Unable to run v0.2.0 in minikube. "pod has unbound PersistentVolumeClaims"

[kohv]

Running helm install --name consul --namespace=pkr -f dev-consul.yaml ./consul-helm with these custom values:

syncCatalog:
  enabled: true
server:
  storage: "8Gi"

It completes successfully, but when running kubectl get pods -n pkr, I see that consul-server-1 & consul-server-2 are Pending.

NAME                                   READY     STATUS    RESTARTS   AGE
consul-4vstb                           0/1       Running   0          15m
consul-server-0                        0/1       Running   0          15m
consul-server-1                        0/1       Pending   0          15m
consul-server-2                        0/1       Pending   0          15m
consul-sync-catalog-587b6859f6-dpj5v   1/1       Running   0          15m

Closer inspection on pods consul-server-0:

kubectl describe pods consul-server-0 -n pkr
Name:           consul-server-0
Namespace:      pkr
Node:           minikube/10.0.2.15
Start Time:     Sat, 29 Sep 2018 22:39:09 +0300
Labels:         app=consul
                chart=consul-0.1.0
                component=server
                controller-revision-hash=consul-server-66479c5df5
                hasDNS=true
                release=consul
                statefulset.kubernetes.io/pod-name=consul-server-0
Annotations:    consul.hashicorp.com/connect-inject=false
Status:         Running
IP:             172.17.0.9
Controlled By:  StatefulSet/consul-server
Containers:
  consul:
    Container ID:  docker://7de44c6027bdb78ba4b7bc73643701aa9e0bbb55abce8ce2c7b8e12e2adf82b0
    Image:         consul:1.2.3
    Image ID:      docker-pullable://consul@sha256:ea66d17d8c8c1f1afb2138528d62a917093fcd2e3b3a7b216a52c253189ea980
    Ports:         8500/TCP, 8301/TCP, 8302/TCP, 8300/TCP, 8600/TCP, 8600/UDP
    Host Ports:    0/TCP, 0/TCP, 0/TCP, 0/TCP, 0/TCP, 0/UDP
    Command:
      /bin/sh
      -ec
      CONSUL_FULLNAME="consul"

exec /bin/consul agent \
  -advertise="${POD_IP}" \
  -bind=0.0.0.0 \
  -bootstrap-expect=3 \
  -client=0.0.0.0 \
  -config-dir=/consul/config \
  -datacenter=dc1 \
  -data-dir=/consul/data \
  -domain=consul \
  -hcl="connect { enabled = true }" \
  -ui \
  -retry-join=${CONSUL_FULLNAME}-server-0.${CONSUL_FULLNAME}-server.${NAMESPACE}.svc \
  -retry-join=${CONSUL_FULLNAME}-server-1.${CONSUL_FULLNAME}-server.${NAMESPACE}.svc \
  -retry-join=${CONSUL_FULLNAME}-server-2.${CONSUL_FULLNAME}-server.${NAMESPACE}.svc \
  -server

    State:          Running
      Started:      Sat, 29 Sep 2018 22:39:10 +0300
    Ready:          False
    Restart Count:  0
    Readiness:      exec [/bin/sh -ec curl http://127.0.0.1:8500/v1/status/leader 2>/dev/null | \
grep -E '".+"'
] delay=5s timeout=5s period=3s #success=1 #failure=2
    Environment:
      POD_IP:      (v1:status.podIP)
      NAMESPACE:  pkr (v1:metadata.namespace)
    Mounts:
      /consul/config from config (rw)
      /consul/data from data (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-fd6r9 (ro)
Conditions:
  Type           Status
  Initialized    True
  Ready          False
  PodScheduled   True
Volumes:
  data:
    Type:       PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
    ClaimName:  data-consul-server-0
    ReadOnly:   false
  config:
    Type:      ConfigMap (a volume populated by a ConfigMap)
    Name:      consul-server-config
    Optional:  false
  default-token-fd6r9:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-fd6r9
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type     Reason                 Age                  From               Message
  ----     ------                 ----                 ----               -------
  Warning  FailedScheduling       31m (x2 over 31m)    default-scheduler  pod has unbound PersistentVolumeClaims
  Normal   Scheduled              31m                  default-scheduler  Successfully assigned consul-server-0 to minikube
  Normal   SuccessfulMountVolume  31m                  kubelet, minikube  MountVolume.SetUp succeeded for volume "pvc-55188bbe-c41f-11e8-b65d-080027750557"
  Normal   SuccessfulMountVolume  31m                  kubelet, minikube  MountVolume.SetUp succeeded for volume "config"
  Normal   SuccessfulMountVolume  31m                  kubelet, minikube  MountVolume.SetUp succeeded for volume "default-token-fd6r9"
  Normal   Pulled                 31m                  kubelet, minikube  Container image "consul:1.2.3" already present on machine
  Normal   Created                31m                  kubelet, minikube  Created container
  Normal   Started                31m                  kubelet, minikube  Started container
  Warning  Unhealthy              16m (x299 over 30m)  kubelet, minikube  Readiness probe failed:

