I have an Azure pipeline that runs torizoncore-builder (tcb) 3.12 to build a custom Tlorizon OS + some bundled containers (from several sources).
Everything works fine.
I’ve just changed "tcb.version": "3.13.0",
in settings.json to use v3.13.0
And now the pipeline cannot be completed anymore, with the following error:
Starting DIND container
Using Docker host "tcp://127.0.0.1:22376"
Connecting to Docker Daemon at "tcp://127.0.0.1:22376"
Attempting to log in to registry 'xxxxxxxx' with username=yyyyyyy
Fetching container image arm64v8/redis:6.2.17-bookworm in service redis
Fetching container image xxxxxxx/yyy:latest in service yyy
Fetching container image xxxxxxx/zzz:latest in service zzz
Fetching container image torizon/chromium-imx8:4.4.0 in service chromium-imx8
Stopping DIND container
Error: container images download failed: 404 Client Error for https://127.0.0.1:22376/v1.40/images/torizon/chromium-imx8:4.4.0/json: Not Found ("no such image: torizon/chromium-imx8:4.4.0: No such image: torizon/chromium-imx8:4.4.0")
Setting up TorizonCore Builder with version 3.13.0.
It seems that other containers from dockerhub and from private container registry can be fetched, but the container from torizon/ cannot.
Is there something different with tcb 3.13?
Hi, I’m the Toradex AI Assistant. The Toradex Support team will follow up with you shortly!
Did you like my answer?
Hello,
It seems you’re encountering an issue while trying to fetch the chromium-imx8 container. This container is indeed a valid Torizon image designed to run web-based GUIs using the Chromium browser. You can find more details in the Torizon documentation.
When working with TorizonCore Builder, especially for i.MX8-based System on Modules (SoMs), it’s important to specify the platform correctly when bundling multi-platform container images. You can do this with the following command:
Additionally, if you’re using private registries such as Azure Container Registry (ACR), ensure that you have configured authentication properly. You can log in using:
This is a known issue that we just uncovered within the past 24 hours. We’re still investigating the issue and discussing possible solutions. For the short-term please go back to using version 3.12.0 of TorizonCore Builder.
By the way, just to confirm the part of your pipeline that is failing is related to the bundling of container images into a custom OS image, correct? Or is it failing on some other part?
The first error comes at the very beginning, just after having pulled torizon/torizoncore-builder the pipeline starts fetching the containers for the bundle.
And here I get the error
Starting DIND container
Using Docker host "tcp://127.0.0.1:22376"
Connecting to Docker Daemon at "tcp://127.0.0.1:22376"
Attempting to log in to registry 'xxxxxxxx' with username=yyyyyyy
Fetching container image arm64v8/redis:6.2.17-bookworm in service redis
Fetching container image xxxxxxx/yyy:latest in service yyy
Fetching container image xxxxxxx/zzz:latest in service zzz
Fetching container image torizon/chromium-imx8:4.4.0 in service chromium-imx8
Stopping DIND container
Error: container images download failed: 404 Client Error for https://127.0.0.1:22376/v1.40/images/torizon/chromium-imx8:4.4.0/json: Not Found ("no such image: torizon/chromium-imx8:4.4.0: No such image: torizon/chromium-imx8:4.4.0")
Setting up TorizonCore Builder with version 3.13.0.
Then the pipeline downloads the Torizon OS base and recompiles with customizations (successfully).
At the end, during the Bundling...
step, it fails again because the bundle directory is not found.