minikube cluster creation error

[mz-pdm]

I'm trying to run a Jumpstarter minikube cluster, following https://jumpstarter.dev/main/getting-started/installation/service/service-local.html#create-a-local-cluster-and-install-jumpstarter. When I run

jmp admin install --minikube --create-cluster

the cluster gets created but I get the following error:

Release "jumpstarter" does not exist. Installing it now.
Pulled: quay.io/jumpstarter-dev/helm/jumpstarter:0.7.0-dev-8-g83e23d3
Digest: sha256:b73c6dafc2456e12ba9c13324a085d4e5020b03d38cf58ccad3650f2a330d77c
Error: values don't meet the specifications of the schema(s) in the following chart(s):
jumpstarter-controller:
- grpc: Must validate at least one schema (anyOf)
- grpc.mode: Must validate at least one schema (anyOf)
- grpc.mode: grpc.mode must be one of the following: "ingress", "route"

Installed Helm release "jumpstarter" in namespace "jumpstarter-lab"

minikube version: v1.36.0

[bennyz]

@kirkbrauer are we missing nodeport in https://github.com/jumpstarter-dev/jumpstarter-controller/blob/a9298a5a9d7f77e89d1815b515a7e801c78fb74c/deploy/helm/jumpstarter/charts/jumpstarter-controller/model.py#L59

[bennyz]

posted jumpstarter-dev/jumpstarter-controller#156

[bennyz]

you can try it with

$ jmp admin install --minikube --create-cluster --version 0.7.0-dev-18-gf8cd704

[mz-pdm]

It fixes the problem, thank you.

However, it still doesn't start:

Release "jumpstarter" does not exist. Installing it now.
Pulled: quay.io/jumpstarter-dev/helm/jumpstarter:0.7.0-dev-18-gf8cd704
Digest: sha256:4d61d6086efcd0fad144b4bc874495d49ac99acda9bb2a755ed26f93ea26f519
Error: context deadline exceeded
Installed Helm release "jumpstarter" in namespace "jumpstarter-lab"

One of the pods doesn't start:

% kubectl get pods -n jumpstarter-lab        
NAME                                      READY   STATUS    RESTARTS   AGE
jumpstarter-controller-786ccdfcfb-7jjcn   0/1     Pending   0          9m7s
jumpstarter-router-898dd7854-dsdsc        1/1     Running   0          9m7s

[bennyz]

Can you share the output of:

kubectl -n jumpstarter-lab describe pods/jumpstarter-controller-786ccdfcfb-7jjcn

[mz-pdm]

Seems to be a hardware/VM resource problem (the pod name is different since I reinstalled the cluster in the meantime):

% kubectl -n jumpstarter-lab describe pods/jumpstarter-controller-55b47995c9-gcwh8
Name:             jumpstarter-controller-55b47995c9-gcwh8
Namespace:        jumpstarter-lab
Priority:         0
Service Account:  controller-manager
Node:             <none>
Labels:           control-plane=controller-manager
                  pod-template-hash=55b47995c9
Annotations:      kubectl.kubernetes.io/default-container: manager
Status:           Pending
SeccompProfile:   RuntimeDefault
IP:               
IPs:              <none>
Controlled By:    ReplicaSet/jumpstarter-controller-55b47995c9
Containers:
  manager:
    Image:      quay.io/jumpstarter-dev/jumpstarter-controller:0.7.0-dev-18-gf8cd704
    Port:       <none>
    Host Port:  <none>
    Args:
      --leader-elect
      --health-probe-bind-address=:8081
      -metrics-bind-address=:8080
    Limits:
      cpu:     2
      memory:  1Gi
    Requests:
      cpu:      1
      memory:   256Mi
    Liveness:   http-get http://:8081/healthz delay=15s timeout=1s period=20s #success=1 #failure=3
    Readiness:  http-get http://:8081/readyz delay=5s timeout=1s period=10s #success=1 #failure=3
    Environment:
      GRPC_ENDPOINT:   grpc.jumpstarter.192.168.39.186.nip.io:8082
      CONTROLLER_KEY:  <set to the key 'key' in secret 'jumpstarter-controller-secret'>  Optional: false
      ROUTER_KEY:      <set to the key 'key' in secret 'jumpstarter-router-secret'>      Optional: false
      NAMESPACE:       jumpstarter-lab (v1:metadata.namespace)
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-xgkth (ro)
Conditions:
  Type           Status
  PodScheduled   False 
Volumes:
  kube-api-access-xgkth:
    Type:                    Projected (a volume that contains injected data from multiple sources)
    TokenExpirationSeconds:  3607
    ConfigMapName:           kube-root-ca.crt
    Optional:                false
    DownwardAPI:             true
QoS Class:                   Burstable
Node-Selectors:              <none>
Tolerations:                 node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                             node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
  Type     Reason            Age                  From               Message
  ----     ------            ----                 ----               -------
  Warning  FailedScheduling  3m26s (x6 over 23m)  default-scheduler  0/1 nodes are available: 1 Insufficient cpu. preemption: 0/1 nodes are available: 1 No preemption victims found for incoming pod.

[bennyz]

the following has worked for me

jmp admin install --minikube --create-cluster --force-recreate-cluster --version 0.7.0-dev-18-gf8cd704 --minikube-extra-args "--cpus 4 --driver=podman"

[mz-pdm]

It works for me too, thank you. Would it be worth to mention in the documentation the need to increase the number of CPUs?

[bennyz]

we can also set this as the default if the default number of CPUs is not enough

[mz-pdm]

That would be even better.