Couchbase Admission Controller not creating cluster

Using helm autonomous operator, I am running the command from the official docs to create all the resources, but it does not create clusters. If I turn off the admission controller, then clusters are created. I have paste the error I get in couchbase-operator when creating the cluster with admission controller enabled.

{"level":"error","ts":1621610869.3155746,"logger":"cluster","msg":"Failed to update members","cluster":"de │
│ fault/couchbase-couchbase-cluster","error":"unexpected status code: request failed GET http://couchbase-couchbase-cluster-0000.couchbase-couchbase-clust │
│ er.default.svc:8091/pools/default 404 Object Not Found: \"unknown pool\"","stacktrace":"*zapLogger).Error\n\t/home/couchbase/go │
│ /pkg/mod/\*Cluster).runReconcile\n\t/home/couchbase/jenk │
│ ins/workspace/couchbase-k8s-microservice-build/couchbase-operator/pkg/cluster/cluster.go:390\*Clus │
│ ter).Update\n\t/home/couchbase/jenkins/workspace/couchbase-k8s-microservice-build/couchbase-operator/pkg/cluster/cluster.go:429\ │
│ uchbase-operator/pkg/controller.(*CouchbaseClusterReconciler).Reconcile\n\t/home/couchbase/jenkins/workspace/couchbase-k8s-microservice-build/couchbase- │
│ operator/pkg/controller/controller.go:90\*Controller).reconcileHandler\n\t/home/couchbase/go/pk │
│ g/mod/\*Controll │
│ er).processNextWorkItem\n\t/home/couchbase/go/pkg/mod/\ │
│ oller-runtime/pkg/internal/controller.(*Controller).worker\n\t/home/couchbase/go/pkg/mod/ │
│ ontroller.go:211\\n\t/home/couchbase/go/pkg/mod/ │
│ t.go:152\\n\t/home/couchbase/go/pkg/mod/\nk8s. │
│ io/apimachinery/pkg/util/wait.Until\n\t/home/couchbase/go/pkg/mod/"}

Hi @Vishvajit_Kher , it’s not clear to me what’s happening from this snippet, can you share cbopinfo of the cluster:

Are you running helm with any overrides?

Hi @tommie, how do I run this command when I have deployed it to Kubernetes cluster via helm chart? No I am not using any overrides. I just ran the command helm install couchbase couchbase/couchbase-operator

I might be wrong, but I think when admission controller is turned on, cluster is not setup so when the server nodes tries to start, they cannot reach port 8091

No, the DAC being on should not affect whether you get a cluster or not.

Whilst testing the upcoming release I found various issues with older kubernetes versions and how defaults are being handled. Can you confirm your kubernetes version used and the version of the helm chart?