How to Fix Docker No Space Left on Device Errors Without Just Deleting Random Images and Hoping the Disk Forgives You

Why this failure gets messy: Docker can fill a machine gradually through old images, stopped containers, layers, build cache, and volumes, which means random cleanup often removes the wrong thing first.

Typical errors:

no space left on device

or builds failing partway through writes.

Step 1: measure before deleting

Start here:

docker system df
docker image ls
docker volume ls

This tells you whether the bulk is in:

1. images
2. containers
3. local volumes
4. build cache

Step 2: remove clearly disposable layers

Unused images:

docker image prune -a

Stopped containers:

docker container prune

Unused build cache:

docker builder prune

Be deliberate. -a is powerful. Do not run it while assuming every image can be re-pulled cheaply in your environment.

Step 3: inspect volumes carefully

Volumes can hide the largest forgotten data sets.

docker volume prune

But only after you know those volumes are disposable. Databases and local app state often live there.

Step 4: rebuild with less churn

If the machine keeps filling up, the real fix may be reducing rebuild waste:

1. smaller base images
2. better layer caching
3. fewer redundant local environments
4. scheduled cleanup

Quick recovery path

If you need fast relief and understand the impact:

docker system prune

For a more aggressive cleanup:

docker system prune -a

Bottom line

No space left on device in Docker is a storage accounting problem, not a mystical daemon curse. Measure first, then prune the specific layer types creating the pressure.

Sources

• Docker prune documentation
• Docker system df