#dedicated-servers

1 messages · Page 55 of 1

flat heron
#

now it fails to initalize steam again

#

ok reinstalled everything works again

flat heron
#

still has that error, which causes it to auto close connection

deep turret
flat heron
#

k k just wanted to make sure

deep turret
#

it's the so called loopback address, it is only on the very machine you are connecting

flat heron
#

do you know why that error occurs?

#

i think it still works if i readd the server tho

deep turret
#

the connection timeout you were talking earlier is ok
is this the server logs ?

flat heron
#

yes

deep turret
#

and also sometimes the client miss the server and show it offline, it's a feature bug of some sort, you can restart your client game (not the server) when that happen, to see if it show the server up again

flat heron
#

ok then its probably just that ig

deep turret
#

or delete/re-add the server

flat heron
#

thats what im doing

#

k i think its working then, ty

deep turret
#

do you have ports setup in the server launch script ?

flat heron
deep turret
#

what port did you setup

#

50568 is not default at all

#

are you using update 7 or update 8 version ?

flat heron
#

im using update 7, i set up the default ports

#

also, the situation with my router is critical. i can connect to the server (in my home network) trough my local ip (192....). the router shows an ip titled "ip address in the internet" behind the opened ports. i can also connect to the server with that. my friend however, who is not in my home network, can't connect trough that ip and neither trough my standard external ip

deep turret
flat heron
#

full log? i can get you that

deep turret
deep turret
flat heron
#

its the ip i get when i google whats my ip

deep turret
#

at first

flat heron
#

starting commandline is -log -unattended

#

if i add the ports in there, it doesn't work

deep turret
#

then you still didnt set the ports as I told you

#

keep the ports in there and tell us when it does not work smh

#

what is the point of asking you to try something if you give no feedback

flat heron
deep turret
#

that lines gives nothing out

#

if it does not work like that it means you probably have multiple instances of the server running

flat heron
#

oh

deep turret
#

or you didnt setup ports correclty

#

but I still dont know because you didnt share your starting commandline with the port setup

flat heron
#

.\FactoryServer.exe -log -unattended -Port=15777 -ServerQueryPort=7777 -BeaconPort=15000

flat heron
#

if i remove the ports, it works in my home network but still not outside of it

deep turret
#

also it needs to be before the -log -unattended

flat heron
flat heron
deep turret
#

.\FactoryServer.exe -ServerQueryPort=15777 -Port=7777 -BeaconPort=15000 -log -unattended

flat heron
#

huh

#

ok changed that, lets see

#

it works in my network but not with the extern ip

#

do you need the log?

deep turret
#

did you forward the 3 ports ?

#

in UDP mode ?

#

and opened it on the server's machine firewall as well ?

#

@flat heron

flat heron
#

firewall

deep turret
#

so I can't really read it

flat heron
deep turret
#

ok then it must be fine, as long as it's listed as enabled

flat heron
#

i really have no clue why it doesn't work

#

do i maybe need like a multihome parameter or smth?

deep turret
#

can you try to check the Selbstständige Portfreigabe box ?

flat heron
#

ofc

deep turret
#

also did you communicate the connect port to your friend or you let them use the default one ?
if they use the default one its fine, but if you communicated the wrong one it is not

flat heron
flat heron
deep turret
#

do you have a save button somewhere on the setting page

deep turret
flat heron
#

you know what i will set it as an exposed host just to try it

deep turret
#

you can reload the setting page to make sure the settings has been registered

flat heron
deep turret
#

is DESKTOP-6OOGP3P your server machine ?

flat heron
#

even as an exposed host, it doesn't work

deep turret
#

wait I just recall you are using you own pc as server probably

flat heron
flat heron
deep turret
#

ok so make sure the firewall settings and the network forward are really setup on the server machine

#

not your usual pc

#

(easy for you to check, but not for me so I still ask)

flat heron
#

its the one in teamviewer

deep turret
#

you can do a ipconfig cmd command on it to check if it match

daring fox
#

yes please that would be wonderful

flat heron
#

nope, thats not the one

deep turret
#

run ipconfig on the server machine and see what MAC match and you will be 100%

flat heron
#

desktop 600 is the server

deep turret
#

you may want to try -multihome=0.0.0.0 then but that's weird

reef lily
#

do updates come out at a consistent time of day when pushed out? trying to set up my dedicated server to check for updates daily, preferably earlier in the day to avoid interrupting players

deep turret
#

it would still mean you are in fact running experimental to me

deep turret
flat heron
#

bye

reef lily
#

appreciate the reply. then ill just have to check early in the day, and set up a remote connect vpn so i can remote in and manually update if needed. want to keep the downtime as low as possible for my friends

deep turret
deep turret
flat heron
#

Just to make sure, the ip I get when i go on that website is the ip my friends need right?

#

Man

deep turret
#

wait a minute I will give you a link to double check your external ip

#

can discord be more bugged than that for message links ??

flat heron
#

Ip is correct

deep turret
flat heron
#

I know

reef lily
twilit fox
#

Maybe not.. but for some reason it works just fine x: I'm guessing the other two parts are opened by the server when it receives a response on 15777

#

so client calls with a valid packet on 15777 server answers either on that port or one of the others. if a connection is opened from inside the network, the port will be opened automatically.

long current
#

Guys, help me, I can’t find the Game.ini file on the dedicated server and make more than 4 players...

grim mica
coral flame
#

running windows satisfactory ded server, U8 experimental, any way i can change the FG.AutosaveInterval, i know the console commands are not working, i tryed putting mIntValues=(("FG.NetworkQuality", 3)) and mFloatValues=(("FG.AutosaveInterval", 3600.000000)) in GameUserSettings.ini, all that did is crash the server when a user joined, i had to delete the ini file and have it recreate it for it to sort itself out and run properly again, any ideas?

grim mica
coral flame
#

i'm using notepad++ with UTF8 as well, is it picky or anything ?, i tryed removing the variables as it crashed and it still crashed doh :), like i said i had to fully delete the file and the server remade it

grim mica
#

Do you have the logs from when it was crashing?
This is what I am running, I was running 3600 but changed to 1800 after the 5.2 update started causing more DS crashes

image.png
coral flame
#

right, i will check it with just the one variable in a bit

#

dam, ded server still crashs' with that one setting changed doh 🙂 lucky i backup the .ini file this time

grim mica
#

What does the factorygame.log say is causing the crash?

#

Are you starting it with a visible console and it just loads and then disappears or as service?

coral flame
#

i'm just running it in remote desktop, i can see the error log, i crashes out just after a user trys to enter server

#

[2023.10.26-20.23.34:450][380]LogGameState: Match State Changed from WaitingToStart to InProgress
[2023.10.26-20.23.34:450][380]LogNet: Join succeeded: Nessus
[2023.10.26-20.23.35:896][381]LogThreadingWindows: Error: Runnable thread Background Worker #15 crashed.

#

my GameUserSettings.ini looks exactly like your one above in screenshot, very odd

#

i edited the GameUserSettings.ini back as it was, it still crash's

grim mica
#

Thats.. interesting - can post the full log?

#

Even better, if you go into the Crashes folder there is an XML in there, let me find one

#

SatisfactoryDedicatedServer\FactoryGame\Saved\Crashes\

#

There should be a crashcontext runtime.xml

coral flame
#

ya can do, lemme grab...

twilit fox
#

Oh yeah for those of you with windows servers... here's a power shell script to check if it's running.. and restart it if it crashes.. maybe useful?

grim mica
#

That is weird.. your crash context shows 0 memory free

#

Is this bare metal Windows or a VM?

twilit fox
#

change the path of your server install.

image.png
#

I'm running it on a VM.. shouldn't matter tbh

grim mica
#

Sorry, wolf not for you 🙂

coral flame
#

server is a potato, windows server 2019 + 'game 2' which is a bit chunky at this point, server has 32gig of ram, it runs ok normally

grim mica
#

I use NSSM and that keeps it always running and dont need to login

twilit fox
#

try running the steamcmd install + validate command again

coral flame
#

it's a bare metal install

#

not in a VM

grim mica
#

His issue is not to do with the install, once the mfloatvariable is set for save interval the server doesnt start

#

but does start with a newly built ini file

coral flame
#

yes, this is correct as i can see, kinda odd

grim mica
#

Are you sure you are stopping the server when you are changing that value? Its as if that startup just didnt have enough memory free/available. Could also try to set the ini and then reboot it and try to start the server to rule out any memory allocation problems

coral flame
#

if i delete the .ini file and run the server, it'll run ok (ya the first client actually crashes instead on connect)

grim mica
#

Seen a bunch of these HAL FMalloc recently and its either CPU extensions not being exposed due to virtualization or not enough memory

coral flame
#

ya i'm editing the ini file when dedicated server is not running, i know it overwrites it

grim mica
#

Interesting, just checked one of my recent crashcontext and it also shows 0 available physical/virtual memory lol

coral flame
#

so what you are saying is: computer says no ? 🙂

grim mica
#

Well I was looking at these a bit more when the first 5.2 came out and all of the 261167 or older crashes list this but the new ones dont, so perhaps not a good indicator that something is wrong

#

If you pop open Task Mgr and look at the Performance tab and Memory specifically, what do you have "Available" with the server stopped?

coral flame
#

ok, i'm running server atm, but not connected yet, the ded server is uing about 3.2gig ram, the overall server is 13.7/32.0

grim mica
#

Should be plenty free after a stop and start, that is an interesting problem.. lol

coral flame
#

ok, it crashed again, with no settings changedf, this time i have not touched the ini file, but i have copied a backup file myself (the dedicated server runs in my username also, no weird setup), ok thats odd, i will now delete ini file and have it remake a fresh one itself

#

oh ya the client crashed as well that time lol

grim mica
#

Thats when you try to join a client? The hell..