This "pod has unbound PersistentVolumeClaims" error is same for all consul servers. Yet, when running kubectl get pvc & kubectl get pv, I see persistent volumes fine:

kubectl get pv
NAME                                       CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS    CLAIM                           STORAGECLASS   REASON    AGE
pvc-55188bbe-c41f-11e8-b65d-080027750557   8Gi        RWO            Delete           Bound     pkr/data-consul-server-0   standard                 37m
pvc-55245df6-c41f-11e8-b65d-080027750557   8Gi        RWO            Delete           Bound     pkr/data-consul-server-1   standard                 37m
pvc-5533ebc2-c41f-11e8-b65d-080027750557   8Gi        RWO            Delete           Bound     pkr/data-consul-server-2   standard                 37m

kubectl get pvc -n pkr
NAME                   STATUS    VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
data-consul-server-0   Bound     pvc-55188bbe-c41f-11e8-b65d-080027750557   8Gi        RWO            standard       38m
data-consul-server-1   Bound     pvc-55245df6-c41f-11e8-b65d-080027750557   8Gi        RWO            standard       38m
data-consul-server-2   Bound     pvc-5533ebc2-c41f-11e8-b65d-080027750557   8Gi        RWO            standard       38m
postgresql             Bound     pvc-56367c46-c41f-11e8-b65d-080027750557   8Gi        RWO            standard       38m

So, I don't understand the error, since pv & pvc outputs seem fine to me.
How should I debug this further? I've tried deleting the minikube cluster and starting over from scratch, but I get this result every time.

kubectl version

Client Version: version.Info{Major:"1", Minor:"11", GitVersion:"v1.11.3", GitCommit:"a4529464e4629c21224b3d52edfe0ea91b072862", GitTreeState:"clean", BuildDate:"2018-09-10T11:44:36Z", GoVersion:"go1.11", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"10", GitVersion:"v1.10.0", GitCommit:"fc32d2f3698e36b93322a3465f63a14e9f0eaead", GitTreeState:"clean", BuildDate:"2018-03-26T16:44:10Z", GoVersion:"go1.9.3", Compiler:"gc", Platform:"linux/amd64"}

helm version
Client: &version.Version{SemVer:"v2.11.0", GitCommit:"2e55dbe1fdb5fdb96b75ff144a339489417b146b", GitTreeState:"clean"}
Server: &version.Version{SemVer:"v2.11.0", GitCommit:"2e55dbe1fdb5fdb96b75ff144a339489417b146b", GitTreeState:"clean"}```

[kohv]

I had to remove affinity rules on the StatefulSet, here

consul-helm/templates/server-statefulset.yaml

Line 40 in 100d14e

affinity:

After removing that anti-affinity rule, I was able to successfully launch consul server & consul-ui. I'm very new to k8s, but it seems to me that the anti-affinity rules are conflicting when running multiple consul server agents?

[ervikrant06]

Yes, I also hit the same issue this is happening because of affinity rules only. for redundancy purpose they want to start the single consul server on each node.

[mitchellh]

This seems like two things:

• We need docs to state you'll need at least N distinct nodes available where N is the number of Consul servers.
• OR: you need to get rid of the affinity rules. We need to add configuration to allow this.

[s3than]

We resolved this issue by making the affinity a variable

    {{- if .Values.affinity }}
      affinity:
{{ tpl .Values.affinity . | indent 8 }}
{{- end }}

And then in the values

### Consul Settings
affinity: |
  podAntiAffinity:
    preferredDuringSchedulingIgnoredDuringExecution:
      - labelSelector:
          matchLabels:
            app: {{ template "consul.name" . }}
            release: "{{ .Release.Name }}"
            component: server
        topologyKey: kubernetes.io/hostname

This then means in minikube you can override with

## Affinity settings, this allows us to configure and run in Minikube
affinity: |
  podAntiAffinity:
    preferredDuringSchedulingIgnoredDuringExecution:
    - weight: 1
      podAffinityTerm:
        topologyKey: kubernetes.io/hostname
        labelSelector:
          matchExpressions:
          - key: component
            operator: In
            values:
            - "{{ .Release.Name }}-{{ .Values.Component }}"

[s3than]

@mitchellh Is there anything I can provide to get this item resolved?

Making affinity a variable allows testing on Minikube

Thanks

[adilyse]

Thanks to @s3than, server affinity is now configurable in the values file! Additional documentation will be added once this PR in the Consul repository is merged.