GKE Conjur GitHub - Application resource definition fails


I’m following the conjur-google-cloud-marketplace GitHub procedure to install Conjur on GKE:

The “Application resource definition” step fails:
kubectl apply -f marketplace-k8s-app-tools/crd/*

error: error validating “marketplace-k8s-app-tools/crd/app-crd.yaml”: error validating data: ValidationError(CustomResourceDefinition.status): missing required field “storedVersions” in io.k8s.apiextensions-apiserver.pkg.apis.apiextensions.v1beta1.CustomResourceDefinitionStatus; if you choose to ignore these errors, turn validation off with --validate=false

I can re-run with the –validate=false flag, but the deployment fails to pull any of the images:
Warning Failed 2m3s (x3 over 2m40s) kubelet, gke-lc-conjur-oss-1-default-pool-2b37a99e-d4jj Error: ErrImagePull
Normal Pulling 2m3s (x3 over 2m40s) kubelet, gke-lc-conjur-oss-1-default-pool-2b37a99e-d4jj Pulling image “gcr.io/cloud-marketplace/cyberark/conjur-open-source:2.0.0

I’d be grateful if someone could point me at what I’m doing incorrectly here. No matter what GitHub project, online guide or quickstart I follow for Conjur, I always run into an issue but in this case it may be a known problem or a pre-req I’m missing.


Hey @leighcee,
The repo you are using here is used for specifically generating the GKE marketplace tile so deploying from this repo directly is not something that will likely work especially with the GKE version changes that happen constantly. Instead, I would highly advise that you use our Helm chart for your deployment which is the basis of the GKE marketplace tile. I think the instructions for the Helm chart are pretty straightforward but please let us know if you get stuck so that we can improve them or if you find that something doesn’t work as expected.

Edit: I just realized that you were having issues on the Helm chart in the other issue so we can move our discussion there.


Hi Srdjan,

Thanks once again and you are right on the money.

I used the Helm chart as described and paid a bit more attention to the detail and finally got it working. I couldn’t login via a conjur-cli container so installed the Ruby CLI, but that’s down to the internals networking of my cluster. So all good.

This therefore solves the other issue too.