#

Oh thats interesting, what is the client crashing with?

coral flame
#

<ErrorMessage>Unhandled Exception: EXCEPTION_ACCESS_VIOLATION reading address 0xffffffffffffffff</ErrorMessage> mmm not very useful

twilit fox
#

could that be something with a full drive or write permissions?

coral flame
#

maybe permissions thing, it's very odd, i'm letting the server make the ini file and editing that

#

ok sanity check, i deleted the ini file, run server, connect to game, neither server or client crashes, works fine

grim mica
#

Was just thinking disk space or permissions.. but if it works when it makes the INI - makes no sense

twilit fox
#

unless when they edit the ini it changes the perms on it

grim mica
#

Using NPP, if it doesnt have permissions it will ask to run as admin iirc

coral flame
#

i could run server as admin, very odd though, it's running under the same user, me

grim mica
#

Though I would think you are running the exe in the same user context where you are changing the INI, so if it is should be the same perms

coral flame
#

c:\steamcmd has not been messed with permission wise

twilit fox
#

you could try to run the server as admin and see if it works maybe

#

windows normally isn't picky about perms.. linux is.. windows server my act a little bit between the two maybe

coral flame
#

ya will do now, i will edit the currently 'working' ini file with no changes whatsoever, just open and save... 🙂

#

not, ran as admin, still fails lol

#

i opened the ini file with notepad++, had to add 1 space char, then deleted it, pressed save, notepad++ is not running as administrator

grim mica
#

Only other thing I can think of is the CPUs age, but why it works when the exe generates the ini vs the edited version is beyond my understanding of Unreal engine mechanics

#

Doesnt make any logical sense and doesnt seem you are doing anything incorrectly.. so not sure

coral flame
#

ya given up for now lol, i'll just make do save every 5 mins doh, prob is the system is a potato as i said, it currently takes 45sec to save 🙂

daring fox
#

was wondering if someone could please help me when they get a chance i had a dedicated server running for 2-3 days just fine minus a couple of crashes but now all of the sudden everytime i try to start the server up it instantly shuts down..... and i cant figure out what i did didfferent or whats changed that now it wont run

coral flame
#

it is using most up to date build etc

#

well, on steam

grim star
#

im having an issue if anyone can help thatd be great, just got a nitrado server. I got it all set into my server manager and created a game then when I click join it says "you are not connected to server" then immediately it goes to a loading screen then says "your connection to the host has been lost" and happens everytime

#

have tried multiple other saves within the server and also changing my network settings within the game

grim mica
daring fox
grim mica
#

Reaper that sounds like a connectivity issue

grim star
#

would that be on nitrado? or my connection?

grim mica
#

Could be either honestly but given its a paid hoster probably your side or your path to them

#

Im not familiar at all with Nitrado as I have always hosted my own but do they have any support system? I would think they could help you with that

grim star
#

I tried contacting them but no help

grim mica
# daring fox

[2023.10.26-22.04.52:129][ 0]LogServer: Bound Server Query Socket to address '[::]:15777'
[2023.10.26-22.04.52:130][ 0]LogServer: Bound Server Query Socket to address '0.0.0.0:15777'
[2023.10.26-22.04.52:130][ 0]LogServer: Error: Failed to initialize server. Could not bind any addresses to port '15777'.This is very likely caused by another server instance already running. Use a different port number with the -ServerQueryPort=<port> startup argument if you intend to run multiple instances.

Your server is failing to start because the port is already in use

grim star
#

I took that clip cuz im not very good at explaining whats going on, but I am connecting to it in order to get into the server then as soon as I try to join it loses connection somehow?

grim mica
#

If you want to DM me the IP I can try to connect to see if its their or your side, otherwise not sure I can help too much

#

Its not completing the connection to the server somehow

daring fox
grim mica
#

Find the thing that has 15777 and then open task manager and find the matching PID (process ID)

#

Actually, sorry I forgot Satis is all UDP - one sec

grim star
daring fox
grim mica
#

If netstat lists the port, there is a process running as that is a current list

daring fox
grim mica
#

Are you running task manager as an admin perhaps? hmm

daring fox
#

i am

grim mica
#

I suppose you can try to kill the task manually with the PID

daring fox
#

i tried normal and admin just in case

grim mica
#

From the command type 'taskkill /pid 4840' and see what it returns

daring fox
#

ok it killed task and i tried to run server again and same problem though

grim mica
#

What did taskkill return? If you re-run netstat does it still show up in the list?

#

If something is using the port it will never start because it has nothing to bind to which is a networking requirement

daring fox
grim mica
#

If taskill didnt do anything try rebooting the box to see if it shows up again in netstat

#

Certainly seen some applications be weird with connection listeners not releasing in the past

coral flame
#

soooo turn it off and no again ? works every time ! ... sometimes... 🙂

daring fox
#

i tried rebooting unfortunately no fix with that but i think i found the issue

grim mica
#

Almost the first rule of IT.. turn it off and on again lol

#

I dont think 15777 is strictly unique, want to say that some other game I was hosting was using that.. maybe Night of the Dead?

coral flame
#

is it an unreal server default or something ?

grim mica
#

Could be

daring fox
#

yep found it.... somehow my old nssm service starting running on its own again so it was automatically running another server i just uninstalled the services in nssm rebooted and its working now thx for your help gunther i greatly appreciate it i hope to buy u coffee someday

grim mica
#

Ah, that makes sense.. sort of - still not sure why it wasnt showing up in the pid list on task mgr.. oh well. at least you got it sorted 😉

daring fox
#

it was under another pid im assuming

#

idk tho

grim mica
#

netstat -ano is all processes on the machine, hence it giving you the PID for it

#

If you run taskmgr as admin it will list processes from all users

daring fox
#

yea thats odd then

grim mica
#

Explains why taskkill didnt do anything, NSSM will just instantly restart it lol

daring fox
#

yea

#

but seriously thank you u really did truly help me find and fix the issue

grim mica
#

No problemo!

deep turret
coral flame
#

it's ok, i gave up for now, if i edit the file at all it makes the dedicated server blow up

deep turret
coral flame
#

ya was thinking something like that, had enough for now, will try again tomorrow maybe, it's notepad++ and set to UTF-8 and windows CR LF

twilit fox
grim mica
thorny shore
#

There has been someone jumping on my server deconstructing shit at random. I assumed the password requirement was set with the save but was wrong. I guess someone decided to port-scan me to see what I had open or something. Not happy about this, but I re-set the password in server management. I will find their body.

red edge
#

Someone have experience issue with gportal experimental server, i try to start new game server, on gportal server is online and i can see it an log on it right after a restart but after less than 1 mn i'm disconnected and server appear offline on server manager console. but gportal say it's online. sometime in server manager i can see le server pass online with join button and immediatly disapear and server appear offline again

timid lintel
#

how do you get linux dedicated server to stop trying to use Audio? [2023.10.27-16.47.01:682][556]LogAkAudio: Error: Failed to post invalid AkAudioEvent on component 'AkComponent_2147478788'.

bronze silo
#

Posting here for comedy reasons: [2023.10.27-17.16.36:883][284]LogGame: Total Save Time took 225565.669 seconds

grim mica
#

Talk about a save hitch

errant gull
#

Looking for some help on connection issue with latest experimental 8 version on a headless Ubuntu 22.04.6 LTS server running on our LAN. We (hubby and I) have no issues running and connecting to the server when we run Satisfactory 7 but can't connect to Satisfactory 8.<latest version>. Constant timeouts even with a 120 tick in the engine.ini file. Any help would be appreciated.
Also, we're both retired IT so hit us with the techie stuff.
Thanks!

sharp umbra
#

@hardy sleet Another person having connectivity issues on Experimental, might be worth looking into it 😄 Maybe you will find some new information this time.

errant gull
#

Yea, and hubby was a network guru. So something is screwed up

sharp umbra
#

Well, the game has some issues currently in Experimental with networking, so it's probably not your fault 😅

#

But still, Baldur is doing a lot of research of this issue, so they will probably ask you later on for some help in testing ^^"

errant gull
#

Do you know if someone hosts version 8 and another joins might help? Basically not use on server

#

Well we do run Cisco routers and have an MTG server

sharp umbra
#

I can technically boot up my private Dedicated server you can try joining if you want for quick testing, but I have no idea how to gather useful information for developers, so that would be just "If it works with different host" test rather than useful testing session :P

desert walrus
errant gull
#

No save, can't even connect

#

Brand new

desert walrus
#

Ah, so you can't add it in the server manager?

errant gull
#

Can add it, see it, but can't connect

desert walrus
#

Okay... so have you created a save on the dedi?

errant gull
#

can't even connect....Can SSH into the server if need be but just timeouts from a client trying to connect to the server

desert walrus
#

You don't need to connect to do that, just have the server recognised in the server manager

errant gull
#

trying again...let's se

desert walrus
#

But you actually have to create a save on the server before you can join the world

errant gull
#

So it's not like version 7 where you select the server then create the game?

desert walrus
#

No, it is

#

I'm confused about where you're running into an issue

#

Are you unable to connect to the server in the server manager or are you unable to join the world

thorny shore
errant gull
#

OK, So I'm running into an issue within the Server MAnager screen when I click on the server....it times out

desert walrus
errant gull
#

I'm doing the force_install_dir /home/sfserver/satisfactoryserver/ experimental

#

is there a different switch? not experimantal?

errant gull
desert walrus
#

This is the command you should use
steamcmd +force_install_dir /home/sfserver/SatisfactoryDedicatedServer +login anonymous +app_update 1690800 -beta experimental validate +quit

errant gull
#

Ah, I see says the blind man...WOman 🙂 forgot the beta switch...will try again

#

Stills times out

#

Well, since it's only 2 of us and the peer2peer (session id) works, we;ll just do it. Thanks again

thorny shore
#

I'm sorry I couldn't help. I run a dedi on my Windows 10 Pro machine (install via Epic, made it a service via NSSM), and my buddy (out of the country from me) never has any issues connecting.

grim mica
verbal tulip
#

wait, so we should be connecting via ipv4 rather than ipv6?

thorny shore
#

my internal network is ipv4 so it just defaults to that, at least I'm assuming it does.

grim mica
#

Most home networks are still going to be running ipv4 in some sense.. the issue is not to do with IPv6 or v4 but how the engine handles connections

#

From what I recall on the issue, when you connect the engine responds with the ipv6 listener even if you connect with ipv4... The disablepacketrouting or multihome fixes that, somehow

#

If you are using IPv6 and its working, nothing wrong with continuing to do so

verbal tulip
#

it wasn't working 😭 hence the question

#

but multihome is working, thank you!

grim mica
#

Are you Julie's other half? lol

verbal tulip
#

is there a way to do that for self-hosting rather than DS? I went for DS cos I wasn't able to connect to self-hosting friends

#

I mean, not yet, but I'm in the market 😂

grim mica
#

Well I was referring to that specific case/issue, at least from what I gathered with what was provided

#

DS uses IP-based connectivity

#

All singleplayer/multiplayer uses the EOS system which is session IDs and all that, much different

#

I dont think there is a similiar option for non-DS

thorny shore
grim mica
#

I do agree.. I need to use multihome or nobody outside my LAN can connect to the server reliably

thorny shore
#

Maybe I should do some connection tests, but I have a buddy halfway across the world and he's never reported problems connecting.

