Replies: 1 comment 1 reply
-
Hi, @ALBDN maybe you can try |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi!
I'm trying to use this great ImageSpec feature with docker hub registry but I'm facing this error:
Failed with Unknown Exception <class 'Exception'> Reason: failed to run command envd build --path /tmp/flyte1ma3cexz/control_plane_metadata/local_flytekit/ab5988b3a69fba5645e9128922535299 --platform linux/amd64 --output type=image,name=alevelbdn/flytekit/flytekit:aLJkDnqKAzK_4X_sEcGCaA,push=true with error b'time="2024-05-24T10:32:55+02:00" level=fatal msg=exit app=envd error="failed to build the image: failed to build: failed to wait error group: failed to solve LLB: failed to solve: server message: insufficient_scope: authorization failed" version=v0.3.45\n'
I know the error shows insufficient_scope: authorization failed but I did a docker login first which turned out to be successful. I've also tried to build the image and push it to the docker registry with:
docker build -t alevelbdn/flytekit:v1 .
docker push alevelbdn/flytekit:v1
and it worked without any problem. I can find the image in my registry.
Now when trying with
image_definition = ImageSpec( packages=["pandas", "numpy", "joblib", "scikit-learn", "xgboost"], registry="alevelbdn/flytekit", builder="envd" )
in my python script and pyflyte run with:
pyflyte run --remote -p house-pricest -d development house_price_predictor.py house_price_predictor_trainer
I end up with the above error.
My environment: Flyte is deployed on a local Kubernetes cluster with the flight-binary helm chart.
I can deploy a simple Flyte pipeline (without dependencies) and run it successfully on the cluster.
As I'm running out of idea It would be really nice if someone who faced the above issue or have any idea could advise...
Thank you!
Beta Was this translation helpful? Give feedback.
All reactions