Joltz
commented
2 years ago Just going to dump my discord messages in here as I was having similar issues on my end until switching from our ISP's provided CGNAT to a static IP address.
Jolts — 11/08/2021 Also been having LNL connection issues that look identical to @Duskitten. Disabled every possible firewall and DMZ'd the router just as a sanity check and got the same results. Logs don't really reveal much but here's a sample from mine.
I had one successful connection via LNL Relay (shown here) but even when it is successful, it quickly becomes unstable and eventually get dropped completely. Wasn't an issue a few patches ago and really only started occurring commonly in the past two weeks. Also something I should note is that steam sockets connect 100% of the time they're available but have the typical issues commonly associated with them with regards to stability and packet queuing. I am also behind a CGNAT which is possibly why it always uses the relay instead of NAT punchthrough.
I am also behind a CGNAT which is possibly why it always uses the relay instead of NAT punchthrough.
Jolts — 11/10/2021 Another interesting thing happened today. I removed our AP router and put a dumb switch in its place and now I get a prompt about LNL Direct-IP when connecting to my BF on our local network (direct LAN connections never worked before) and it's still failing with the same error message.
10:49:02 PM.523 ( 45 FPS) Joining session: lnl://172.16.0.195:58190/ 10:49:02 PM.526 ( 45 FPS) NetworkInitStart 10:49:02 PM.526 ( 45 FPS) Network manager: NetX.LNL_Manager - priority: 0 10:49:02 PM.526 ( 45 FPS) Network manager: FrooxEngine.SteamNetworkManager - priority: -100 10:49:02 PM.529 ( 45 FPS) Connecting to: lnl://172.16.0.195:58190/ 10:49:08 PM.442 ( 45 FPS) Disconnected: 172.16.0.195:58190, reason: ConnectionFailed, socketErrorCode: Success 10:49:08 PM.442 ( 45 FPS) Connection failed: ConnectionFailed 10:49:08 PM.442 ( 45 FPS) All protocols failed to establish connection
Here's what the log looks like when it uses SNS fallback
Jolts — 11/11/2021 Another entry into this saga. We decided to swap routers entirely and LNL is still nonfunctioning. LAN connections are working again but LNL over WAN is still barely working. Going to request a static IP from our ISP tomorrow to try and remove the CGNAT from the equation and see if that helps.
Talt — 11/12/2021 it always goes like Loading... Establishing connection LNL NAT something 1 2 3 4,LNL relay and goes failed only some session im able to join
Jolts — 11/12/2021 You are probably experiencing the same problem @Duskitten and I are having. You can check my message history for more details. I'm currently working on getting a static IP from my our ISP set up and seeing if that resolves the issue.
Rottex — 11/12/2021 For the sake of completeness I just want to inform that I also have these wierd issues since I have come back from holiday. Can't connect to our VM any more (Can't punch-through and get 03:26:15.038 (100 FPS) NAT Punchthrough failed, Connecting to Relay 03:26:21.581 (100 FPS) Disconnected: , reason: ConnectionFailed, socketErrorCode: 0 03:26:21.581 (100 FPS) Connection failed: ConnectionFailed 03:26:21.581 (100 FPS) All protocols failed to establish connection )
plus also reported the UNITYTLS_X509VERIFY_NOT_DONE / SSL Problem I had the last days.
Also tried everything /different routers (even with mobile-hotspot 🙂 ), double-checked time on the VM in the hostingcenter and in my LAN. reinstalled complete neos and Steam. Double-checked the f+cking M$ firewall on my PC. No joy. I even see the relayed packets from my client coming in on the device if I check with "tcpdump -i ens3 -nn dst 172.28.231.14 and not port 22" ?(
Strange, my colleagues can work with the headless server with no problems.
Jolts — 11/12/2021 Well it seems we finally ended our saga. Getting off the CGNAT and getting a static IP fixed all of our LNL issues (and as an added bonus, nat punchthrough actually works now).
I'm guessing this is a unintended regression when switching from the old LNL library to the updated one but I can only speculate. I know most people aren't in a position where they can just switch off CGNAT on a whim (especially if you're using something like StarLink).
I should also note that I couldn't connect to lnl-nat sessions at all most of the time as opposed to connecting and then getting dropped soon after (although that was always guaranteed if I did somehow manage to connect).
kulzae
shiftyscales
3x1t-5tyl3
Describe the bug?
When using LNL-Nat instead of pure LNL (Direct ip connect lnl://ipadress/) the connection times out exactly 5 minutes after connecting.
Relevant issues
None applicable. There's some that seem similar but aren't.
To Reproduce
None/Not reproducable easily
Expected behavior
For the connection to last or for it to use LNL directly
Log Files
Basically empty. I'd have to enable verbose logging to get more relevant info.
Screenshots
No response
How often does it happen?
Always
Does the bug persist after restarting Neos?
Yes
Neos Version Number
Beta 2021.11.10.1253
What Platforms does this occur on?
Windows, Linux
Link to Reproduction Item/World
No response
Did this work before?
Yes
If it worked before, on which build?
Pre-lnl update build seemed to work fine
Additional context
When connecting via direct IP Adress to my headless (lnl://whateveriphere/ or steam-neos://) they seem to be perfectly able to connect and keep the connection. When connecting with lnl-nat:// specifically it seems to timeout after exactly 5 minutes. Not exactly sure why?
Reporters
Duskitten#44553x1t_5tyl3#0001