#

plus I think I've already had some rando come on from time-to-time randomly deconstructing shit.

grim mica
#

It either is, or isnt an issue depending on your OS and NAT device it seems, for me its a problem probably because of my firewall

thorny shore
#

right

grim mica
#

Should be able to tell if that happens.. check the logs?

thorny shore
#

anyway, enough Satisfactory for now.

errant gull
#

Does version 8 use different ports than 7? From I what I seen, it doesn't.

grim mica
#

No, not different ports

#

The multihome doesnt affect the visibility of the server on WAN or LAN, just how Unreal engine networking deals with connections

errant gull
#

Yea, I was just worried that maybe Epic had changed a port for validation, etc.

grim mica
#

Negative, same ports

#

And I had an issue with DS when upgrading to 8 and had to use multihome to get mine working properly

errant gull
#

The thing is that the ping is 1ms. Can't get much better than that. It sees the server but we just can't get connected

#

Multihome on your LAN?

grim mica
#

Are you on the same broadcast domain as the server or being NATed?

#

Its been a hot minute since I had to deal with this but I believe it was only WAN-based clients that had issues connecting in my case

errant gull
#

Same. It's our same domain from our IT business we had before retiring.

grim mica
#

Humm.. dunno, id give the multihome a shot.. worse case it doesnt change anything anyway

errant gull
#

Not WAN either....server, hubby, and mine are all on our gigabyte CAt6

grim mica
#

Right, broadcast domain means on the same switch with no router/nat device between

errant gull
#

correct

#

I can have my hubby look a bit closer, he was a Cisco Engineer

#

Maybe capture some packets

thorny shore
grim mica
#

It would complain about a version mismatch on th e server manager screen if the version was wrong

errant gull
#

Correct

grim mica
#

Nor would it let them try to join and get the timeout

errant gull
#

So we updated it 🙂

thorny shore
#

but if they are connecting via U7....

grim mica
#

Yes, the client talks to the server before hand

#

If the server is EXP and client is EA it will tell you the builds are wrong

#

Same with the inverse

thorny shore
#

so by logical deduction, if they are connecting via EA with no problems, vs all the problems with exp

errant gull
#

Yea, and it's a pain to reload the clients....server easy.

grim mica
#

Yep

#

This is what the client does when its mismatched with the server

image.png
thorny shore
#

Yeah, I've seen that before.

#

so I'm familiar with that.

errant gull
#

A couple of weeks ago saw that but then just did an update/force

thorny shore
#

first I got to fix the VMs screen resolution. This tiny box is for the birds.

errant gull
#

The weird thing is that when we go to join the server (8), it says it Game Ongoing.

grim mica
#

Yes, the client talks to the server in the browser with different connection than the running game

#

Its the exact issue I had with WAN clients trying to join, they can see the server in the manager and even click join. Some would get in but then after moving around for 5-10 seconds it would kick, others never connected and just got a generic timeout

errant gull
thorny shore
#

VM server? I'm running the VM locally.

errant gull
errant gull
thorny shore
#

Right, I just use them for projects. I had one running a PXE server at one point.

#

anyway, VM talk is OT. grumbles at forgetting passwords

thorny shore
desert walrus
#

Shouldn't be an issue

thorny shore
#

shouldn't, going to be doing some connection tests.

desert walrus
#

Let's put it this way: if there's an issue, it's probably not because of the os

thorny shore
#

Oh I'm certain of it, but this is going to be fun. If this works like I hope it'll motivate me to boot up my office workstation, transfer the VM to it, and have two-three Satisfactory dedicated servers going.

#

same timeout issue. Will try the multihome switch.

#

Yeah, that's not right. It looks like it's mixing up IPv6 and IPv4.

image.png
#

I really need more monitors.

#

Yep, multihome fixed that.

#

Interesting, started with this

image.png
grim mica
#

Weird right?

thorny shore
#

Extremely

grim mica
#

That looks normal for a quick start

#

You are on LAN with the machine as well and it does that weird ipv4 <> ipv6 thing?

thorny shore
#

VM hosted on same machine

grim mica
#

Well, least I am not crazy

#

Granted - the fix is extremely simple its just not obvious what is going on

thorny shore
#

Yeah, that was super simple and following the Wiki and using Fight or Flight's command (modified to reflect my own directory structure, ofc). ezpz beautiful covergirl.

#

I mean, ezpz lemon squeezy

#

Both hosted on this machine, one in VM, the other not. I love computers.

image.png
#

Eventually I'll add the Linux one as a service and make sure the vm restarts w/e computer restarts.

verbal tulip
#

so what exactly is packet routing in this context?

errant gull
thorny shore
#

yes, but it won't after I get my office machine running.

errant gull
#

and how many NICs to you have on that PC?

thorny shore
#

1

errant gull
#

OK, that's why you had to use multihome

#

If you had 2 NICs with different IPs, you wouldn't need multihome

thorny shore
#

The server - for w/e reason - tried parsing the ipv4 address as ipv6

#

For the top server 'Joe's Gaming Paradise' I don't have -multihome set.

errant gull
#

Ah, Our Statisfactory server is on a dedicate server in our house (on our LAN)

#

So you think Satisfactory 8 is only using IPv6?

thorny shore
#

I don't know.

errant gull
#

Ah ok.

thorny shore
#

I use my external IPV4 address to connect to it

#

but if you set -multihome to w/e ipv4 address is of the computer the server is from, it seems to 'fix' that particular issue.

verbal tulip
#

have you tried the packet routing option instead of multihome?

thorny shore
#

Nope.

#

I could do that next time I feel like messing with it.

verbal tulip
#

I'm planning on spooling down the droplet and spooling it back up for the next session, so I expect the IP to change, so I don't want to use multihome, more work

errant gull
#

our server is connected to the same switch as out printers, PCs, laptops and on the same network and subnet

verbal tulip
#

I'll try packet routing next time around and see if that works

errant gull
#

Cool! And I'll play around with it tomorrow and see what I can do before having to get back pushing out a beta release for another older game

#

See ya'all tomorrow. Thanks everyone for the help!!!

verbal tulip
#

thank y'all for the help! not being able to play with my friends was driving me batty, I would have never found out multihome if not for youse

errant gull
thorny shore
#

well, you could if your router supports it

errant gull
#

We use DHCP with reservations

verbal tulip
#

this is a VPS, so I can't haha

thorny shore
#

yeah, I have my router set to assign IPs via MAC address and all the devices are set to use DHCP, so the router sets certain devices (desktops, laptops, printers) with the same IP on successive restarts/connections.

thorny shore
verbal tulip
#

people are still playing HOI2? damn

#

🫡

errant gull
#

A lot, They don't like HOI4

verbal tulip
#

if I keep renting the VPS then yes, I'd expect the IP to not change, but for my use case I totally expect it to change, that's reasonable

thorny shore
#

fair

#

Yeah, that makes sense then.

#

I think it's a Linux thing.

errant gull
#

Again, thanks to ya'll

verbal tulip
#

bit more hassle, but it's so cheap this way - I think the last play session was like, 7 cents

thorny shore
#

np Julie. I love helping when I can.

verbal tulip
#

plus 3 cents per mo for the snapshot

thorny shore
#

That's wicked cheap. AWS?

verbal tulip
verbal tulip
thorny shore
#

Ah

#

That's insane crazy cheap.

verbal tulip
#

I am cheaping out and running a 2GB instance, which runs fine for our world save so far

#

was 0.018/h for 3-4 hours

thorny shore
#

I was using OVH for Minecraft and it wasn't near that cheap.

#

Granted I was prepaying monthly

verbal tulip
#

when this world scales up I might need to bump up the droplet, but hell, even a 16 gig instance is $0.125/h

thorny shore
#

Heh

verbal tulip
#

84/month is a little steep, but if we keep doing what I'm doing and spooling it up only when we play, it's gonna be like, a dollar a week, which is peanuts

thorny shore
#

That's going to make progression really slow.

verbal tulip
#

it only matters if you turn off 'pause game when last player leaves'

#

plus it's functionally the same as we're doing right now, the previous solution was we launch the game when it's time for the play session, we don't idle that save and I think we're pretty alright with that

#

plus if we really need to I can idle the save on a local DS

#

I wanted to have a VPS for play session mostly for ping, so it's in the middle of all of us

#

but yeah, that's not really a massive consideration rn, might re-evaluate when things changes

thorny shore
#

I should mention I'm going to be redoing the VM to have a 192.168.1.* ip address and have the router forward the port to the proper machine when I move it to my office machine. Then I'm going to set it up as a service. I may put that one up for LFG here.

grim mica
#

The use of multihome just makes Satisfactory bind to that specific IP, but it ALSO disables packet routing

#

I believe the default (without startup switches) will bind to 0.0.0.0 which is akin to 'any/all' in terms of interfaces and would respond on whatever the client connected to

#

So dont think of multihome as telling it use this interface but just disables the unreal packet routing "feature" that seems to cause these issues in the first place

#

I just use -multhome=0.0.0.0 since I dont need it to bind to a specific IP (server machine is static and has just one active NIC) but rather just all available. I am fairly sure you need to put it before the -log and -unattended for it to work as well..

thorny shore
#

I didn't. I put it at the end of the command string.

toxic belfry
#

Is it just me or satisfactory produces unholy amount of log/journal files ? I am running just satisfactory on a Debian server.

image.png
thorny shore
#

use logrotate to help manage/delete old log data unless you're keeping them for some reason.

toxic belfry
#

yhea gonna delete logs, i was just wondering is is normal for satisfactory to produce this amount of logs?

