Container keeps restarting
It comes up, dies a few seconds later, and comes back. The restart count climbs and the logs scroll past before you can read them.
What it does. It reads the container state, the restart policy, the health check and the port map, then tells you which of those is actually failing rather than listing all four as possibilities.
Container exited with code 137
The service dies under load and nothing in the application log explains it, because the application never got to write one.
What it does. Exit 137 means the kernel killed it, almost always the memory limit. It checks the configured limit against usage history and tells you what the ceiling is and what the job needs.
docker compose up fails
It works on one machine and not another, and the error mentions a file or a port rather than your code.
What it does. It reads the compose file, resolves the project name the way Compose does, and checks whether the port is taken, the bind-mount path exists, and the image is present.
Port is already allocated
A container will not start because something else is holding the port, and it is not obvious what.
What it does. It lists what is bound to that port, whether the holder is another container or a host process, then shows you the options before changing anything.
The app cannot reach the database
One container cannot talk to another, and every guide online tells you to check five different things.
What it does. It checks the network both containers are on, whether the target is actually up, and whether the hostname resolves inside the container. It reads the log line rather than any credential value.
The server is out of space
Deploys start failing and the disk is full, but the offender is not obvious.
What it does. It finds what is consuming the space, which is usually unrotated container logs or old images, and offers to clear it with a retention limit so it does not recur.