I'm trying to install Redis cluster (StatefulSet) out of GKE and when getting pvc I've got
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning ProvisioningFailed 10s persistentvolume-controller Failed to provision volume with StorageClass "slow": Failed to get GCE GCECloudProvider with error <nil>
Already added "--cloud-provider=gce" on files /etc/kubernetes/manifests/kube-controller-manager.yaml and sudo vi /etc/kubernetes/manifests/kube-apiserver.yaml. Restarted but still the same. Can anyone help me please? What's the trick for making k8s work on GCP?
My manifest taken from here:
---
apiVersion: v1
kind: ConfigMap
metadata:
name: redis-cluster
labels:
app: redis-cluster
data:
fix-ip.sh: |
#!/bin/sh
CLUSTER_CONFIG="/data/nodes.conf"
if [ -f ${CLUSTER_CONFIG} ]; then
if [ -z "${POD_IP}" ]; then
echo "Unable to determine Pod IP address!"
exit 1
fi
echo "Updating my IP to ${POD_IP} in ${CLUSTER_CONFIG}"
sed -i.bak -e "/myself/ s/[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}/${POD_IP}/" ${CLUSTER_CONFIG}
fi
exec "$@"
redis.conf: |+
cluster-enabled yes
cluster-require-full-coverage no
cluster-node-timeout 15000
cluster-config-file /data/nodes.conf
cluster-migration-barrier 1
appendonly yes
protected-mode no
---
kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
name: slow
provisioner: kubernetes.io/gce-pd
parameters:
type: pd-standard
replication-type: none
zone: "us-west2-a"
reclaimPolicy: Retain
---
apiVersion: v1
kind: Service
metadata:
name: redis-cluster
labels:
app: redis-cluster
spec:
ports:
- port: 6379
targetPort: 6379
name: client
- port: 16379
targetPort: 16379
name: gossip
clusterIP: None
selector:
app: redis-cluster
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: redis-cluster
labels:
app: redis-cluster
spec:
serviceName: redis-cluster
replicas: 5
selector:
matchLabels:
app: redis-cluster
template:
metadata:
labels:
app: redis-cluster
spec:
containers:
- name: redis
image: redis:5.0-rc
ports:
- containerPort: 6379
name: client
- containerPort: 16379
name: gossip
command: ["/conf/fix-ip.sh", "redis-server", "/conf/redis.conf"]
args:
- --cluster-announce-ip
- "$(POD_IP)"
readinessProbe:
exec:
command:
- sh
- -c
- "redis-cli -h $(hostname) ping"
initialDelaySeconds: 15
timeoutSeconds: 5
livenessProbe:
exec:
command:
- sh
- -c
- "redis-cli -h $(hostname) ping"
initialDelaySeconds: 20
periodSeconds: 3
env:
- name: POD_IP
valueFrom:
fieldRef:
fieldPath: status.podIP
volumeMounts:
- name: conf
mountPath: /conf
readOnly: false
- name: data
mountPath: /data
readOnly: false
volumes:
- name: conf
configMap:
name: redis-cluster
defaultMode: 0755
volumeClaimTemplates:
- metadata:
name: data
labels:
name: redis-cluster
spec:
accessModes: [ "ReadWriteOnce" ]
storageClassName: slow
resources:
requests:
storage: 5Gi
Google offers two main types of persistent disk, which are provisioned automatically on kubernetes:
Standard storage (labeled pd-standard)
SSD storage (labeled pd-ssd)
By default, GKE will provision standard storage persistent disks. In fact, that’s the only storage class even available at first.
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: test-storageclass
provisioner: kubernetes.io/gce-pd
parameters:
type: pd-ssd
you can tell a persistent volume to use the new ssd storage class with the following key/value pair: storageClassName: ssd.
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: ssd-storageclass
spec:
accessModes:
- ReadWriteOnce
storageClassName: ssd
resources:
requests:
storage: 1Gi
Please verify your "StorageClass: slow", it seems there is an indentation problem (starting with reclaimPolicy)
kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
name: slow
provisioner: kubernetes.io/gce-pd
parameters:
type: pd-standard
replication-type: none
zone: "us-west2-a"
reclaimPolicy: Retain
# Update:
Please add --cloud-provider=gce
into: kube-apiserver.yaml, kube-controller-manager.yaml, KUBELET_KUBECONFIG_ARGS. You can also enable enable-admission-plugins=DefaultStorageClass
Verify in your "VM instance details" permissiosn in "Cloud API access scopes" permissions.
Verify if your storage class pv and pvc are working properly.
kind: StorageClass
apiVersion: storage.k8s.io/v1beta1
metadata:
name: slow
annotations:
storageclass.beta.kubernetes.io/is-default-class: "true"
provisioner: kubernetes.io/gce-pd
parameters:
type: pd-standard
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: pvc-test
spec:
accessModes:
- ReadOnlyMany
storageClassName: slow
resources:
requests:
storage: 1Gi