I am using a container-based Jenkins architecture, where the controller is only responsible for scheduling jobs, and the agents execute the actual tasks. Recently, I encountered an issue while running a job that involves a long-running process.
When the job was executed on an agent, it was terminated automatically after waiting for 30 minutes without response due to a timeout. However, I noticed that the process, which was supposed to be running on the agent, was actually found running inside the controller’s application container.
Could anyone help me understand why this might be happening? Is there something wrong with my configuration, or is this a potential issue with the container setup?