#

you know just in case i have funky stuff going on. don't wanna crash game and end up with corrupted safe file

vague trail
#

™️

deep turret
runic turtle
#

Hi, we have a private dedicated server with 16gb of ram, the server uses only 2gb and often the avarage tick rate drop to 15/crash the server. Does anyone know why?

tall bough
#

how can you make a other server than dedicated online multiplayer server

desert walrus
#

Not sure what you mean... your options are dedicated server or non-dedicated multiplayer

tall bough
#

and how do i get a non dedicated multiplayer server

thorny shore
#

you don't

#

you open up a normal game and invite your friends to play

#

!wikisearch multiplayer

rancid etherBOT
#

https://satisfactory.wiki.gg/index.php?search=multiplayer

Satisfactory Wiki
Index.php

Satisfactory can be played in multiplayer, with up to four players officially supported, but with no hard player-limit see below.[1] The game is co-op only and there will be no PvP elements, although pioneers can damage each other.[2] This means that HUB, MAM and Space Elevator progress as well as AWESOME Shop unlocks are shared among all players.

rich palm
#

We have quite a powerful CPU and also with 3 players the tickrate goes down easily

runic turtle
rich palm
grim mica
#

The main loop is still on a single thread, so yes you will be CPU limited and generally not memory limited given you have enough

grim mica
rich palm
grim mica
desert walrus
rich palm
deep turret
grim mica
#

Right, but the statement that DS is somehow 'unoptimized' is just incorrect. They are constantly improving and optimizing as they go and dont release unoptimized builds to us. Stating that DS is unoptimized leaves open room for improvement in terms of tickrate with multiple clients that is very unlikely to occur.

desert walrus
deep turret
#

you have to sync every calculations at some point to set the frame to network and make the client able to know its state depending on that
otherwise everything would stay very inconsistent
and that's the main loop

desert walrus
#

I'm aware of that

grim mica
#

Not stating that there isnt more than a single thread, obviously there is but if there is a thread that doesnt proceed until a single loop is completed, then well.. it is absolutely single threaded with regard to performance. Saying that a 64-core 2Ghz CPU (ignoring all other factors aside from core and frequency) vs an 8-core at 5Ghz - which would be better at running the game?

desert walrus
#

That still doesn't mean it's single threaded. The limitation can be on the execution of any one thread on any given frame

deep turret
#

if your main loop is loosely tied to the workload in term of threads for the syncing, then you will add syncing overhead on top of that and decrease resource usage

#

you need it to be managed as worker queue with main loop handling tight dependencies

grim mica
#

I think we are confusing single thread and main loop - i'm refering to tick rate on DS as that was the main topic of discussion here and tickrate is directly correlated to the 'main loop', regardless of how many threads may be active at any given time.

rich palm
grim mica
#

Right, and im saying that is NOT the case

#

I would not expect major performance improvements moving forward

desert walrus
deep turret
#

it's a common optimization mistake to misconceive the dependency chain loading/queuing
and ending up with more struggles yet more idling

rich palm
rich palm
rich palm
desert walrus
#

No, that was when dedi performance was worse

#

U6 was the last time I ran a dedi on a regular basis

deep turret
#

it runs fine on ryzen 5900x

#

So I don't really get what are the 'latest i9' struggles supposed to be

rich palm
desert walrus
rich palm
#

You see it clearly with trucks and when you drive and hit each other

desert walrus
#

That's not a problem of server performance, much more due to the way it handles client syncing, which needs a lot of work

deep turret
rich palm
deep turret
#

oh

rich palm
desert walrus
#

Playercount doesn't actually have that significant of an impact on dedi performance

deep turret
#

I was not active this week because I had lot of work to do
I was wondering what I could do tomorrow to cool down my brain
maybe I'll run satisfactory 👍

rich palm
deep turret
#

more player usually means more spread cpu usage and bigger network and ram usage

#

are you using a vps or windows host ?

rich palm
#

No I’m using a dedicated machine with Linux (I’m a cloud engineer, so I know nothing is wrong with the hardware or setup)

#

However what I noticed on the current branch, sometimes the console gets spammed with error dumps. Esp when fighting with creatures. (The errors mention something like animations). It seems to affect it heavily too

marsh mango
#

anyone else running into issues with running dedi server after this most recent update? Mine gets stuck at [2023.10.29-02.23.56:526][ 18]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 2.513000, Update Interval: 328.534210

I even went in and renamed the savedGames folder in hopes that it would launch. no dice.

grim mica
marsh mango
marsh mango
# grim mica I dont think that log is relevant to loading, what is happening with the server?

Just checked it again and here is what it is doing.

[2023.10.29-02.36.14:554][ 15]LogEOSSDK: LogEOSAnalytics: Start Session (User: ...)
[2023.10.29-02.36.14:687][ 19]LogEOSSDK: LogEOS: SDK Config Product Update Request Successful, Time: 7.291040
[2023.10.29-02.36.14:688][ 19]LogEOSSDK: LogEOS: SDK Config Data - Watermark: -1544435436
[2023.10.29-02.36.14:688][ 19]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 7.291040, Update Interval: 335.660889
[2023.10.29-02.41.50:347][947]LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 342.988464
[2023.10.29-02.41.50:619][955]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
[2023.10.29-02.41.50:619][955]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 343.225922, Update Interval: 335.203094
[2023.10.29-02.47.29:093][981]LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 681.732544
[2023.10.29-02.47.29:390][990]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
[2023.10.29-02.47.29:390][990]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 681.996216, Update Interval: 321.180450
[2023.10.29-02.52.55:352][665]LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 1007.989929
[2023.10.29-02.52.55:487][669]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
[2023.10.29-02.52.55:487][669]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 1008.090576, Update Interval: 338.684052

#

repeats about every 6 minutes. (the rest of it is to long to post).

grim mica
#

But what is the actual problem? Is it not loading the save or no response in server manager?

marsh mango
#

The server never comes online for people to join.

#

this behavior is the same on a new instance or loading the saved game.

grim mica
#

post the rest of the logs?

marsh mango
#

I stand corrected. when I switched it back to the saved games folder, I get a few more lines.

[2023.10.29-03.27.18:689][ 19]LogEOSSDK: LogEOS: SDK Config Data - Watermark: -1544435436
[2023.10.29-03.27.18:689][ 19]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 5.495795, Update Interval: 350.613739
[2023.10.29-03.27.22:700][139]LogNavigationDataBuild: Display: FRecastNavMeshGenerator::ProcessTileTasksAndGetUpdatedTiles build time: 4.66s
[2023.10.29-03.27.23:932][176]LogNavigationDataBuild: Display: FRecastNavMeshGenerator::ProcessTileTasksAndGetUpdatedTiles build time: 5.89s
[2023.10.29-03.27.25:458][222]LogNavigationDataBuild: Display: FRecastNavMeshGenerator::ProcessTileTasksAndGetUpdatedTiles build time: 7.42s
[2023.10.29-03.27.28:112][302]LogNavigationDataBuild: Display: FRecastNavMeshGenerator::ProcessTileTasksAndGetUpdatedTiles build time: 10.07s
[2023.10.29-03.27.29:971][358]LogNavigationDataBuild: Display: FRecastNavMeshGenerator::ProcessTileTasksAndGetUpdatedTiles build time: 11.93s
[2023.10.29-03.27.32:420][432]LogNavigationDataBuild: Display: FRecastNavMeshGenerator::ProcessTileTasksAndGetUpdatedTiles build time: 14.38s

then it stops.

#

Sure, let me get the log file and I'll upload it here.

#

Here is the log file from an attempt to launch the server using a previous saved game.

📎 FactoryGame.log
grim mica
#

Dont see any obvious issues but also dont see any clients attempting to connect

#

Where is the server, on your LAN or hosted?

marsh mango
#

its on my LAN

#

prior to this recent update, my daughters friends were able to connect to it just fine.

grim mica
#

Recent meaning an u8 Patch?

marsh mango
#

yes. Looks like the most recent patch was released on 10/25.

grim mica
#

Or switching from U7 to U8? Ah okay

marsh mango
grim mica
#

When you add the IP in server manager is just says not responding?

marsh mango
#

The server is offline

#

or the server appears offline.

#

Can't recall the verbiage.

grim mica
#

Yep all good, hmm

marsh mango
#

but looking at the server lines, there usully is an acknowledgement that the server is up and ready. I never get that message with this update. I'm going to let it run over night and see what comes in the morning. maybe it just needs to work through something...

grim mica
#

I mean its loading the map
[2023.10.29-03.27.15:197][ 0]LogLoad: LoadMap: /Game/FactoryGame/Map/GameLevel01/Persistent_Level??sessionName=Tuna-Tito?startloc=Rocky Desert?Visibility=SV_Private?listen?loadgame=Tuna-Tito_autosave_0

#

I see no connection entries though

#

Lemme compare it against one of my logs, just in the middle of a sons of the forest cave

marsh mango
#

sons of the forest is a great game.

#

got it when it first came out, it was missing quite a bit of content. went back to it a few months later and it was a whole new gam.e

#

Looks like the server is up and accepting connections. I guess this is a non-issue. I'm use to seeing the server is ready message.

grim mica
#

Allrighty, good to know lol

molten widget
#

how do i add blueprints to a dedicated server?
the guy hosting the server added the blueprints to the save location where they are meant to be placed but for some reason they don't appear in-game
Server is on EA U7, Blueprints are also made in EA-U7

deep turret
#

looks like a bug maybe, current tickrate is between 11 and 15

#

and the last manual save didnt save anything

#

it basically just duplicated the old save

deep turret
#

wait looks even worse than that, it looks like it's partially saved, some things are from the save point, others are not, looks like it saved everything bug my bodys and equipments

#

can confirm it at least saved partially

#

weird that it didnt save my equipments

#

didnt succeed at reproducing it though

#

wait I found my piece of equipment

image.png
#

so it just means I had both my body/ids at the old location (the one on the screenshot was at the same starting point than the one I have now before)

#

so it saved fine

#

I'm feeling better now

vivid pine
#

cant add blueprints

#

my friend did that first, it workied for him

#

(hes on EG im on Steam

thorny shore
#

Have y'all unlocked the BP designer? Was the server restarted afterward?

deep turret
reef lily
#

does aanyone have automatic crash recovery on their server??

#

i wrote a batch file that automatically closes, backs up the save, check for updates, and relaunch the server. i also wrote another one that checks for the server running every 15 seconds and relaunches it if if crashes. would anyone be interested in these files?

deep turret
#

" i also wrote another one that checks for the server running every 15 seconds and relaunches it if if crashes" you can do that a lot more easier than this

#

you can put it in the launch script

#

and basically just loop over the launching bat

#

when the server exits it just relaunch itself

reef lily
#

didnt know that there was a syntax for that. i dont see it on the wiki. whats the syntax?

deep turret
#

I think there could be a way to wait on the process to end

reef lily
#

lol thats a more elegant solution than i took. im not great at batch scripting 😅

#

i believe the script there is the only way to 'wait on the process to end'.

deep turret
#

I was more thinking of something like that

while($true)
{
launch.server -args
server process |wait end
}```
reef lily
#

your starting to make me feel dumb. 😂 i know very basic batch scripting. only know what iv had to learn to get accomplish a goal.

deep turret
amber storm
loud saddle
#

running server on windows 🤢

deep turret
amber storm
loud saddle
#

with tons of issues 🤷‍♂️

amber storm
#

Really? I've had exactly none.

toxic belfry
#

On Linux how do I turn off logs ? on wiki it says -log is implicit on Linux but does not mention anything about how to disable it. This morning I woke up to 9GB of log files after server was running for about about 24h.

vivid pine
#

How do i download save games

#

from a server

deep turret
rough arrow
#

how up to date is the wiki on dedicated servers on fandom? there is a reference to update 5 and also references to ini files that i don't have (or aren't in the places they specify) on linux dedicated server

#

i am trying to sort out a server build that i built the vm months ago, the server has been offline but we have stood it up again and updated things but now can't get connected.

