安装kubernetes时，Flannel启动失败

Question

我正在安装kubernetes，但是docker和kube-apiserver都无法启动。通过日记，发现原因是法兰绒的。所以我试着开始使用法兰绒，但是失败了。以下是输出信息：

Job for flanneld.service failed because a timeout was exceeded. See "systemctl status flanneld.service" and "journalctl -xe" for details.

然后我输入systemctl status flanneld.service，结果如下：

● flanneld.service - Flanneld overlay address etcd agent

Loaded: loaded (/usr/lib/systemd/system/flanneld.service; enabled; vendor preset: disabled)
Active: activating (start) since Fri 2018-01-26 03:28:08 EST; 51s ago
Main PID: 5870 (flanneld)
Memory: 6.5M

CGroup: /system.slice/flanneld.service

           └─5870 /usr/bin/flanneld -etcd-endpoints=http://127.0.0.1:2379 -etcd-prefix=/atomic.io/network


Jan 26 03:28:50 node-1 flanneld-start[5870]: E0126 03:28:50.755931    5870 network.go:102] failed to retrieve network config: 100: Key not found (/atomic.io) [23858]

有没有人能在这方面帮我？

Answer 1

卷曲已定义的端点。您可能需要将flannel指向您的主节点。

/etc/sysconfig/flanneld
FLANNEL_ETCD_ENDPOINTS="http://<master ip>:2379"
FLANNEL_ETCD_PREFIX="/kube-centos/network"

试一试，然后重启flanneld。