#Linux Help

1 messages · Page 2 of 1

karmic basin
#

Like this 100%

#

on a fresh debian 12 container

vivid bay
#

but its for Ubuntu

#

you have to fix the repos

#

sudo add-apt-repository multiverse; is Ubuntu.

I mean on Debian its non-free

karmic basin
#

Ah my bad I used my command for the steamcmd install it was like this :
apt update; apt install software-properties-common; apt-add-repository non-free; dpkg --add-architecture i386; apt update
apt install steamcmd

#

i copied his guide starting step 4

#

steamcmd works out of the box

#

For me the errors that might cause the problem are these 2 that stand out for me :
[E 00:00:00,206] [steam] Failed to initialize Steamworks system...Generic Error
[E 00:00:00,206] [steam] Message from Steam: No SteamClient020

Should I simply install the steam client via apt install steam -y ?

#

I had the same error on a ubuntu 22 that's why I did a fresh install

#

Until now I used wine, but that is not possible anymore because every person who joins after the 3rd one gets 10k ping the first 3 have a normal 20 ping but everyone after laggs like crazy, so strange

vivid bay
#

steamclient itself is not needed

karmic basin
#

I did an ln link from the linx63/steamclient.so to the enshrouded server file no?

vivid bay
#

try it. idk atm the correct path, where to put the .so

karmic basin
#

No I dont need to try it cause I did it already ^^

vivid bay
karmic basin
#

I but the linux64 client in the enshrouded folder in the user folder and in the Steam folder nothing helps

#

I think I just need to know what id the steamworks program has in steamcmd to install it

vivid bay
#

app_id 1007

karmic basin
#

yeah tried didnt help sadly i put up a widows vm now 🤢

grand cloud
#

Hey guys, can someone maybe give me some info if there is a fix for the network problems (high latency) when using wine?

vivid bay
grim egret
#

this is showing up in the logs today "src/steamnetworkingsockets/clientlib/steamnetworkingsockets_lowlevel.cpp (2427) : Assertion Failed: SteamnetworkingSockets service thread waited 101ms for lock! This directly adds to network latency! It could be a bug, but it's usually caused by general performance problem such as thread starvation or a debug output handler taking too long." proton egg, server rubberbanding slightly with just one player on it

#

anyone know the cause?

#

with two players its closed to unplayable, server has been restarted a few times

grim egret
#

base load today on the server after building some is 250 % of 800 %, yesterday with a smaller buildings it had a base of 80-100 % when no one was logged in...

snow token
#

Lecssl Could not add component e3b96393 for enti
ty 611270
[ecss] Could not add component e3b96393 for enti
ty 611314
[ecss] Could not add component e3b96393 for enti
ty 611394

#

Anyone know why server load warning occurs with 3 ppl in server?

snow token
#

Make sure the ports are assigned to the egg server

cerulean mica
#

hey everyone, I tried out the pterodactyl egg for enshrouded today and the server never starts:

[app] start creation step Init
[app] start creation step Init_Steamworks
[steam] Failed to initialize Steamworks system...Generic Error
[steam] Message from Steam: No SteamClient020
[app] Failed to create Steamworks.
[app] start destruction step Init
[enshrouded] Close Log file
#

is anyone having luck with th eegg?

grim egret
# snow token Big building?

medium sized, like enough for 6 players, a crafting npc area etc, spacious but not castle like. think it was like 26x58m (in-game meters)... tore it down, problem still persisted, even after multiple restarts etc

snow token
#

So it didn’t matter after all?

golden dust
#

Kind of surprised there arent any native linux binaries for dedicated server

errant gust
#

Hey guys, I am currently hosting a serveur for friends, we are playing at 6, but when we are 6 one of us start lagging really really hard. We have no problem at 5 everything run somethly. You can find my server config below. What surprise me is that nothing seems over top when we are 6, the load average is aroung 7 (close to the top but got no cpu at 100%), the ram usage is really low, same for io (both disk and network). So my supposition is the cpu but not totally sure about it, do you some feedback/help to give ? I am running the server on linux with wine64 directly.