#

so i think that means i am only on the networking step and not ready to actually configure game settings via ini, however one of the steps specified is to update the timeout to 60s via an ini that i don't seem to have.... so kind of stuck

abstract bay
#

I have the itch to play experimental with multiplayer. Is it bad right now? Does having a dedicated help?

thorny shore
#

I've not tried multiplayer, I just went straight to Dedi because I wanted the world running 24/7

rough arrow
#

experimental is much more fun than stock for multiplayer we have been on it since update 4

dim nebula
#

[2023.10.29-19.01.20:640][ 0]LogStreaming: Error: Couldn't find file for package /Game/FactoryGame/Schematics/Story/Schematic_ObjectScanner_WAT1 requested by async loading code. NameToLoad: /Game/FactoryGame/Schematics/Story/Schematic_ObjectScanner_WAT1
[2023.10.29-19.01.20:642][ 0]LogStreaming: Error: Found 1 dependent packages...
[2023.10.29-19.01.20:642][ 0]LogStreaming: Error: /Game/FactoryGame/Prototype/WAT/Desc_WAT1
[2023.10.29-19.01.20:643][ 0]LogStreaming: Error: Couldn't find file for package /Game/FactoryGame/Schematics/Story/Schematic_ObjectScanner_WAT2 requested by async loading code. NameToLoad: /Game/FactoryGame/Schematics/Story/Schematic_ObjectScanner_WAT2
[2023.10.29-19.01.20:643][ 0]LogStreaming: Error: Found 1 dependent packages...
[2023.10.29-19.01.20:643][ 0]LogStreaming: Error: /Game/FactoryGame/Prototype/WAT/Desc_WAT2 can anyone help me with my server it wont launch

grim mica
vagrant turtle
#

I still can't get my DS running since the last patch.

0x00007fadb27a7677 libUnrealServer-UnixCommonStartup-Linux-Shipping.so!CommonUnixMain(int, char**, int (*)(char16_t const*), void (*)()) [C:/BuildAgent/work/78a794e88763017d/UE4/Engine/Source/Runtime/Unix/UnixCommonStartup/Private/UnixCommonStartup.cpp:269]
0x00007fadb259fd90 libc.so.6!UnknownFunction(0x29d8f)
0x00007fadb259fe40 libc.so.6!__libc_start_main(+0x7f)
0x0000000000217dc9 UnrealServer-Linux-Shipping!_start()

[2023.10.29-19.25.00:262][  0]LogExit: Executing StaticShutdownAfterError
Engine crash handling finished; re-raising signal 6 for the default handler. Good bye.
Aborted (core dumped)

It all looks like it's about to fully load properly, then boom, after the start(), it shuts down after error with no error listed

#

Ubuntu 22.04, everything updated to latest

rough arrow
#

can the

-ServerQueryPort
-BeaconPort
-Port```
options be used with the server running via systemctl service?
https://satisfactory.fandom.com/wiki/Dedicated_servers
dim nebula
vagrant turtle
#

I just scrolled back up further in the logs. It shows that is opening the ports and then gives the Signal 6:

[2023.10.29-20.17.00:206][  0]PacketHandlerLog: Loaded PacketHandler component: DTLSHandlerComponent ()
[2023.10.29-20.17.00:206][  0]PacketHandlerLog: Loaded PacketHandler component: Engine.EngineHandlerComponentFactory (StatelessConnectHandlerComponent)
[2023.10.29-20.17.00:207][  0]LogNet: GameNetDriver NetDriverEOSBase_2147482210 IpNetDriver listening on port 7777
Signal 6 caught.
Malloc Size=262146 LargeMemoryPoolOffset=262162  
CommonUnixCrashHandler: Signal=6
[2023.10.29-20.17.00:251][  0]LogCore: === Critical error: ===
Unhandled Exception: SIGABRT: abort() called

[2023.10.29-20.17.00:251][  0]LogCore: Fatal error!

0x00007f0a4e9619fc libc.so.6!pthread_kill(+0x12c)
0x00007f0a4e90d476 libc.so.6!raise(+0x15)
0x00007f0a4e8f37f3 libc.so.6!abort(+0xd2)
0x00007f0a4d08cb93 libUnrealServer-Chaos-Linux-Shipping.so!TransformedAABB(+0x92)
rough arrow
rough arrow
dim nebula
vagrant turtle
rough arrow
#

ok well in that case thanks for helping me rule that out. i was concerned about 0.0.0.0

grim mica
# dim nebula

This is a dedicated server log? Seems like a client log

#

You want the FactoryGame.log in the server directory, depending on how you are hosting the dedicated you may not have access to that

#

Are you on Windows? Linux? Paying for the host? Hosting it on the same as the client? Need some more detail to help you find it..

dim nebula
#

I am on windows 11 and hosting on the same client

grim mica
#

Okay based on that log you have 16GB of memory, trying to run a dedicated and the client probably wont work well, but without the server log I cannot tell you why it isnt running

#

The server log will be in the steamapps\common\SatisfactoryDedicatedServer\FactoryGame\Saved\Logs folder

torn sparrow
#

Hi how do i set enemies to passive on dedicated server

desert walrus
#

Using the normal client side setting

torn sparrow
#

im using a paid one from shokbyte and i cant find the option there

dreamy nexus
#

Is there some new way to change the autosave interval on dedicated servers? The wiki says to type FG.AutosaveInterval xx in the console but I'm getting a command not recognized.

deep turret
royal quail
thorny shore
#

yes

#

or it'll start off the animation at the wrong end (or wrong tube system).

rough arrow
#

i don't use tubes because i am ALWAYS next to or underneath the tube

thorny shore
#

weird

rough arrow
#

and it can be game-breakingly bad and crash the client

#

never had it crash the server, but it does crash the client after (my guess) a mile of tube

thorny shore
#

I've not had it crash my client, but I have had to relog.

rough arrow
#

i consider those kind of the same, a 'soft crash' you need to log out to clear

#

but yes i have had both

thorny shore
#

fair

rough arrow
#

i would like it if anyone has a running DS on debian/ubuntu/similar could let me pick their brain whenever is convenient

#

my vm's are debian 12 with 4 cores & 32GB. I can increase both as needed, both on the same host, so i will have to take port forwarding into consideration

royal quail
#

SO Random update, found by complete accident... my friend spammed F while his hand was off mouse and it stopped lol

thorny shore
rough arrow
#

my server 'runs' now that i have the port/sqp/beaconport and multihome added to match my fw rules

#

however i can't create a game, it goes from letting me choose which region to start, to 'server is preparing a new game' to 'the server appears offline'

#

i am digging through ini files now to see if the game has populated an ini that i can edit with the timeout thats suggested in a few different reddit posts

rough arrow
# 
[2023.10.30-03.36.47:397][327]LogNet: IpConnection_2147482275 setting maximum channels to: 32767
[2023.10.30-03.36.47:398][327]PacketHandlerLog: Loaded PacketHandler component: DTLSHandlerComponent ()
[2023.10.30-03.36.47:398][327]PacketHandlerLog: Loaded PacketHandler component: Engine.EngineHandlerComponentFactory (StatelessConnectHandlerComponent)
[2023.10.30-03.36.47:398][327]LogNet: NotifyAcceptedConnection: Name: OnlineBeaconHost_2147482297, TimeStamp: 10/29/23 23:36:47, [UNetConnection] RemoteAddr: x.x.x.x:63952, Name: IpConnection_2147482275, Driver: IpNetDriver_2147482295 IpNetDriver_2147482295, IsServer: YES, PC: NULL, Owner: NULL, UniqueId: INVALID
[2023.10.30-03.36.47:398][327]LogNet: AddClientConnection: Added client connection: [UNetConnection] RemoteAddr: x.x.x.x:63952, Name: IpConnection_2147482275, Driver: IpNetDriver_2147482295 IpNetDriver_2147482295, IsServer: YES, PC: NULL, Owner: NULL, UniqueId: INVALID
[2023.10.30-03.36.47:833][340]LogDTLSHandler: EnableEncryption: Server
[2023.10.30-03.36.47:834][340]LogDTLSHandler: Warning: Empty certificate identifier
[2023.10.30-03.36.48:160][343]LogDTLSHandler: DTLSCertVerifyCallback: Verified
[2023.10.30-03.36.48:160][344]LogDTLSHandler: UpdateHandshake:  Handshaking completed
[2023.10.30-03.36.48:462][352]LogNet: NotifyAcceptingChannel Control 0 server OnlineBeaconHost /Game/FactoryGame/Map/DedicatedserverEntry.DedicatedserverEntry:PersistentLevel.OnlineBeaconHost_2147482297: Accepted
[2023.10.30-03.36.48:462][352]LogNet: Remote platform little endian=1
[2023.10.30-03.36.48:462][352]LogNet: This platform little endian=1
[2023.10.30-03.36.48:462][352]LogBeacon: OnlineBeaconHost_2147482297[IpConnection_2147482275]: Beacon Hello
[2023.10.30-03.36.48:529][354]LogBeacon: OnlineBeaconHost_2147482297[IpConnection_2147482275]: Client netspeed is 120000
[2023.10.30-03.36.48:529][354]LogBeacon: OnlineBeaconHost_2147482297[IpConnection_2147482275]: Beacon Join FGServerBeaconClient Steam:1 (76561197983978788) (unauthenticated)
[2023.10.30-03.36.48:529][354]LogReplicationGraph: Warning: No Replication Lists were returned for connection
[2023.10.30-03.36.48:562][356]LogReplicationGraph: Warning: No Replication Lists were returned for connection
[2023.10.30-03.36.48:596][356]LogBeacon: OnlineBeaconHost_2147482297[IpConnection_2147482275]: Handshake complete.

putting this here in hopes maybe someone will be able to help me understand why the server will allow me to set admin pw, game pw, set a starting point, and then once i try to create the session it tells me its offline

grim mica
#

Can you get the full factorygame.log and share?

rough arrow
#

sure give me a moment to rejoin the vpn

grim mica
#

A client timeout shouldnt be the issue, but the UI in the client isnt the best atm and exhibits finicky behavior.. you cant refresh the server state for example and it doesnt re-check unless the clients restarts OR you add the server again

#

We've been asking for a refresh button for awhile, just DS not getting much attention currently. As they get closer to 1.0 it hopefully will get better.

rough arrow
#

do you mind if i dm it, too large to obfuscate every instance of ip address like above

desert walrus
#

You can just find/replace if you're worried about it

grim mica
#

Yeah you can DM if you want, or like Fight said just CTRL+H and replace

rough arrow
#

i forgot textedit gives me that option, i have too many big projects open in other text editor to bring it open tonight

grim mica
#

Notepad++ FTW 😄

#

One sec, its formatted a bit.. interestingly

rough arrow
#

i could fix that if needed i literally just pasted from less probably line breaks everywhere

grim mica
#

So it just goes offline once you select the start location hmm.. not even seeing it try to get that far

#

The startup looks normal and then it holds at waitingtostart since it finds no existing save files

rough arrow
#

this morning i had no ping and had to fight with fw rules, we had a windows server in that rack so i have made some decent progress, but with nearly identical settings i have only 1/2 vms able to get that far

grim mica
#

Assuming the process stays running after it disappears from server manager?

