Ignore: 🐛 Distributed Lock Failure Causing Duplicate Chat Room Joins

[psychology50]

작업 이유

Initially, I thought the issue was due to a missing validation check for users already joined in a chat room, so I added it. However, after further investigation, I realized that the problem lies elsewhere. (I haven't found the root cause yet.)

작업 사항

I DON'T KNOW WHY THIS MAKE A ISSUE!!!!!!!!!! 🤯

I found the root cause of the issue through debugging—duplicate user entries were saved in the chat room.

However, this shouldn’t be possible
I had already anticipated this issue and applied a distributed lock at the service level.
Additionally, the validation to check for duplicate users was being executed in ChatMemberRdbService.createMember()

But it still failed.

Logs Analysis

[2025-02-05 17:03:06:433949759] INFO  1 --- [io-8080-exec-10] k.c.p.d.c.aop.DistributedLockAspect      : 88 : Redisson Lock 진입 : execute LOCK:chat-room-join-674887267360394154
[2025-02-05 17:03:35:433979121] INFO  1 --- [io-8080-exec-27] k.c.p.d.c.aop.DistributedLockAspect      : 3429 : Redisson Lock 진입 : execute LOCK:chat-room-join-674887267360394154
[2025-02-05 17:03:37:433981077] INFO  1 --- [io-8080-exec-10] k.c.p.d.c.aop.DistributedLockAspect      : 88 : Redisson Lock 해제 : execute LOCK:chat-room-join-674887267360394154
[2025-02-05 17:03:37:433981321] INFO  1 --- [io-8080-exec-27] k.c.p.d.c.aop.DistributedLockAspect      : 3429 : Redisson Lock 해제 : execute LOCK:chat-room-join-674887267360394154
[2025-02-05 17:03:39:433982954] ERROR 1 --- [io-8080-exec-10] k.c.p.d.c.aop.DistributedLockAspect      : Redisson lock is already unlocked: execute LOCK:chat-room-join-674887267360394154

1. Thread ID 88 acquires the lock and starts execution at 17:02:50.
2. Thread ID 3429 acquires the lock and starts execution at 17:03:34.
3. The lock for Thread ID 88 is released at 17:03:37.
4. The lock for Thread ID 3429 is released at 17:03:37.

These logs clearly indicate that two threads acquired the lock simultaneously. 😂😂😂

I’ve been trying to reproduce the issue under the same conditions, but I haven't been able to.

리뷰어가 중점적으로 확인해야 하는 부분

• For now, I’m pausing the investigation on this issue since I need to focus on another urgent task.
• Will revisit once I have more bandwidth.

발견한 이슈

• Redisson lock failure allowed simultaneous thread execution.
• Need further investigation to determine why the distributed lock mechanism failed.