Graded lab “Implementing Canary Releases of TensorFlow model deployments with Kubernetes and Istio”
The step kubectl apply -f tf-serving/service.yaml
fails with a CrashLoopBackOff for tf-serving
Graded lab “Implementing Canary Releases of TensorFlow model deployments with Kubernetes and Istio”
The step kubectl apply -f tf-serving/service.yaml
fails with a CrashLoopBackOff for tf-serving
Downgrading tensorflow/serving to 2.8.3 helps. You should really stop heaving (implicit) latest tags on all the images. It. Will. Break. Eventually.
Was able to finish the lab by downgrading both deployments to tensorflow/serving:2.8.3