#

What if you just try to add the server again from the menu after selecting the start loc?

rough arrow
# 
satisfactory.service - Satisfactory dedicated server
     Loaded: loaded (/etc/systemd/system/satisfactory.service; enabled; preset: enabled)
     Active: active (running) since Sun 2023-10-29 23:36:32 EDT; 39min ago
    Process: 817 ExecStartPre=/usr/games/steamcmd +force_install_dir /home/steam/satisfactory +login anonymous +app_update 1690800 -beta experimental validate +quit (code=exited, status=0/SUCCESS)
   Main PID: 862 (FactoryServer.s)
      Tasks: 28 (limit: 38471)
     Memory: 304.1M
        CPU: 2min 23.021s
     CGroup: /system.slice/satisfactory.service
             ├─862 /bin/sh /home/steam/satisfactory/FactoryServer.sh -multihome=10.6.66.101 -ServerQueryPort=15781 -BeaconPort=15001 -Port=7781
             └─869 /home/steam/satisfactory/Engine/Binaries/Linux/UnrealServer-Linux-Shipping FactoryGame -multihome=10.6.66.101 -ServerQueryPort=15781 -BeaconPort=15001 -Port=7781

Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-03.53.44:867][687]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-03.53.44:867][687]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 1031.838989, Update Interval: 322.159729
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-03.59.16:602][588]LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 1363.547363
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-03.59.16:735][592]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-03.59.16:735][592]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 1363.647705, Update Interval: 334.572815
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-04.05.10:113][138]LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 1717.078613
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-04.05.10:247][142]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-04.05.10:247][142]LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 1717.179321, Update Interval: 347.416931
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-04.11.28:688][436]LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 2095.589844
Oct 30 00:11:28 fg1 FactoryServer.sh[869]: [2023.10.30-04.11.28:822][440]LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change

for context specifically with the ports

rough arrow
#

i have htop going i will try it and let you know about the pid

#

same results in client, checking server

#

the pid changes

grim mica
#

Odd, that log doesnt indicate any crash hm

rough arrow
#

so the service is restarting but i couldn't figure out where/how in the log

grim mica
#

PID change would indicate that it is closing and restarting, no?

#

I would simplify and just run it in an SSH temporarily or something, forget the service part

#

See what you actually get in the console

rough arrow
#

give me a moment and i will try that

# 
0x0000000000219c3a UnrealServer-Linux-Shipping!GuardedMain(char16_t const*) [C:/BuildAgent/work/78a794e88763017d/UE4/Engine/Source/./Runtime/Launch/Private/Launch.cpp:190]
0x00007f017645b677 libUnrealServer-UnixCommonStartup-Linux-Shipping.so!CommonUnixMain(int, char**, int (*)(char16_t const*), void (*)()) [C:/BuildAgent/work/78a794e88763017d/UE4/Engine/Source/Runtime/Unix/UnixCommonStartup/Private/UnixCommonStartup.cpp:269]
0x00007f01712461ca libc.so.6!UnknownFunction(0x271c9)
0x00007f0171246285 libc.so.6!__libc_start_main(+0x84)
0x0000000000217dc9 UnrealServer-Linux-Shipping!_start()

[2023.10.30-04.25.43:612][994]LogExit: Executing StaticShutdownAfterError
Engine crash handling finished; re-raising signal 6 for the default handler. Good bye.
Aborted

how far up would you like me to go

grim mica
#

oh goodie lol

#

at least the start of the crash i suppose

#

can see what portion its crashing on

#

I think you mentioned VM and I see a virtualized CPU in the other log, what hypervisor is it?

rough arrow
#

proxmox

grim mica
#

I almost knew you were going to say that lol

rough arrow
# 
[2023.10.30-04.25.43:551][994]LogStreaming: Error: Couldn't find file for package /Game/CB/FactoryGame/Map/GameLevel01/Persistent_Level/_Generated_/StreamingObject_251CVL0ABL1HOH6ZF3KRICZIW requested by async loading code. NameToLoad: /Game/CB/FactoryGame/Map/GameLevel01/Persistent_Level/_Generated_/StreamingObject_251CVL0ABL1HOH6ZF3KRICZIW
[2023.10.30-04.25.43:551][994]LogStreaming: Error: This will hitch streaming because it ends up searching the disk instead of finding the file in the pak file.
[2023.10.30-04.25.43:551][994]LogStreaming: Error: Found 0 dependent packages...
#

i think this might be relevant

grim mica
#

There is some CPU setting for proxmox, though it was so far affecting starting the server - then again its likely people were updating and resuming an existing save and not starting new world

#

"must set CPU in host mode" on proxmox

#

That error looks actually familiar, im not sure that would cause the crash though

rough arrow
#

there is another AMD option that is cut off both these images, that isn't relevant since these are Xeons

image.png image.png
grim mica
#

Might have to stop the VM but I am guessing "type"?

rough arrow
#

i didn't even know that was an option lets see how bad we can break it now

image.png
grim mica
#

There must be some extensions that setting it to host vs kvm64/emulation exposes that EXP needs

#

Dont know what that would be.. but you also dont see those emu CPUs on VMWare or HyperV.. wonder why 😐

rough arrow
#

alright its back up one moment

#

Ficsit Tip no 14
Life needs things to live

#

never been so glad to see an ADA message

#

thank you so much for the help i loaded into the world

#

then i got a timeout message that i was able to dismiss

grim mica
#

Yeah thats probably just server manager being.. itself

rough arrow
#

so far so good i'll let it idle for a moment and make sure but i'm going to edit server 2 the same now

grim mica
#

i would run around and start building random shit, seems to be the best test

#

ive had people be connected to my DS and once they move or try to build, it DCs

rough arrow
#

yep so still wacky with server 2 but i can deal with that another day

#

i wanted to disco/reco in addition to testing that way

#

nope booted back to main menu

#

i appreciate all your help i can't do any more tonight but i will dig through the logs tomorrow and see if i can find anything more relevant

grim mica
#

That the 2nd or 1st server?

rough arrow
#

the first one that i did load into

grim mica
#

Interesting, allright

#

Well progress at least - and np you can @ me and if im around will take a look with you

rough arrow
#

thanks

harsh eagle
#

Any reason why I can't connect to a Nitrado hosted server for Satisfactory Exp?

#

"The Server Appears To Be Offline"

rich cliff
#

are there any issues with experimental dedicated server radar tower? it's not showing all nodes any more

rich cliff
rough arrow
#

does the game really need 100 ports range?

rich cliff
#

don't know. but it wasn't working for me before i did that

rough arrow
#

what was the next smallest range you tried before moving to 100 wide?

glad nest
#

Should be sufficient with 3 ports open, maybe even 2...

#

7777, 15000 and 15777

rich cliff
rich cliff
glad nest
#

They should be

#

Game port may vary if you have anything running on the port already

thorny shore
glad nest
#

Then it will be +1

rich cliff
#

i'll test today, I think dedicated is just using another then p2p

#

probably 7778 or smth

glad nest
#

It will be +1 if you have something occupying game port already

#

Its on the guide

thorny shore
#

!wikisearch Dedicated+Server

rancid etherBOT
glad nest
thorny shore
#

yes

glad nest
#

Neat

thorny shore
#

well, until the client decides to futz out.

glad nest
#

Right

#

Would still recommend to forward the 3 ports as long as the guide still says so

toxic belfry
#

How do I disable logs on Linux? Wiki says -log is implicit on Linux but does not mention on how to turn it off. My log files are getting out of hand.

glad nest
#

Don't think you can

#

how many logs / large ones we talking?

#

havent had any issues with log size since the vehicle logs spam

thorny shore
#

Otherwise you could always use a cron job to find and remove logs of a certain age.

toxic belfry
#

Well am taking about the one on the satisfactory folder

thorny shore
#

yes

#

I know.

#

man logrotate

#

anyway

toxic belfry
#

Oh you can you logorotate to manage that as well damm okay

#

Did dint know

thorny shore
#

also, you could symlink the game's log folder into the system's log folder (assuming you have the perms to do so)

#

that should work (note, I haven't done this, just an idea)

toxic belfry
#

Yhes I do, I am ruining it on my old gaming laptop.

thorny shore
#

anyway, back to game

toxic belfry
#

Hmm that's an idea. True.

deep turret
cunning field
#

Hi guys, when the game saves I have huge lag and the game save could take 5 seconds, wich part of my server hardware should upgrade to reduce the lag or make the save process faster? Are there any software fixes for this?

desert walrus
#

Cpu is the biggest limiting factor in save time(ram speed to a lesser degree)

sly shard
#

I update my server just now and it doesn't start

i start with this cmd in windows

start FactoryServer.exe -NOSTEAM -log -unattended -multihome=192.168.1.102

¿Someone know anything?

deep turret
#

check that the ip is still the one of the server machine

sly shard
#

The ip has not change, y try remove the param and doesn't work

But if open from steam launcher it's works correctly

cunning field
loud minnow
desert walrus
cunning field
#

so basically a faster single core speed cpu with less cores/thread should be good enough?

thorny shore
#

I'm on a Ryzen 7 3800x. Pretty happy with performance in terms of save-time (though my factory is small). Am running a dedi server on same machine, fwiw.

cunning field
#

I wonder that should I run on my main PC tho I dont wanna loose graphic performance

thorny shore
#

Dedi server shouldn't touch gfx, only client. But I could be wrong.

cunning field
#

i had problems when I ran with my prev pc

#

but as I realized my prev GPU was faulty

tall citrus
#

Oh yeah you are definitely good with that setup. I would call the manufacturer of the gpu

#

And ask for a replacement

#

If it's faulty

cunning field
#

I should just send back

#

always forget

#

..>

#

actually I did the RMA e-mailing with gygabite

#

thanks for reminding me

tall citrus
#

No prob

sharp umbra
#

Ofc you need good PC for that, but I have nearly identical PC as you, so I think you should see similar results

cunning field
#

hmm okay, thank I'll might try that

sharp umbra
#

Just remember Dedi is way less stable MP experience than client hosted game >.> I've decided to loose performance for stability for example

cunning field
#

that would have been my next question

#

the save structure is the same?

