How to fix OCI runtime create failed: container init process failed in Docker

The OCI runtime error is usually followed by more specific information. Look for the trailing part of the error:

docker run your-image 2>&1 | tail -20

Common trailing messages and their meanings:
- executable file not found in $PATH - Missing binary
- permission denied - File permission issue
- exec format error - Wrong architecture or shebang issue
- no such file or directory - Missing file or bad line endings
- read-only file system - Volume mount issue

Inspect the image to see what command it's trying to run:

docker inspect --format='Entrypoint={{.Config.Entrypoint}} Cmd={{.Config.Cmd}}' your-image

Then verify the executable exists inside the image:

# Start a shell in the image (if available)
docker run --rm -it --entrypoint /bin/sh your-image -c "which your-command"

# Or list files in the expected path
docker run --rm -it --entrypoint /bin/sh your-image -c "ls -la /path/to/entrypoint"

If using a minimal base image like alpine or scratch, ensure all required binaries are present.

Shell scripts created on Windows often have CRLF line endings that cause "no such file or directory" errors on Linux:

# Check for CRLF in your script
file entrypoint.sh
# Or
cat -v entrypoint.sh | head -5  # Look for ^M characters

Fix: Convert to Unix line endings:

# Using dos2unix
dos2unix entrypoint.sh

# Or using sed
sed -i 's/\r$//' entrypoint.sh

# Or in Dockerfile
RUN sed -i 's/\r$//' /entrypoint.sh

Prevent future issues: Configure Git to handle line endings:

git config --global core.autocrlf input

Scripts need execute permissions to run. Add this to your Dockerfile:

COPY entrypoint.sh /entrypoint.sh
RUN chmod +x /entrypoint.sh
ENTRYPOINT ["/entrypoint.sh"]

Or fix permissions before building:

chmod +x entrypoint.sh
git update-index --chmod=+x entrypoint.sh  # Preserve in Git

Outdated Docker versions can have runc compatibility issues. Docker 25+ resolves many OCI runtime issues:

# Check current version
docker version

# Update on Ubuntu/Debian
sudo apt update && sudo apt upgrade docker-ce docker-ce-cli containerd.io

# Update on CentOS/RHEL
sudo yum update docker-ce docker-ce-cli containerd.io

After updating, restart Docker:

sudo systemctl restart docker

Corrupted containers or images can cause this error. Clean up and retry:

# Stop all containers
docker stop $(docker ps -aq) 2>/dev/null

# Remove stopped containers
docker container prune -f

# Remove unused images
docker image prune -a -f

# Restart Docker daemon
sudo systemctl restart docker

# Try running your container again
docker run your-image

If AppArmor or SELinux is causing the issue, test by temporarily disabling security confinement:

docker run --rm --security-opt seccomp=unconfined --security-opt apparmor=unconfined your-image

Warning: Only use this for testing. If it works, investigate the proper security policy fix rather than running production containers without security confinement.

For SELinux systems, check audit logs:

sudo ausearch -m avc -ts recent | grep docker

If running an image built for a different architecture, you'll get "exec format error":

# Check your system architecture
uname -m

# Check image architecture
docker inspect --format='{{.Architecture}}' your-image

Solutions:
- Pull the correct architecture: docker pull --platform linux/amd64 your-image
- Use multi-arch builds
- Enable QEMU emulation: docker run --privileged --rm tonistiigi/binfmt --install all