k8s通过virtual-kubelet部署带cinder后端动态pv的pod到openstack zun时报volume node affinity conflict
1、环境信息
作者：vxlinux2019
同 k8s+virtual-kubelet部署带动态pv的pod到openstack-zun
2、创建带cinder动态pv的pod具体创建过程参考
k8s+virtual-kubelet部署带动态pv的pod到openstack-zun
pod处于pending状态
3.1、pod状态查看pod列表
# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
testpvcpod 0/1 Pending 0 22s
3.2、pod详细信息
查看出问题的pod详细信息，问题出在调度上，调度失败的原因有两个
3.2.1、4 node(s) had taints that the pod didn’t tolerate这个问题的原因是4个master和1个普通node都有taint，本例主要验证使用两个virtual-kubelet node部署pod，所以这儿不做修改
3.2.2、2 node(s) had volume node affinity conflict该问题是本例要解决的问题
# kubectl describe pods testpvcpod
...
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning FailedScheduling default-scheduler persistentvolumeclaim "testpvc" not found
Warning FailedScheduling default-scheduler persistentvolumeclaim "testpvc" not found
Warning FailedScheduling default-scheduler 0/6 nodes are available: 2 node(s) had volume node affinity conflict, 4 node(s) had taints that the pod didn't tolerate.
Warning FailedScheduling default-scheduler 0/6 nodes are available: 2 node(s) had volume node affinity conflict, 4 node(s) had taints that the pod didn't tolerate.
3.3 查看pv和pvc信息
3.3.1、列出pv和pvc
# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
testpvc Bound pvc-c9e46b6b-cd52-474d-9835-8f5757ac5ec7 2Gi RWO standard 26m
[root@k8s-m1 test]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pvc-c9e46b6b-cd52-474d-9835-8f5757ac5ec7 2Gi RWO Delete Bound default/te stpvc standard 26m
可以看到pv和pvc信息都正常创建了
3.3.2 查看pv详细信息从pv详细信息中，pv有节点亲和的需求
Node Affinity:
Required Terms:
Term 0: failure-domain.beta.kubernetes.io/zone in
# kubectl describe pv pvc-c9e46b6b-cd52-474d-9835-8f5757ac5ec7
Name: pvc-c9e46b6b-cd52-474d-9835-8f5757ac5ec7
Labels: failure-domain.beta.kubernetes.io/zone=nova
Annotations: kubernetes.io/createdby: cinder-dynamic-provisioner
pv.kubernetes.io/bound-by-controller: yes
pv.kubernetes.io/provisioned-by: kubernetes.io/cinder
Finalizers: [kubernetes.io/pv-protection]
StorageClass: standard
Status: Bound
Claim: default/testpvc
Reclaim Policy: Delete
Access Modes: RWO
VolumeMode: Filesystem
Capacity: 2Gi
Node Affinity:
Required Terms:
Term 0: failure-domain.beta.kubernetes.io/zone in [nova]
Message:
Source:
Type: Cinder (a Persistent Disk resource in OpenStack)
VolumeID: 9ac04a59-146c-4300-9030-58c1e153c2f1
FSType:
ReadOnly: false
SecretRef: nil
Events:
4、解决问题
通过第三节分析，pv有节点亲和的需求
所以需要给节点打上相应的标签
# kubectl label node virtual-kubelet failure-domain.beta.kubernetes.io/zone=nova
node/virtual-kubelet labeled
4.2、重新部署pod
# kubectl delete -f testpvc.yaml
pod "testpvcpod" deleted
# kubectl apply -f testpvc.yaml
pod/testpvcpod created
persistentvolumeclaim/testpvc created
4.3、pod部署成功
# kubectl get pods
NAME READY STATUS RESTARTS AGE
testpvcpod 1/1 Running 0 4m29s
作者：vxlinux2019
