# docker "Error response from daemon: Conflict. The container name is already in use"

TL;DR: A stopped container with that name still exists. Remove it with `docker rm [name]`, or run one-off containers with `--rm` so they delete themselves. In CI, add a cleanup step before the run so one failed build does not poison the next.

```text
Error response from daemon: Conflict. The container name "/web" is already in use by container "a1b2c3d4e5f6"
```

## Steps

1. List it: `docker ps -a --filter name=[name]`. Confirm the stale container and whether it is running or exited.
**Expected:** You see the conflicting container and its state.

2. If it is exited and disposable, remove it and rerun:
```
docker rm [name]
```
**Expected:** The next `docker run --name [name]` starts clean.

3. If it is running and should not be, stop it first, then remove it. If it should be running, you have two things fighting over one name; rename one of them.
**Expected:** The name is free, or the real owner is identified.

4. Prevent recurrence for one-off runs: add `--rm` so finished containers delete themselves.
**Expected:** No stale containers accumulate between runs.

5. In CI, add a pre-step that removes any container with the job's name, or generate a unique name per run from the build id.
**Expected:** Name conflicts stop flaking the pipeline.

6. Compose users: `docker compose down` removes the project's containers. Conflicts across compose projects mean two projects share a name; give them distinct project names with `-p`.
**Expected:** Clean slate, no cross-project collision.

## Use this when
- `docker run --name X` fails with the Conflict message

## Not for this skill when
- The error is `port is already allocated`. That is a port conflict, same family but a different resource
- The error is `unable to delete ... image is being used`. That is an image reference conflict, not a name
- The daemon is down. Nothing runs at all

## Variant phrasings
### Conflict. The container name "/name" is already in use by container "abc123"
The full form of the error with both names shown.

## Why it happens
Container names are unique per daemon, and stopped containers keep their names until removed. A crashed or interrupted run leaves the name claimed, so the next identical run collides with the leftover of the last one.

## Edge cases
- `--rm` and `--name` work fine together; the container is removed when it exits
- `docker rm -f` handles stuck containers that ignore a plain stop
- Two compose projects in directories with the same name share a default project name and will fight; `-p` fixes it permanently

## Provenance

Resolved from the public thread: https://vectle.com/posts/pst_vEqb8Blbe0xtYwdVVXVgQQ