Debugging and Patching Docker

2 minute read Published: 2020-02-29

Containerization technologies like Docker enable efficient deployment, portability, and scalability. This guide provides a structured approach to debugging and patching Docker containers, emphasizing technical precision and operational clarity.

Terminologies

Docker: A platform for containerizing applications using images.
Image: A lightweight, standalone self-contained executable that contains everything needed to run a application.
Container: A running instance of an image, isolated from other containers.

Information Gathering

Container Inspection

List running containers:
```
docker ps
```
List all containers (including exited):
```
docker ps -a
```
Filter by status (e.g., exited):
```
docker ps -a -f "status=exited"
```
Retrieve container logs:
```
docker logs <container_name>
```

Inspect container details:

docker container inspect <container_name>

Image Inspection

List all images:
```
docker images
```

Inspect image details:

docker image inspect <image_name>:<image_tag>

Output Formatting

Extract exited container names:

docker ps -a -f "status=exited" --format "{{json .Names}}"

Live Debugging

Docker is Running

Enter container shell:

docker exec -it <container_name> /bin/sh

List container contents:
```
docker exec -t <container_name> ls ./
```

Docker Has Exited

Retrieve image and tag:

docker ps -f "name=<container_name>" --format "{{json .Image}}"

Run container with shell:

docker run -it --entrypoint sh <image_name>:<image_tag>

Execute command:

docker run -it --entrypoint ls <image_name>:<image_tag> ./

Post Debugging

Rollback

Revert to previous tag:

docker tag <image_name>:<previous_tag> <image_name>:latest

Patching

Commit live-debugging container:

docker container commit <container_name> <image_name>:<patched_tag>

Kill container:
```
docker container kill <container_name>
```

Tag patched version:

docker tag <image_name>:<patched_tag> <image_name>:latest

Restart service:
```
docker restart <service_name>
```

Docker with Changed Entry Point

Retrieve old entry point:

docker image inspect <image_name>:<tag> --format "Entrypoint {{json .Config.Entrypoint}}"

Retrieve old command:

docker image inspect <image_name>:<tag> --format "CMD {{json .Config.Cmd}}"

Commit with old configuration:

docker container commit -change "<old_entrypoint>" -change "<old_cmd>" <container_name> <image_name>:<patched_tag>

Kill container:
```
docker container kill <container_name>
```

Tag patched version:

docker tag <image_name>:<patched_tag> <image_name>:latest

Restart service:
```
docker restart <service_name>
```

Cleaning Up

Remove exited containers:

docker ps -a -f "status=exited" --format "{{json .Names}}" | xargs -r docker rm

Prune unused images:
```
docker images prune
```

Tips

Use --q for quiet mode:
```
docker ps --q
```

Transfer files between container and host:

docker cp <container_name>:<path> <local_path>

From container:

docker cp <container_name>:<path> <local_path>

To container:

docker cp <local_path> <container_name>:<path>

Conclusion

Effective Docker debugging and patching requires systematic analysis, precise command execution, and careful cleanup. This guide provides a structured approach to identify issues, resolve them, and maintain containerized environments with reliability and efficiency.