SYS-5-SSD-32
Intel Xeon E3-1270v6 - 4 c / 8 t - 3,8 GHz / 4,2 GHz
32 Go DDR4 ECC
500 Mbit/s

snow token
errant gust
#

thanks for your answer, if i may ask, do you know how i can install with debian ?

snow token
obtuse cedar
karmic basin
errant gust
#

Thanks a lot

vivid bay
still ember
#

anyone have any VPS/dedi machine recommendations? been looking into hosting and I have a tendency to be too cheap. So would appreciate any advice/recommendations anyone has. Looking to run pterodactyl over linux with the proton egg.

gritty willow
#

Being cheap is fine as long as you meet the requirements for the number of players you have.

#

I personally went with Vultr hosting then moved to Google Cloud because 300$ free credit

#

Their interface is more complicated however and I wouldn't recommend it for people not confident in troubleshooting

obtuse cedar
true mauve
gritty willow
true mauve
gritty willow
#

On Vultr when I was paying out of pocket it was between 80-100$ per month but lower stats

true mauve
#

I did basically the same thing with azure a couple years ago hosting valheim on credit. not sure if I want to go through the effort of setting it up though haha

gritty willow
#

The plan never was to play for an entire month so the costs made sense just for the convenience of a dedicated server

#

Azure would be a great option as well I imagine, because of free credit

#

Actually I don't know why I didn't think of Azure from the start 😂 I use it every day at work

true mauve
#

eh free money is free money

agile timber
#

Switched from the Wine based egg to the Proton based one... much better performance (less CPU usage, and no occasional short lag spikes for remote players). Nice! Running it on old Dell Optiplex i5 (with Proxmox, Plex and many other tasks)

obtuse sorrel
#

Hi, i'm looking for a guide to install Proton on an Ubuntu 22.04 server, does anyone know one please ?

true mauve
#

Has anyone figured out if there's anyway to make server announcements? I'd like to set up server restarts and would prefer to have a warning in game for a heads up

strong pulsar
#

Steal your users login, force them to use ssh, send xdialog popups! ^^ 🤭

vivid bay
snow token
#

I have mine restarting 4:30 am

true mauve
snow token
#

There you go

true mauve
#

Thinking about adding another to catch the update window during the week.

snow token
true mauve
jade quartz
#

Hello everyone, i ran into a strange issue. I'm using a dedi linux server for enshrouded that worked fine since the release. I'm using the PR3SIDENT solution (big thanks to them) for the server. But today i'm noticing an issue where the server can't be found in the server list when using filter by name. The server is running fine when i check it on my linux service and i can ping it just fine using the server window of steam. Does anyone ran into the same issue?
Thanks for any answers 🙂

vivid bay
jade quartz
obtuse cedar
#

Has anyone set up a server on proxmox? I can't get it to show up on the server list despite it being in my steam favorites and I can see that networking is fine and packets are coming in and answered correctly if I monitor it with tcpdump.

vivid bay
obtuse cedar
vivid bay
obtuse cedar
# vivid bay No, i never use UFW

Me neither, I manage security with iptables and the server is not showing up despite seeing network traffic incoming and outgoing. This is why I was wondering if anyone else has done it and if there were any ‘special’ actions needed to get it to work. The strange thing is that I can query the server using steam but it does not show up in-game.

vivid bay
strong pulsar
#

iptables +1

obtuse cedar
# vivid bay are you hosting at home?

