Bug Report

After upgrading the server to 3.1.1, I'm unable to login. No client or server crashes, just disconnects. I have the 3.0.8 and 3.1.1 (post conversion) world save files available on dropbox if they are needed.

Expected Behaviour

To be able to successfully login to the server.

Possible Solution

N/A

Steps to Reproduce (for bugs)

1. Turn on server
2. Convert 3.0.8 world save to 3.1.1
3. Try to login

Logs

• Client/Server Log:
  https://gist.github.com/abortu/d30f78617e4df9975ba2e02654935998
• Crash Log:
  N/A, no crashes occur.

Client Information

• Modpack Version: 3.1.1
• Java Version: 64-Bit Server VM, version 1.8.0_191, running on Linux:amd64:4.20.15-200.fc29.x86_64
• Launcher Used: MultiMC 0.6.4
• Memory Allocated: 8GB
• Server/LAN/Single Player:
  MP server (local network)
• Optifine Installed: No
• Shaders Enabled: No

World Information

• Modpack Version world created in: 3.0.8

• Additional Content Installed: N/A

Server Information

• Java Version: OpenJDK 64-Bit Server VM, version 1.8.0_171, running on Linux:amd64:4.20.16-200.fc29.x86_64
• Operating System: Linux - Fedora 29
• Hoster/Hosting Solution: N/A
• Sponge (Non-Vanilla Forge) Server: N/A

Hello, it appears your client log is incomplete so I'm not seeing if there's an error on the client side causing this - it only goes up to some point during init and does not include your connection attempts.

Your server log only flags that a time out is occurring during connection. Are you using the bundled startup script? If not you might consider adding the argument -Dfml.readTimeout=180 to your environment. You can also try adding this to the client.

If this does not resolve your issue, you will probably find you get more and quicker support for this issue on the discord. https://discord.gg/darkosto in the #server-discussion channel

You need to view the raw version, gist truncates large files.

I'll try to add that option and report back.

I tried that, besides it normally warns when it's truncating.

Weird, I just saw the raw version is messed up. I'll re-upload and send a link.

Re-uploaded client log link: https://gist.githubusercontent.com/abortu/fd04717afccdea04e3c5f03d7181c466/raw/df61fa3673436b239a8b8c91dc1496c9ad9288de/sevtech_311_client.log

Yeah, your connection attempt is lasting exactly 30 seconds before being timed out and I'm not seeing anything else particularly worrying. I suspect the argument I mentioned above will solve this. If not, please do check out the discord, as I need sleep!

I added the argument to both client and server, it still fails. I'll hit up discord, thanks for the help!

New Logs:
https://gist.github.com/abortu/920c33249c2b770e3bcfc0903e559455

It appears this is an issue with the server running on OpenJDK 1.8 under Linux. I'm still testing and will submit my findings once I'm done. Switching to Oracle's java has let me login. I tried this locally and and am now going to test this on my server to see if the results are the same.

I'm able to login now. Here are the steps I tried and the results.

Removed Docker from the equation, ran server as usual (ServerStart.sh)

OpenJDK 1.8_201: Same behavior, unable to connect, no errors. Just looks like I connected and disconnected normally.

Oracle 1.8_201:

• Running on the already converted save (done by OpenJDK): logged in, but lost all of my stage information.
• Re ran the conversion process using a fresh copy of my 3.0.8 save. Connects and staging appears to be correct now. No manually giving stages needed.

Re-ran using docker (with Oracle 1.8)

• Ran as intended, connects, staging is fine.

Thanks for all the help on here and discord! I'll close this issue, just putting the info out there in case someone comes across a similar situation.