
yeah ok not normal
How can I fix it?
Seems a ddos
20 users would hardly spend 1tb in a month, a typical layer7 attack on your server could easily reach that in a matter of minutes or seconds depending on the power on repetitive times
I feel like that, only probable cause
I guess your solution would be to use some quick mitigation system if your host doesn't give it to you, improve your firewall or use some reverse proxy that filters the traffic before it reaches your machine
How can I do this? since I'm selling shared hosting but all of the proxy that use waterfall in my node have the same problem
only few that use velocity show normal usage
<-> InitialHandler - overflow in packet detected! A string was longer than allowed. For more information, launch Waterfall with -Dwaterfall.packet-decode-logging=true
wondering if anyone has seen this before with geyser -> TCPShield -> waterfall
You sent a handshake with more bytes on the string than on the limit, i just think that Geyser doesn't work with TCPShield on free plans, also run the proxy with the flags and send again it
I don't think Waterfall uses that bandwich, seems a tipic flood?
I had a "big" network running more than 600 players on the same proxy and the bandwich usage was much less than yours
Tipic flood?
Yes what i said you before^^
I think here i specified a bit
Should I buy anti ddos services?
Probably?
Wait
Is your server premium rigth?
I would recommend you use free TCPShield plan, wich can handle 20-30 players
Will this work?
https://github.com/theraw/The-World-Is-Yours
Bruh
Probably that all bandwich is the totally of all your hosters?
The one on the screen shot is the worst offender
There's more with blazing 40gb+ bandwith usage under 24hour
How much customers do you have
Spigot counts as traffic tho
This is from one proxy btw
BungeeCord + spigot = double traffic
I know that
I guess that you can switch to ovh then
Is there really no other choices? ;-;?
Will this work tho?
Look at your logs, and check if they are not full of random proxies, if so you know what caused the bandwidth to be so high
I guess? At a low scale should work
Okay
Which log?
My wings log?
Or the servers log?
Server logs
Alr
So to ovh it is then
Well dat's fine
Thanks for your help tho
Oh one more. Will it work of I just use anti ddos services. But they mostly offer l4 protection
L7 is quite rare and expensive
eh, well expensive because is harder to detect i think
yeah ;-;
the problem with ovh is, Indonesian biggest ISP seems to hated it
with player having +100ms pings
whilst the ones using other isp getting 20ms
with which datacentre?
well do a traceroute and see at which point the ping loss occurs
ok well I mean your biggest issue is probably that you don't have good connections between indonesia and singapore lol
pretty insane how there is only a small, single sub-sea cable: https://is.papr.best/gqUvFguV.png
ok that data is from 2014 so might be better now 
could just be a routing issue (like it going via the us or something)
idk about that. but any other dedi provider doing just fine with the ping
it's just sometimes ovh dedi just have huge ping
sometimes they're good
and I've seen ppl hating ovh to the bones because of this
again: check your traceroute