No I host it on a dedi but on a separate VLAN that has WAN access and the traffic goes through a separate gateway and from there it gets routed out to the internet. You could say there are 2 chokepoints for traffic and neither of these points show any dropped packets. I must be forgetting something though, I am just not sure what it could be.

vivid bay
obtuse cedar
#

Yeah so what I thought I would do is just check if packets are coming in and if they are going out. So for the query port they do, I can see the traffic but I am not seeing anything for the game port. I also host palworld on it and that works fine but it uses Epic Games servers for their matchmaking, which also makes me believe that I am missing something to make iptables work with steamworks here (just speculating)

agile timber
obtuse cedar
strong pulsar
#

bloat ontop of some more bloat ^^

obtuse cedar
#

You're not wrong but sometimes we need fancy things 😛

strong pulsar
#

Sometimes, but i try to live without them as far as i can ^^ less areas to troubleshoot and headache about ^^

#

I just wait for a proper linux executsble with a working console command interface to setup things over ssh ^^

#

Not feeling the website gui thingie ^^

obtuse cedar
#

I hope Enshrouded will get a linux executable, that would be nice.

strong pulsar
#

Yeah ^^

agile timber
#

Our current list of servers (running in an LXC on Proxmox, which is also hosting other stuff like Plex.... on a 6yr old i5 small form factor Dell Optiplex PC)

image.png
mighty badger
#

any good solution for the Session failed for peer #0 with error 4.? error?

karmic basin
#

I found in my setup the problem, somehow i blocked in my network outbound communication and only allowed http/https protocol when I opend all ports to the outside it worked without a problem.

graceful goblet
#

Hello, trying to setup Enshrouded on my ubuntu server (22.04, heztner) with Docker (mornedhels image). I'm having trouble connection to it (good old Session failed for peer #0 with error 4)

But I can connect when Tailscale is running on my client. So i'm guessing it's a firewall issue or network interface issue but can't seems to find what is causing it.

I don't have ufw enabled and iptables is cleaned and flushed (input, output, forward at ACCEPT). And on the Hetzner side, I believe I didn't fail to set up the firewall 🤔

Do you have any idea for further debugging this ?

2024-02-08_12-24-35_vivaldi.png
gritty willow
graceful goblet
gritty willow
#

Weird, everything looks in order, I've used that image myself to host with docker compose so I'll say it's an issue on your end. Has to do with firewall or something

obtuse cedar
#

Use tcpdump to capture traffic on your game port and see what’s going wrong. Perhaps packets are being dropped.

graceful goblet
#

Alright, I found something, on Hetzner's firewall, port 32768-65535 was only accepting TCP ACK, when I removed the filter on TCP flags, I was able to go further in the connecting process but end up crashing the server

2024-02-08 16:38:04.494 supervisord: enshrouded-server STEAMPS3 - TCPPort::OnAcceptEvent (tcpport=0x7f6562d72e20), (socket=0x7f6562d730b8), (incoming.addr=<my_ip>:40209)(incoming_=0)
2024-02-08 16:38:04.494 supervisord: enshrouded-server STEAMPS3 - AsyncTCPSocket created
2024-02-08 16:38:04.767 supervisord: message repeated 36 times: [ enshrouded-server STEAMPS3 - AsyncTCPSocket created]
2024-02-08 16:38:04.774 supervisord: enshrouded-server *** buffer overflow detected ***: terminated
2024-02-08 16:38:05,479 INFO reaped unknown pid 73 (exit status 0)
2024-02-08 16:38:05,479 INFO reaped unknown pid 77 (exit status 0)
2024-02-08 16:38:05,479 INFO reaped unknown pid 80 (exit status 0)
2024-02-08 16:38:05,480 INFO reaped unknown pid 97 (exit status 0)
2024-02-08 16:38:05,480 INFO reaped unknown pid 108 (exit status 0)
2024-02-08 16:38:05,480 INFO reaped unknown pid 114 (exit status 0)
2024-02-08 16:38:05,481 INFO reaped unknown pid 122 (terminated by SIGSEGV (core dumped))
2024-02-08 16:38:05,481 INFO reaped unknown pid 124 (exit status 0)
2024-02-08 16:38:05,481 INFO reaped unknown pid 129 (exit status 0)
gritty willow
#

And which image tag of mornedhels/enshrouded-server are you using?

