Styjun

I tried to use helm on docker for windows on the local machine. When I used a storage class as local storage, persistent volume, and persistent volume claim without helm, it works fine. But when I used this setting with helm, CrashLoopBackOff happened.

localStrageClass.yaml

kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
 name: local-storage
provisioner: kubernetes.io/no-provisioner
volumeBindingMode: WaitForFirstConsumer

pv.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
 name: local-pv002
 labels:
 type: local
spec:
 capacity:
 storage: 1Gi
 accessModes:
 - ReadWriteMany
 #storageClassName: hostpath
 mountOptions:
 - hard
 persistentVolumeReclaimPolicy: Delete
 storageClassName: local-storage
 local:
 path: /c/k/share/mysql
 nodeAffinity:
 required:
 nodeSelectorTerms:
 - matchExpressions:
 #- key: docker.io/hostname
 - key: kubernetes.io/hostname
 operator: In
 values:
 - docker-desktop

pvc.yaml

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
 name: local-mysql-claim
spec:
 accessModes:
 - ReadWriteMany
 resources:
 requests:
 storage: 1Gi
 storageClassName: local-storage

mysqlConf.yaml

persistence:
 enabled: true
 storageClass: local-storage
 existingClaim: local-mysql-claim
 accessMode: ReadWriteOnce
 size: 1Gi
 annotations: 

$ helm install --name mysql stable/mysql -f mysqlConf.yaml

$ kubectl describe pod mysql

Containers:
 mysql:
 Container ID: docker://533e4569603b05fac83a0a701da97898b3190503618796678ac5db6340c4dce6
 Image: mysql:5.7.14
 Image ID: docker-pullable://mysql@sha256:c8f03238ca1783d25af320877f063a36dbfce0daa56a7b4955e6c6e05ab5c70b
 Port: 3306/TCP
 Host Port: 0/TCP
 State: Waiting
 Reason: CrashLoopBackOff
 Last State: Terminated
 Reason: Error
 Exit Code: 1
 Started: Thu, 28 Mar 2019 13:24:25 +0900
 Finished: Thu, 28 Mar 2019 13:24:25 +0900
 Ready: False
 Restart Count: 2
 Requests:
 cpu: 100m
 memory: 256Mi
 Liveness: exec [sh -c mysqladmin ping -u root -p$MYSQL_ROOT_PASSWORD] delay=30s timeout=5s period=10s #success=1 #failure=3
 Readiness: exec [sh -c mysqladmin ping -u root -p$MYSQL_ROOT_PASSWORD] delay=5s timeout=1s period=10s #success=1 #failure=3

 Environment:
 MYSQL_ROOT_PASSWORD: <set to the key 'mysql-root-password' in secret 'mysql'> Optional: false
 MYSQL_PASSWORD: <set to the key 'mysql-password' in secret 'mysql'> Optional: true
 MYSQL_USER:
 MYSQL_DATABASE:
 Mounts:
 /var/lib/mysql from data (rw)
 /var/run/secrets/kubernetes.io/serviceaccount from default-token-dccpv (ro)
Conditions:
 Type Status
 Initialized True
 Ready False
 ContainersReady False
 PodScheduled True
Volumes:
 data:
 Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
 ClaimName: local-mysql-claim
 ReadOnly: false
 default-token-dccpv:
 Type: Secret (a volume populated by a Secret)
 SecretName: default-token-dccpv
 Optional: false
QoS Class: Burstable
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
 ---- ------ ---- ---- -------
 Normal Scheduled 39s default-scheduler Successfully assigned default/mysql-698897ff79-n768k to docker-desktop
 Normal Pulled 38s kubelet, docker-desktop Container image "busybox:1.29.3" already present on machine
 Normal Created 38s kubelet, docker-desktop Created container
 Normal Started 38s kubelet, docker-desktop Started container
 Normal Pulled 18s (x3 over 37s) kubelet, docker-desktop Container image "mysql:5.7.14" already present on machine
 Normal Created 17s (x3 over 37s) kubelet, docker-desktop Created container
 Normal Started 17s (x3 over 37s) kubelet, docker-desktop Started container
 Warning BackOff 13s (x5 over 35s) kubelet, docker-desktop Back-off restarting failed container

When storageClassName was hostpath or did not used the configuration file as

$ helm install --name mysql stable/mysql

it worked fine.

Please tell me how to fix this problem.

I think you are having a mismatch of accessModes between what you claim in PVC definition (ReadWriteOnce) and what your Storage Class offers (ReadWriteMany).

Please mind also that PersistentVolume(s) of HostPath type does not support ReadWriteMany mode (see spec here).

I would propose you to create PV similar to this one:

# Create PV of manual StorageClass
kind: PersistentVolume
apiVersion: v1
metadata:
 name: task-pv-volume
 labels:
 type: local
spec:
 storageClassName: manual
 capacity:
 storage: 10Gi
 accessModes:
 - ReadWriteOnce
 hostPath:
 path: "/C/Users/K8S/mysql" 

and override default PVC storageClassName configuration during helm install like this:

helm install --name my-sql stable/mysql --set persistence.storageClass=manual