I had a wg-easy (https://github.com/wg-easy/wg-easy) docker container running well for months, but one day I could not connect. I found the container stopped so I attempted to restart it, but it complained that the port 51820/udp was already in use.
I could not find a process associated with the port using netstat -tulpn.
First thing I tried was making sure I reinstalled the proper docker from the website's documentation (since I was using snap before). Then I made sure all the docker containers were removed. After that, I did some reboots and restarted networking a few times. This was repeated after each of the following attempts to fix the issue.
I tested a solution that seemed promising from studgeek: netstat shows a listening port with no pid but lsof does not
However, after using the inode associated with the port, I was only able to find an associated PID once. After attempting to kill that PID, no other PIDs associated with that port appeared again (I reran ss -tulpne to get the new inode for the open port).
netstat -tulpne shows
udp UNCONN 0 0 0.0.0.0:51820 0.0.0.0:* ino:10007 sk:5 cgroup:unreachable:138a <->
I also tried using fuser and rpcinfo -p as suggested by several answers on stack overflow but to no avail.
Appreciate any help! The fact that it persists across reboots really has me feeling a little stuck just short of completely reinstalling Ubuntu. I also tried reboots with docker/docker.socket disabled and stopped.
Additional details: This is running on Ubuntu 24.02 Server on a mini pc with 8 GB RAM.