#

I recommend mornedhels/enshrouded-server:stable-proton if you're not using it already

graceful goblet
#

Forgot to update, got it working, it was the port range 32768-65535 needed to be open in UDP as well.

Spec wise, it's a i7 8700, 64GB, it's running perfectly. I built the image myself using the proton version yes.

#

Thanks both of you for your help fell_love

strong pulsar
#

Feels a bit overkill to open thousands of ports, kinda defeats the idea of having a firewall 🤔

agile timber
#

Yeah. I just have the 2 ports open

#

(but both TCP and UDP)

graceful goblet
#

Well I agree but I have no idea why without this range of port being open (UDP/TCP) it couldn't work 🤔

obtuse cedar
#

Hetzner's FW is weird, pretty much just that. You could also just disable it / allow everything and then use iptables.

strong pulsar
#

Errything is better to manually/not in a frontend imo, to have a clear view of what goes on ^^

graceful goblet
obtuse cedar
#

I have a bash script here that automatically downloads the latest version of Proton-GE and then extracts the files into the specified folder.

You could use this to automatically update proton without any kind of manual actions 🙂

Requires jq and curl packages

#!/bin/bash

# Set the target directory where you want to extract Proton-GE
TARGET_DIR="/proton/bin"

# Create the target directory if it does not exist
mkdir -p "$TARGET_DIR"

# Use GitHub API to get the download URL of the latest Proton-GE release
# that ends with .tar.gz
LATEST_RELEASE_URL=$(curl -s https://api.github.com/repos/GloriousEggroll/proton-ge-custom/releases/latest | jq -r '.assets[] | select(.name | endswith(".tar.gz")) | .browser_download_url')

# Extract the filename from the URL
FILENAME=$(basename "$LATEST_RELEASE_URL")

# Check if we got a valid URL
if [[ $LATEST_RELEASE_URL == null ]] || [[ $LATEST_RELEASE_URL == "" ]]; then
    echo "Failed to find the latest Proton-GE release with a .tar.gz extension."
    exit 1
fi

# Download the latest release
echo "Downloading Proton-GE from $LATEST_RELEASE_URL..."
curl -L $LATEST_RELEASE_URL -o "/tmp/$FILENAME"

# Determine the directory name inside the tar.gz (without the .tar.gz extension)
DIR_NAME="${FILENAME%.*.*}"

# Extract the specific directory from the tar.gz into the target directory
echo "Extracting the contents of $DIR_NAME to $TARGET_DIR..."
tar -xzf "/tmp/$FILENAME" --directory "$TARGET_DIR" --strip-components=1 "$DIR_NAME/"

echo "Proton-GE has been successfully updated."
strong pulsar
#

Looks solid at a look-over ^^

golden dust
#

Any word on ever getting native Linux binaries for dedicated server?

#

Started playing again with some other friends but looks like nothing yet

strong pulsar
#

No word on that sadly

south ivy
#

has a lot of quality of life built in

#

I have it set up right now with podman+quadlet(systemd) and it self-runs, self-updates, self-restarts

#

completely hands off

#

pretty good performance too

golden dust
#

That's the one I'm using

#

😢

#

Everything else is great

#

But the performance is so bad. We have issues with only 3 people and 5 it is painful

south ivy
#

I don't have problems with 3 people

#

I used to have server load notifications in-game with just the WINE based image

#

but when I switched to proton, even before release 1.0 of the image, it was all fixed

#

Maybe its a resource/hardware problem?

