kubeadm HA master集群master重置故障恢复

文章楔子

对于一个具有HA master的集群来说，发生单点故障通常不会影响集群的正常运行，只要及时复原单点故障，就可以避免潜在的数据、状态丢失。本文旨在指导读者，在kubeadm搭建的HA master集群中，某一master主机遭遇硬件更换、系统重置、k8s配置重置的情况下，应当如何恢复K8s HA master集群。

前置需求

• 一个HA master的Kubernetes集群，搭建过程可以参考kubeadm HA集群搭建指南

单点重置恢复

故障重现

一个健康的HA master集群

首先登陆到其中一台master上，运行下面的命令以模拟单点重置的场景，随后k8s HA master陷入单点故障。

kubeadm reset -f

HA master进入单点故障

故障恢复

本章节的全过程可以在此处查看演示视频：DEMO

首先在一台健康的master上运行下面的命令获取etcd集群中故障member的ID

ETCD=`docker ps|grep etcd|grep -v POD|awk '{print $1}'`
docker exec \
  -it ${ETCD} \
  etcdctl \
  --endpoints https://127.0.0.1:2379 \
  --ca-file /etc/kubernetes/pki/etcd/ca.crt \
  --cert-file /etc/kubernetes/pki/etcd/peer.crt \
  --key-file /etc/kubernetes/pki/etcd/peer.key \
  cluster-health

10.130.29.81连接被拒绝，故障成员ID为19c5f5e4748dc98b

本例中的故障member ID为19c5f5e4748dc98b，由于故障节点已经被重置，因此相当于该ID对应的ETCD实例已经丢失，无法再取得联系。因此直接运行下面命令将故障的member从etcd集群中删除。

ETCD=`docker ps|grep etcd|grep -v POD|awk '{print $1}'`
docker exec \
  -it ${ETCD} \
  etcdctl \
  --endpoints https://127.0.0.1:2379 \
  --ca-file /etc/kubernetes/pki/etcd/ca.crt \
  --cert-file /etc/kubernetes/pki/etcd/server.crt \
  --key-file /etc/kubernetes/pki/etcd/server.key \
  member remove 19c5f5e4748dc98b

移除故障节点成功

随后将新的（重置过的）节点加入到集群中，重新组成三节点的HA master，注意重建master的过程中使用了kubeadm的配置文件，该配置文件为HA master首次部署过程中使用过的，此处直接复用该配置文件。本例中，文件内容如下，可以参考，请注意其中一条配置可能需要根据集群的现有状态进行修改：

apiVersion: kubeadm.k8s.io/v1alpha2
kind: MasterConfiguration
kubernetesVersion: v1.11.0
apiServerCertSANs:
- 10.130.29.80
- 10.130.29.81
- 10.130.29.82
- centos-7-x86-64-29-80
- centos-7-x86-64-29-81
- centos-7-x86-64-29-82
- 10.130.29.83
kubeProxy:
 config:
 mode: ipvs
etcd:
 local:
 extraArgs:
 listen-client-urls: https://127.0.0.1:2379,https://10.130.29.81:2379
 advertise-client-urls: https://10.130.29.81:2379
 listen-peer-urls: https://10.130.29.81:2380
 initial-advertise-peer-urls: https://10.130.29.81:2380
 # 注意此处需要修改，确保包括该重置节点在内的所有etcd节点的HOST=IP地址对都被列出在该配置中，不然新节点的etcd启动失败
 initial-cluster: centos-7-x86-64-29-80=https://10.130.29.80:2380,centos-7-x86-64-29-81=https://10.130.29.81:2380,centos-7-x86-64-29-82=https://10.130.29.82:2380
 initial-cluster-state: existing
 serverCertSANs:
 - centos-7-x86-64-29-81
 - 10.130.29.81
 peerCertSANs:
 - centos-7-x86-64-29-81
 - 10.130.29.81
networking:
 # This CIDR is a Calico default. Substitute or remove for your CNI provider.
 podSubnet: 172.168.0.0/16

如果读者是使用kubeadm HA集群搭建指南该教程部署的，该文件存放在各个master机器的/etc/kubernetes/kubeadm-config.yaml，否则请在下面的命令中修改所有的”/etc/kubernetes/kubeadm-config.yaml”至读者被重置的master机器上的kubeadm-config.yaml文件路径。同时，需要将host变量和ip变量设为故障节点的hostname和IP地址

ETCD=`docker ps|grep etcd|grep -v POD|awk '{print $1}'`
host=centos-7-x86-64-29-81
ip=10.130.29.81
 
bash -c """
ssh $host 'mkdir -p /etc/kubernetes/pki/etcd'
scp /etc/kubernetes/pki/ca.crt $host:/etc/kubernetes/pki/ca.crt
scp /etc/kubernetes/pki/ca.key $host:/etc/kubernetes/pki/ca.key
scp /etc/kubernetes/pki/sa.key $host:/etc/kubernetes/pki/sa.key
scp /etc/kubernetes/pki/sa.pub $host:/etc/kubernetes/pki/sa.pub
scp /etc/kubernetes/pki/front-proxy-ca.crt $host:/etc/kubernetes/pki/front-proxy-ca.crt
scp /etc/kubernetes/pki/front-proxy-ca.key $host:/etc/kubernetes/pki/front-proxy-ca.key
scp /etc/kubernetes/pki/etcd/ca.crt $host:/etc/kubernetes/pki/etcd/ca.crt
scp /etc/kubernetes/pki/etcd/ca.key $host:/etc/kubernetes/pki/etcd/ca.key
scp /etc/kubernetes/admin.conf $host:/etc/kubernetes/admin.conf
docker exec \
 -it ${ETCD} \
 etcdctl \
 --ca-file /etc/kubernetes/pki/etcd/ca.crt \
 --cert-file /etc/kubernetes/pki/etcd/peer.crt \
 --key-file /etc/kubernetes/pki/etcd/peer.key \
 --endpoints=https://127.0.0.1:2379 \
 member add $host https://$ip:2380
ssh ${host} '
 kubeadm alpha phase certs all --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase kubeconfig controller-manager --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase kubeconfig scheduler --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase kubelet config write-to-disk --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase kubelet write-env-file --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase kubeconfig kubelet --config /etc/kubernetes/kubeadm-config.yaml
 systemctl restart kubelet
 kubeadm alpha phase etcd local --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase kubeconfig all --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase controlplane all --config /etc/kubernetes/kubeadm-config.yaml
 kubeadm alpha phase mark-master --config /etc/kubernetes/kubeadm-config.yaml'
"""

被重置的master被重新加入集群

至此，HA master单点重置故障恢复完毕。