## 3.Kubelet启动时Failed to start ContainerManager systemd version does not support ability to start a slice as transient unit
CentOS系统版本7.2.1511
kubelet启动时报错systemd版本不支持start a slice as transient unit。
尝试升级CentOS版本到7.3,看看是否可以修复该问题。
与[kubeadm init waiting for the control plane to become ready on CentOS 7.2 with kubeadm 1.6.1 #228](https://github.com/kubernetes/kubeadm/issues/228)类似。
## 4.kube-proxy报错kube-proxy[2241]: E0502 15:55:13.889842 2241 conntrack.go:42] conntrack returned error: error looking for path of conntrack: exec: "conntrack": executable file not found in $PATH
## 5. [Pod stucks in terminating if it has a privileged container but has been scheduled to a node which doesn't allow privilege issue#42568](https://github.com/kubernetes/kubernetes/issues/42568)
Error validating pod kube-keepalived-vip-1p62d_default(5d79ccc0-3173-11e7-bfbd-8af1e3a7c5bd) from api, ignoring: spec.containers[0].securityContext.privileged: Forbidden: disallowed by cluster policy
kubelet 的配置文件 `/etc/kubernetes/kubelet` 中的配置中将集群 DNS 的 domain name 配置成了 `––cluster-domain=cluster.local.` ,虽然对于 service 的名字能够正常的完成 DNS 解析,但是对于 headless service 中的 pod 名字解析不了,查看 pod 的 `/etc/resolv.conf` 文件可以看到以下内容:
1h 12s 441 {persistentvolume-controller } Warning ProvisioningFailed Failed to provision volume with StorageClass "ceph-web": failed to create rbd image: executable file not found in $PATH, command output:
- [Helm: Error: no available release name found - StackOverflow](https://stackoverflow.com/questions/43499971/helm-error-no-available-release-name-found)
- [Helm 2.2.3 not working properly with kubeadm 1.6.1 default RBAC rules #2224](https://github.com/kubernetes/helm/issues/2224)