#

or network problem

#

I host from a fiber 1g/1g up/down connection

golden dust
#

Started having htop open when I played and whenever the server was getting really bad I noticed one core is at 100% while the rest are fine

south ivy
#

I have mine running on a 12900K so maybe my CPU is just able to handle it

#

I suspect being able to leverage more cores would obvs improve your experience though

#

ok for now, but nobody is logged into game atm

image.png
golden dust
#

I have it running with 20 cores

#

2x Xeon E5-2650

south ivy
#

Well I mean, wine isn't going to help distribute the process load across the cores

golden dust
#

I'm running it with proton

south ivy
#

the 2650 is a very old Xeon nowadays

#

you can get better perf out of a lot of mini-NUC Atom/Celeron dies

#

from 2022+

#

2650 is from 2014/2015 era and it's low clock

golden dust
#

Hmmm

south ivy
#

the 12900K is literally like 4x core efficient over the 2650

#

Is this like, old homelab gear?

golden dust
#

Yeah I figured having more cores might help but it just fixates in the one core

#

Dell R720

south ivy
#

yeah at most you're gonna be running off of one core

#

yeah I used to have an R720 based homelab... 4 years ago

#

it's almost a decade old

#

they're power hungry, heat-heavy boys

#

not really worth running anymore

golden dust
#

It is heat heavy lol

#

I do have an unused 4690k machine you think that would give better performance?

south ivy
#

That NUC is 4x as powerful CPU wise and takes maybe a quarter of the TDP

#

Minisforum makes it, highly recommend

#

they are releasing the MS-01 machines with 13th gen intel chips if you are interested

golden dust
#

I was looking at some of the Intel NUCs

south ivy
#

when I ditched my R720 years ago, nobody wanted to even buy it

#

I had to donate it

golden dust
#

LOL

south ivy
#

Intel NUCs are overpriced imo

#

I used to have a fairly sizeable rolling 24U cabinet with R720, Unifi stack, multiple R620s

#

got rid of it all

#

went to NUCs - quiet, doesn't destroy my electricity bill, and plenty fast

#

I think your perf issue is because of that E5 2650 😉

#

def recommend moving on from it

#

they are costly to run

golden dust
#

I guess I'll try the 4690k since I have it on hand

#

I will look into some NUCs for the future though

south ivy
#

not 12900K, that's in my desktop, but the server is on 12700H

golden dust
#

Oh nice

south ivy
#

I just got the barebones and slapped in a 64GB DIMM kit in there

golden dust
#

What do you run on it? A hypervisor?

south ivy
#

It's just running fedora server 39, I have 3 Minisforum MS-01's on the way since they're shipping the first batch today actually

#

that will be a 3-node bare metal Openshift cluster

#

i'll run Kubevirt on there

#

and migrate this game server to it later after migrating the PV

golden dust
#

Oh wow neat

south ivy
#

I'll have the option of running a headless windows nano VM on kubevirt when I get my cluster set up

#

I have a single node openshift cluster on this UN1270 NUC before

#

but it was just too clunky

golden dust
#

No PCIe slot looks like tho

south ivy
#

not the UN1270, but the MS-01's do

golden dust
#

I kind of want something for NAS options

south ivy
#

the MS-01's also have 2x 10gbit, 2x 2.5gbit, U2 NVMe slot, 40gbps Thunderbolt as well

#

can do literally anything with that kind of setup

golden dust
#

I have an HBA card

#

Kind of want something to be able to slap that in

south ivy
#

doing ZFS?

#

or just FreeNAS

golden dust
#

Yeah, truenas scale

#

So zfs

south ivy
#

I am going to be doing Ceph across 3 nodes

#

with the triple MS-01's

#

put a 2TB NVMe in each

#

full redundancy

#

any compute node can go down and still have full parity

#

I used to have ZFS on linux with my R720 since my 720 was one of those 16 bay SFF drive chassis

#

I did a flashed HBA to just JBOD+ZFS

#

but the SPOF is still the server running it, as well as the HBA

golden dust
#

Good food for thought, the MS-01 looks interesting

#

But I'm in the middle of upgrading my desktop so that will be a later solution

strong pulsar
#

Did linux server help die out? ^^