记录一次ETCD集群故障恢复

k8s集群部署的etcd集群（3节点）由于集群扩容，宿主机重启，导致一个节点起不来。

{"level":"fatal","ts":"2025-07-29T06:58:31.624750Z","caller":"etcdmain/etcd.go:183","msg":"discovery failed","error":"member da5cb2321ac7cbbd has already been bootstrapped","stacktrace":"go.etcd.io/etcd/server/v3/etcdmain.startEtcdOrProxyV2\n\tgo.etcd.io/etcd/server/v3/etcdmain/etcd.go:183\ngo.etcd.io/etcd/server/v3/etcdmain.Main\n\tgo.etcd.io/etcd/server/v3/etcdmain/main.go:40\nmain.main\n\tgo.etcd.io/etcd/server/v3/main.go:31\nruntime.main\n\truntime/proc.go:272"}

错误信息如上，其实是由于节点部署时 inital_cluster_state是new，导致集群节点识别已存在，拒绝再次初始化。

处理方式，

先改成existing。然后用etcd可用节点登陆集群，移除故障节点。

然后重新添加原节点。

节点成功加入集群，并启动成功。

过程的命令不再记录了，问豆包吧。

如果不用命令移除节点，也会报错，错误如下：

panic: tocommit(35) is out of range [lastIndex(0)]. Was the raft log corrupted, truncated, or lost?

以上

加一下吧

etcdctl member list 查看节点信息

etcdctl member remove 节点ID 移除节点

etcdctl member add 节点名称 --peer-urls=https://10.0.0.3:2380