The error you’re experiencing indicates that Docker is incorrectly trying to pull the ARM64 version of the torizon/binfmt image instead of the AMD64 version. This is similar to the issue documented in the community forums (1).
Root Cause Analysis
Based on the community reports, this problem occurs when Docker thinks it’s running on an ARM64 system when it’s actually running on x86_64 (1). The issue is that the DOCKER_HOST environment variable might be set incorrectly, causing Docker to connect to a remote ARM64 system instead of your local x86_64 WSL environment (1).
Solution Steps
Step 1: Check Docker Environment Variables
In your WSL terminal, run:
printenv | grep DOCKER
Step 2: Check Docker Info
Run this command to see what Docker thinks your system architecture is:
docker info
Look for the “Architecture” field in the output (1). If it shows aarch64 instead of x86_64, that’s the problem.
Step 3: Unset DOCKER_HOST (if set)
If you find that DOCKER_HOST is set, unset it:
unset DOCKER_HOST
Step 4: Restart VS Code
Sometimes the environment variables get cached. As mentioned in the community forum, “have you tried turning it off and on?” often works (1). Close VS Code completely and restart it.
Step 5: Verify Docker is Running Locally
After unsetting any problematic environment variables, verify Docker is connecting to your local WSL instance:
docker info
The output should show:
- Architecture:
x86_64 - Operating System: Should reference your WSL Ubuntu installation
Step 6: Re-enable ARM Emulation
Once Docker is connecting to the correct local instance, run the ARM emulation command:
docker run --rm -it --privileged torizon/binfmt
VS Code Settings Check
Also check your VS Code settings in .vscode/settings.json and ensure DOCKER_HOST is not set or is empty:
"docker.environment": {
"DOCKER_HOST": ""
}
This issue is commonly caused by VS Code extension settings that redirect Docker commands to the target device instead of the host development machine (1).
Best regards, Toradex AI Assistant.
If I answered your question, please tick the “Solution” checkbox.
For follow-up questions to me, please use @ToradexAI in your post.