BentoML

image.png

Because of the yatai process cannot connect to the k8s api-server, can you restart your yatai deployment?

So when using my own namespace, the image builder completes successfully but I don't see anything being deployed in the namespace

Seems like switching to `yatai` would work, but using some other namespace still doesn't work properly yet

Hmmm... spoke too soon:

```  Warning  Failed     3m39s (x3 over 4m19s)  kubelet            Error: ErrImagePull
  Warning  Failed     3m15s (x6 over 4m19s)  kubelet            Error: ImagePullBackOff
  Normal   BackOff    3m (x7 over 4m19s)     kubelet            Back-off pulling image "127.0.0.1:5000/bentos:yatai.ktp_ocr.uipktisj5sjyzz3e"```

You can log in to the node where the pod is located and manually pull the image with docker, suspecting that the node does not have the docker-private-registry-proxy pod started

i restarted the daemonset and it seems to be working