#

dedicated and client hosted

#

I'm running dedicated so the machines can work while I am AFK

#

or playing other game

sharp umbra
#

It's the same between Dedi and Client-Hosted, it's not the same between U7 and U8.
You can pretty much load a Singleplayer save on dedicated server any time, or load dedi server on Client if you are hosting it on the same machine

cunning field
#

that is nice to hear, I'll try to load in my server single player then

jagged harness
#

i like how you feel like a god then unlock the next tier and fully absorb what you need to do....

thorny shore
#

I've been ignoring the next tier and just building stuff.

rigid sun
#

excuse me whatttt

image.png
#

412GB log file

toxic belfry
#

Yhea I been having the same log file issue. My solution for now is just delete it when ever I remember as I haven't still figured out of its possible to disable them on Linux

thorny shore
#

is logrotate not a possibility?

grim mica
#

Doesnt happen on Windows at least..

glad nest
#

never seen that big of a save file either

#

how long has it been running for

deep turret
glad nest
#

I meant log file

west sun
#

how much does it cost to build an own dedicated satisfactory game server?

#

(does that belong in tech-talk?)

deep turret
#

if you are talking about setup it belongs here

west sun
#

I want to build my own satisfactory server, bc i dont want to pay for it monthly

deep turret
#

base cpu need is ryzen 3600, you can find some pretty cheap
though would recommend starting with 7600 instead
it is lot more expensive rn
all over 3600 will be able to run it even if it does not run it very good
check your spare hardware first

west sun
#

I dont have spare Hardware

#

sorry

#

ill wait until i upgrade my pc

#

but it will take a whilr

#

a long while

glad nest
#

I just bought a used office computer with a i7-6700K, works like a charm, got it for 90$ so doesn't really need to be expensive

edgy crag
#

My boys and I play together, but, since we must play experimental to use dedicated servers, we are losing alot of enjoyment because of incessant downloads and bugs. We've lost at least a hundred power shards and no telling how many harddrives to updates wiping our characters.
I would hope that, IN THE FUTURE, dedicated servers are divorced from experimental mode.

desert walrus
#

Or a 125/600k

desert walrus
edgy crag
desert walrus
edgy crag
desert walrus
edgy crag
#

I intially tried to set it up in stable/EA, but after support got involvrd, they said it had to be experimental.
We had originally had a server with them prior to U8, and stable was fine

#

after u8, it wouldn't work anymore

deep turret
#

I can understand the greed to go to U8 for how much RAM is saved

desert walrus
deep turret
#

but at least make sure the networking is fine first

#

smh

kind lion
#

hello can someone help my friend dedicated server keeps crashing on start

deep turret
kind lion
deep turret
#

how much is available when you start the server

kind lion
deep turret
#

you mean your machine have more than 64GB RAM ?

kind lion
deep turret
#

so not all of it is available to the dedicated satisfactory server

#

you may need 12GB is available depending on the save file

#

maybe you are right and plenty is available, but you saying that all of it is available does not make me assured about the informations you give me

kind lion
deep turret
#

and no other services are running I guess

#

then ok, you have far enough

kind lion
#

yes

deep turret
#

it does not crash on starting the server ?

#

only some times after you launched a game ?

kind lion
#

it crashes on start

#

the weird thing that it was working before

deep turret
#

make sure you have only one instance of the server running (that it is not already launched in the background)

kind lion
#

there's none

deep turret
#

some other people had trouble running on windows server too iirc
maybe it is an issue with some windows dependencies

#

the game is prepared to run on client versions

#

I dont know if people successfully ran it on windows server

#

I would let people knowing that help you

wicked pine
#

Do you have a recommendation of a cloud server

fading quarry
#

Not really allowed to advertise here, but all I can say is stay away from the "x slot" servers, satisfactory dedicated server performs very poorly in such environments. So you should be looking at root/dedicated servers.

desert walrus
#

The one exception to that seems to be indifferent broccoli, they appear to have very reasonable resource allocation.

fading quarry
#

Oh I am quite sure there exist cloud hosters who do care about balancing, but truth is that most will over provision commit their hardware to absurd lengths and you are left in a set and forget situation, unfortunately this is something they need to do since they are dirt cheap and practically only way to make a profit.

cloud mesa
rich palm
# fading quarry Oh I am quite sure there exist cloud hosters who do care about balancing, but tr...

Im actually working for a cloud provider. And we indeed never overprovision on machines with >2 vcpus. But thats not the problem, most CPUs of cloud providers are not suitable for single core applications, because the hypervisior usually consists of a CPU with a lot of cores (>32). This way the the turbo boost can practically never really kick in, which makes up the majority when testing single core perf.
Afaik some cloud providers have special offerings for that, but thats then very expensive (much more than simply renting a dedi)

fading quarry
# rich palm Im actually working for a cloud provider. And we indeed never overprovision on m...

I worked for a cloud hosting company as well for several years, and in that company it was all about maximizing profit and we did A LOT of overcommitting, since we knew that 90% of the time customers would have their vm's run practically on idle. Most heavy workloads would happen during night since most do reporting during that time, and we would just work out schedules for the customers in a order so that the entire data center would not kneel.

But you are absolutely correct most servers are not designed to deal with heavy single core workloads, though they exist but are extremely expensive, these just do not make financial sense.

twilit fox
#

I keep getting dropped from my factory server ):

twilit fox
#

oml... so I swapped ethernet cords.. because that's the only thing that's changed in the last week.. and now the client crashed. Version: 211839, IsEditor: No, IsPerforceBuild: No, BuildConfiguration: Shipping, Launcher: Steam, NetMode: Client, IsUsingMods: No

Fatal error: [File:D:/BuildAgent/work/78a794e88763017d/UE4/Engine/Source/Runtime/Core/Private/Windows/WindowsPlatformMisc.cpp] [Line: 475] Pure virtual function being called

VCRUNTIME140

thorny shore
#

that little bit tells nothing. More log please?

deep turret
#

it tells

#

Update your vc runtime maybe

#

it's a problem with windows functions

#

vcredist

#

https://aka.ms/vs/17/release/vc_redist.x64.exe
latest 64b version (this is Microsoft link)

#

you can check here :
https://support.microsoft.com/en-us/help/2977003/the-latest-supported-visual-c-downloads

#

if it's not a problem with vcredist then it's a bad code condition
(an abstract like base class is casted on a derived object
then base part of the object is stripped out/copied somewhere else where it is attempted to be used as if the object was whole :
it calls polymorphic methods which can not reflect in a whole object as the whole object is broken
so it fails by missing derived methods of the whole object)

lament cape
#

how to fix this issue

#

it spams console

grim mica
#

You dont atm, known issue

#

Others have mentioned they have used logrotate to keep it from eating all of your disk space

fading quarry
#

you can also create a link on the file that points to /dev/nul

thorny shore
#

I wonder if you could make it point to /dev/random too snuttstach_think

twilit fox
# deep turret it tells

Could that be causing issues with my dedi too? I didn't share the log because I didn't really see anything.. It seems to be working better now that I've switched the ethernet cables.. the old cable was still connected at 1000mbps and wasn't dropping at least that I've noticed. That's really the only thing that's changed since my server started having issues. I changed the cables monday.. so maybe that could be it

whole meadow
#

Sorry Have not been super active here; it would be great to know what are the critical isssues you guys are currently facing with dedicated server. I have seen couple of crashes on Proxmox VM; Log spamming issue on some servers hosted on linux.
Traveling through hypertubes have not been a great experience in DS. Trees/foliages growing back on DS and you cannot destroy them.
Also feel free to directly ping me here if there some critical issue you want to address.

desert walrus
#

Log spamming seems to be one of the worst of those you listed, I've seen several people report 5gb+ after just a day or two

whole meadow
desert walrus
#

It all seems to be along a similar vein at the very least

whole meadow
#

spamming only on servers hosted on a linux servers ?

desert walrus
#

It appears that way but I'm assuming that's just because -log is implicit?

#

I've never hosted on windows so not entirely sure of the functional difference there

grim mica
#

Confirmed does not happen on Windows

#

I'd say the log issue is the most common one, the other being Proxmox based Linux VMs not being set to "host" and instead using some sort of KVM emulated CPU

#

Everything else that i've seen/helped with has been some form of user error or the machine just not being up to spec (few cases of trying to run client+server on 16/32gb memory w/ U7)

#

The multihome/disablepacketrouting thing is also.. annoying since it doesnt default to that behavior (feel like it should be the default) as everyone that has an issue, once they apply multihome no longer has issues connecting. Easy fix at least however, so not too big of a deal imo.

whole meadow
grim mica
#

I just kind of avoid hypertubes on DS if I can, they are all over the place. Sometimes they work fine, sometimes they dont. Sometimes the cannon works, sometimes it kills me instantly. Normal hypertube path does weird things where it sounds like im entering the hypertube like 50 times over 10 seconds then resumes normally, sometimes I get stuck in the tube and simply cant get out without disconnecting and rejoining as it removes all inputs

#

#dedicated-servers message In this area for the most recent Proxmox stuff, the user provided a screenshot as well showing what needs to change

#

Talking about pinned messages.. the pin in here still has the fandom wiki page

vagrant turtle
#

hmm.. my DS uses proxmox as the host, but ubuntu 22.04 for the VM ... what kind of changes?

grim mica
#

The guest VMs running the Satis DS on U8 need to be set to 'host'

#

Not an issue on U7

vagrant turtle
#

It was running U8 just fine before this last patch, so that's what is throwing me off

grim mica
#

If yours is using the KVM64 or whatever.. that would be interesting if its working as its the first instance ive heard of that working

#

Then again, people only come here when they have issues, so we are a sample of problems only and not successes

vagrant turtle
#

trying to figure out what you mean by "set it to host" .. as in the VM to be set as host somehow?

grim mica
#

Type setting

#

I dont and have never run Proxmox, so no idea where/how to get to that config. I assume its similar to VMWare and there is some console with VM settings/configuration page

vagrant turtle
#

I'm looking for it, and not really seeing it yet, but I'll keep looking.. thanks

grim mica
vagrant turtle
#

ok, found it.

