Unable to start after upgrading to RC11.1 from RC10

[RobotHanzo]

Stacktrace

Error opening zip file or JAR manifest missing : /home/ubuntu/server/temp/caches/wrapper.jar

Actions to reproduce

Start the server

CloudNet version

.11 02:55:21.610] INFO :
[06.11 02:55:21.610] INFO : CloudNet Blizzard 4.0.0-RC12-SNAPSHOT 7aaa1cb
[06.11 02:55:21.610] INFO : Discord: https://discord.cloudnetservice.eu/
[06.11 02:55:21.611] INFO :
[06.11 02:55:21.611] INFO : ClusterId: c6b7bcad--401c--069f7903704c
[06.11 02:55:21.611] INFO : NodeId: Node-1
[06.11 02:55:21.611] INFO : Head-NodeId: Node-1
[06.11 02:55:21.611] INFO : CPU usage: (P/S) 3.18/18.86/100%
[06.11 02:55:21.611] INFO : Node services memory allocation (U/R/M): 8448/12544/60000 MB
[06.11 02:55:21.611] INFO : Threads: 61
[06.11 02:55:21.612] INFO : Heap usage: 86/256MB
[06.11 02:55:21.612] INFO : JVM: Eclipse Adoptium 23 (OpenJDK 64-Bit Server VM 23.0.1+11)
[06.11 02:55:21.612] INFO : Update Repo: CloudNetService/launchermeta, Update Branch: nightly

Other

I tried using both the beta and nightly branch, though none of which worked

Issue uniqueness

• Yes, this issue is unique. There are no similar issues.

[0utplay]

Does this happen with a clean restart and deleting the temp folder before starting?

[RobotHanzo]

Does this happen with a clean restart and deleting the temp folder before starting?

Yes, I tried deleting temp and redownloading the core jars, unfortunately neither worked

[RobotHanzo]

Does this happen with a clean restart and deleting the temp folder before starting?

FYI, the full error is as below (found in the log file)

Error occurred during initialization of VM
agent library failed Agent_OnLoad: instrument
Error opening zip file or JAR manifest missing : /home/ubuntu/server/temp/caches/wrapper.jar

Also, I use the docker module with the image eclipse-temurin:23-jdk-noble

[0utplay]

Does it also happen without docker? Can you provide an minimal reproducer?

[RobotHanzo]

Does it also happen without docker? Can you provide an minimal reproducer?

I tried creating a completely new instance both with and without docker, interestingly they both worked
I cant figure out whats causing this (I tried disabling all my custom modules, it didn’t help)

[RobotHanzo]

Does it also happen without docker? Can you provide an minimal reproducer?

However, disabling docker on the broken instance fixed the issue

[RobotHanzo]

Something ever weirder:
Copying the entire instance folder to somewhere else like /tmp fixed the issue
But as long as the instance folder is in the original parent directory (i.e. /home/ubuntu/), it wouldn't start even if renamed
Note: I made sure wrapper.jar can be read by the user running cloudnet