C4W3 Deployment creation taking too long

[Alexandre_Abela]

Hi,
As part of the second graded exercise of the third week I tried to create TensorFlow serving deployment of the ResNet50 model. However, when I run the command “kubectl apply -f tf-serving/deployment-resnet50.yaml” and I try to check the status of the deployment with “kubectl get deployments” it never seems to end and I got the following results :

Then the lab ends as time runs out. Could someone help me with that ?

[rishu21]

@Alexandre_Abela

kubectl get deployments

I am having the exactly same issue, am basically stuck

[chris.favila]

Hi! We are looking into this Qwiklabs issue and will update as soon as we have more info. Thank you for reporting!

[chris.favila]

Hello again! I just redid the lab and was able to get past the checkpoint. Maybe there was a temporary issue before and you just need to retry now.

Make sure though that you update the configmap file to have your model bucket name in the MODEL_PATH as mentioned in the instructions. Without that, the deployment might indeed hang. If it’s still taking more than 2 minutes to see the deployment status as READY, you can run kubectl logs deploy/image-classifier-resnet50 or kubectl logs deploy/image-classifier-resnet50 -c istio-init and see if there are any error messages that are being printed in the console.

Hope these help!

[weiqiang_mu]

I followed the instruction, but stuck with the same issue. The following is the configuration file and the deployment log

There is an error from server. Should I try again later or something wrong?

[ZaheedaT]

@chris.favila

kubectl logs deploy/image-classifier-resnet50 -c istio-init

Having the exact same issue

[ZaheedaT]

I followed the instruction, but stuck with the same issue. The following is the configuration file and the deployment log
Screen Shot 2022-07-29 at 10.43.20 AM
Screen Shot 2022-07-29 at 10.43.20 AM
2006×400 86.4 KB
There is an error from server. Should I try again later or something wrong?
Exact same issue

[ZaheedaT]

Hi there, I 've figured out the issue. What they have is

What we have in the command line is this

But on the Lab it says the name must be tfserving-configs , it should be resnet50-configs .

It will work

[chris.favila]

Thank you Zaheeda! There is indeed a mismatch between the instructions and what is currently in the repo. The metadata should not be changed and you should only modify the MODEL_PATH. We’ve now reported this to Qwiklabs for fixing. Thanks again!

[Harish_Fegade]

Hi,

I have folllowed all instructions upto deployment, but for me it is taking more time to deploy

Do I need to wait for some more time or can anyone suggest what has gone wrong here?

I can view below message in logs