#

now to reboot and test.

grim mica
#

Was yours broken or working with it set to KVM64?

vagrant turtle
#

It was working with it set to KVM64(default)

#

before this last patch, that is on U8

grim mica
#

This only started with UE 5.2 iirc, it didnt happen with 5.1 so the early U8 was working but.. QA site is down so cant get build numbers lol

vagrant turtle
#

it is looking like it is firing up. sweet!

grim mica
#

Okay so worked on 249836 but not on 259435 or later without the CPU change

vagrant turtle
#

hmm.. now I'm getting a version mismatch .. 2632332 (game version) 261598 (server version) .. time to check for more updates...

grim mica
#

Speaking of issues.. just setting up my first resource well for this save and cannot get it to work at all.. Placed the pressurizer, powered it, then placed the nodes and it never started a cycle. Saved and restarted the DS and now the pressurizer shows 0/min output and no connected nodes. Deleted all of the nodes and pressurizer and still showing 0/min even after reloading the save in between the deletion. I guess i'll try to place it in single player now.

image.png
vagrant turtle
#

ok, there was a patch.. firing it back up now

#

ROFL.. now my client is patching LOL

#

hrm.. now the server is giving the error "could not open new router for an IPV6 type address, but shows my ipv4 address in it.. I have IPv6 disabled on the server

grim mica
#

Using multihome switch?

vagrant turtle
#

will have to double check, but I think so.

thorny shore
vagrant turtle
#

now to find out where the setting is again LOL

#

been a while since I re-configured this thing

grim mica
#

Its the launch parameter on the process/exe

vagrant turtle
#

I thought it was in a config file. I remember seeing that in the past..maybe I'm thinking of Icarus for that one...

#

ok, thanks.. got the -multihome=<local IP> working..

#

Thanks again. Everything is working again. I'll just have to remember to add the extra to the end of the command

thorny shore
#

You could make a script that executes it the same way everytime when you launch it

#

that way you don't have to remember all of the particulars (unless something is constantly changing).

vagrant turtle
#

yeah, I could. but just glad it is finally working.. won't have to leave my client open in order to let things go while I sleep LOL

#

that's the main reason to run a DS for me. let it do things while I'm not actually logged in

deep turret
#

So it crash and says "that feature of that object can't be used like that in that context and it makes absolutely no sense, the execution is completely ill and stale"

#

a bit like if I was trying to go make errands with a car on a mechanic's elevator

#

it makes no sense

#

it just absolutely fail any ongoing process

thorny shore
twilit fox
#

This is pretty much the wrong attitude to have.. wabash is correct.. you may see it as just a game server, but an attacker sees it as an open port. There's probably thousands of servers out there that do nothing but scan for open ports. I've been hit by them a few times..

calm cave
#

hi ! did anyone know well the dedicated servers ?

#

Got multiples bug like on dediacted non experimental there is bugs 😦

#

And on experimental i've got a timeout

twilit fox
#

That says it seems that the satisfactory server will drop packets if someone tries to scan it... which is really good

calm cave
#

My start is :
./FactoryServer.sh -Port={{SERVER_PORT}} -ServerQueryPort={{QUERY_PORT}} -BeaconPort={{BEACON_PORT}} -multihome=0.0.0.0

calm cave
#

Personally i've

image.png
twilit fox
calm cave
#

But the dedicated server look working soooo bad

twilit fox
#

So you're able to join, and then you get kicked?

calm cave
#

The thing is

No firewall on my pc,
No filter on the port of the dedicated so i see no problem 😦

twilit fox
#

It seemed pretty straight forward on windows for me. I think I got it working on Linux to.. sorta.

#

If you're using Linux did you allow satisfactory in ufw?

calm cave
calm cave
twilit fox
#

That just says the connection was closed. I've got no idea why. Maybe it'll say in your server log?

calm cave
calm cave
#

maybe packet routing is needed ?

#

Packet routing is fucking useless when you're working on you're bgp net 😂

#

[2023.11.01-19.44.30:140][409]LogDTLSHandler: EnableEncryption: Server
[2023.11.01-19.44.30:140][409]LogDTLSHandler: Warning: Empty certificate identifier
[2023.11.01-19.44.30:269][412]LogDTLSHandler: DTLSCertVerifyCallback: Verified
[2023.11.01-19.44.30:270][412]LogDTLSHandler: UpdateHandshake: Handshaking completed
[2023.11.01-19.44.59:751][295]LogNet: Warning: UNetConnection::Tick: Connection TIMED OUT. Closing connection.. Elapsed: 30.01, Real: 30.01, Good: 30.01, DriverTime: 110.09, Threshold: 30.00, [UNetConnection] RemoteAddr: 82.66.**.:57601, Name: IpConnection_2147482284, Driver: IpNetDriver_2147482295 IpNetDriver_2147482295, IsServer: YES, PC: NULL, Owner: NULL, UniqueId: INVALID

#

[2023.11.01-19.44.59:751][295]LogNet: Error: UEngine::BroadcastNetworkFailure: FailureType = ConnectionTimeout, ErrorString = UNetConnection::Tick: Connection TIMED OUT. Closing connection.

deep turret
calm cave
#

I think sts dedicated didn't like docker

deep turret
calm cave
#

i'm over pterodactyl panel

#

on the same machine on lgsm it work but not over the docker

deep turret
#

then there is probably something wrong with your docker networking

calm cave
#

that the fact 😂

deep turret
#

do you use -multihome setting ?

#

in the docker ?

calm cave
#

maybe that can be because docker proxy ? but i think not

#

docker proxy permit my node to have multiple ip's

twilit fox
#

My bad. I'm at work, and once you mention docker you're over my head

calm cave
twilit fox
#

I'd like to set something like that up on my server.. one day

deep turret
#

in the start script in the docker

#

you mean you use it with the ip ?

#

I'm asking that because I dont know if the ipv6/ipv4 connection bug got sorted out

thorny shore
#

Explorer bugged out when I logged back into the server, almost killed me.

image.png
#

I managed to dismantle it before it fell through the world.

drifting trellis
#

Mine (jiggling truck) was bouncing into space, taking damage from out of bounds because "too high" and had to bail and jet pack land... I rode it three times... wheee.

thorny shore
#

oof. I had left mine in a building, and when I came back it was invisible, but I figured, why not. Then when I entered it, it fell through the floor (took some damage), bounced and jiggled a lot (took more damage). I finally jumped out and carefully aimed my dismantle gun (it was also invisible most of the time).

rough arrow
#

@grim mica thanks again for your help the other day. turns out some of the networking issues were due to isp at the server site but both vm are serving games properly now. its not 100% perfect i do believe the last time i built an experimental server (update 6?) the game was more stable than now, but they are functinal and thats all i want for now.

grim mica
#

Nice, good to hear

rich palm
#

So the performance issues are really annoying. At some parts of my world there are basically none, at other parts you really notice the performance going down because for example the train gets slower (including a lot of rubber banding) and the tickrate down. Somehow it seems to be related to the creatures; as in those parts I see a lot of log messages about creature animations

mighty panther
#

Im, trying to set up a dedicated server. I have been trying this before but didn't really get far I think.
RN I have the server up and running with an already existing save from when I tried before.

My current issues are

  • Create game, Server setting, Manage Saves all require Authentication password. This I don't have and dont remember setting up. Can I reset it in any way?
  • I want to upload a local save from my friends computer, how do I do this the easiest way?
mighty panther
#
  • I am trying to locate ServerSetting.15777 but I can't find it. There is no folder for Saved games under FactoryGame/Saved/ other than Config and Logs
deep turret
# mighty panther Im, trying to set up a dedicated server. I have been trying this before but didn...
  1. you should have set up password, otherwise the server instance is not initialized, and so it will ask to set one
  2. the easiest way is to connect to the server through game's server manager and with admin authentication (the password we already are talking about) : go to manage save panel and upload a savefile
  3. make sure to look at the dedicated server's user local files
    for instance on windows when using nssm it is under system32/service/local/..
    on linux it is in ~/.config/Epic/FactoryGame/..
mighty panther
#

Thank you so much

deep turret
#

you can join ?

#

are you on latest update ?

kind lion
deep turret
#

tbh there are lot of networking issues lately, are you using a rented server ?

kind lion
#

nope but managet to fix it

deep turret
#

gj

scenic musk
# calm cave yes and no the both do the same

@deep turret multihome is for more complex networking / multiple gateways. Your 0.0.0.0 means its accessible to any IP reaching the host. You might find something like Portainer easier for managing docker.

deep turret
#

especially with ipv6 mixing

scenic musk
#

Oh I see gotcha

errant geyser
#

What issues are we talking? (other than the log spam that makes the servers unusable atm)

deep turret
#

first update 8 updates confused itself over initialization of client connection by asking the client to reconnect on an other selected ip address
which
1: included ipv6 addresses for ipv4 clients
2: included loopback addresses (with ipv6)
3: having address handling AFTER the connection is recognized is .. weird

I don't know if it's still happening
I never removed my -multihome=0.0.0.0 argument

#

btw client still is not able to retry/check the server connection regularly
which force user to restart the game or re-add the server for it to be checked (and found)
and I don't know if the server manager still fail to check ipv4 connectivity when using dns with ipv6(AAAA) records

#

I should test that

errant geyser
#

I don't think the server ever asks the clients to reconnect using the given IP, the clients always use the hostname that the user specified to connect to the beacon, the game and do lightweight query

deep turret
#

then the server was confusing itself alone

errant geyser
#

The only thing that might go wrong is if your DNS record resolves to multiple addresses

#

In that case the client will try to ping them all using the lightweight query, which might result in inconsistent results if they resolve to different servers

#

The client not reattempting the beacon connection regularly is a known issue that will hopefully be resolved before the next major update

calm cave
verbal sundial
#

when better dedicated server? :C