当前位置：首页 > news >正文

K8s 集群高可用master节点ETCD挂掉如何恢复?

news 2025/11/2 16:42:57

写在前面

很常见的集群运维场景，整理分享
博文内容为 K8s 集群高可用 master 节点故障如何恢复的过程
理解不足小伙伴帮忙指正

不必太纠结于当下，也不必太忧虑未来，当你经历过一些事情的时候，眼前的风景已经和从前不一样了。——村上春树

遇到了什么问题

今天做实验发现，集群其中一个 master 节点上的 etcd 和 apiserver 都挂掉了

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get nodes
NAME                          STATUS   ROLES           AGE    VERSION
vms100.liruilongs.github.io   Ready    control-plane   415d   v1.25.1
vms101.liruilongs.github.io   Ready    control-plane   415d   v1.25.1
vms102.liruilongs.github.io   Ready    control-plane   415d   v1.25.1
vms103.liruilongs.github.io   Ready    <none>          415d   v1.25.1
vms105.liruilongs.github.io   Ready    <none>          415d   v1.25.1
vms106.liruilongs.github.io   Ready    <none>          415d   v1.25.1
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

vms100.liruilongs.github.io 这个节点上的 apiserver 和 etcd

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep apiserver
kube-system          kube-apiserver-vms100.liruilongs.github.io            0/1     CrashLoopBackOff   1448 (3m23s ago)   415d   192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms101.liruilongs.github.io            1/1     Running            272 (3h18m ago)    415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms102.liruilongs.github.io            1/1     Running            246 (3h18m ago)    415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep etcd
kube-system          etcd-vms100.liruilongs.github.io                      0/1     CrashLoopBackOff   1244 (3m6s ago)    415d   192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms101.liruilongs.github.io                      1/1     Running            167 (3h18m ago)    415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms102.liruilongs.github.io                      1/1     Running            173 (3h18m ago)    415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>

查看 keepalived 对应的静态Pod运行正常

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep keep
kube-system          keepalived-vms100.liruilongs.github.io                1/1     Running            63 (3h50m ago)    415d   192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          keepalived-vms101.liruilongs.github.io                1/1     Running            54 (3h51m ago)    415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          keepalived-vms102.liruilongs.github.io                1/1     Running            60 (3h51m ago)    415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

所以可能是 etcd 数据不同步，或者什么原因导致etcd 挂掉了。因为每个 master 节点的 apiserver 只和本节点的 etcd 进行通信(每个 etcd 的写请求会转发到 etcd 的领导节点)，etcd 挂掉，apiserver 无法提供能力，所以也会挂掉。

通过 etcdctl 可以发现 vms100.liruilongs.github.io 上的 etcd 彻底死掉了

┌──[root@vms100.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379 \--cert="/etc/kubernetes/pki/etcd/server.crt" \--key="/etc/kubernetes/pki/etcd/server.key"  \--cacert="/etc/kubernetes/pki/etcd/ca.crt" \member list -w table
Error: dial tcp 127.0.0.1:2379: connect: connection refused

如何排查

这里我们换一个 etcd 节点执行命令

查看 etcd 集群成员

┌──[root@vms100.liruilongs.github.io]-[~]
└─$ssh vms101.liruilongs.github.io
Last login: Sat Mar  2 09:52:01 2024 from 192.168.26.100
┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379 \--cert="/etc/kubernetes/pki/etcd/server.crt"  \--key="/etc/kubernetes/pki/etcd/server.key" \--cacert="/etc/kubernetes/pki/etcd/ca.crt" \member list -w table
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
|        ID        | STATUS  |            NAME             |         PEER ADDRS          |        CLIENT ADDRS         |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
|  ee392e5273e89e2 | started | vms100.liruilongs.github.io | https://192.168.26.100:2380 | https://192.168.26.100:2379 |
| 70059e836d19883d | started | vms101.liruilongs.github.io | https://192.168.26.101:2380 | https://192.168.26.101:2379 |
| b8cb9f66c2e63b91 | started | vms102.liruilongs.github.io | https://192.168.26.102:2380 | https://192.168.26.102:2379 |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+

查看节点状态

┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379 \--cert="/etc/kubernetes/pki/etcd/server.crt" \--key="/etc/kubernetes/pki/etcd/server.key"  \--cacert="/etc/kubernetes/pki/etcd/ca.crt" \endpoint status --cluster  -w table
Failed to get the status of endpoint https://192.168.26.100:2379 (context deadline exceeded)
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
|          ENDPOINT           |        ID        | VERSION | DB SIZE | IS LEADER | RAFT TERM | RAFT INDEX |
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
| https://192.168.26.101:2379 | 70059e836d19883d |   3.5.4 |   88 MB |     false |       603 |   22208417 |
| https://192.168.26.102:2379 | b8cb9f66c2e63b91 |   3.5.4 |   88 MB |      true |       603 |   22208417 |
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+

确定 ETCD 节点故障

┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  \--cert="/etc/kubernetes/pki/etcd/server.crt"  \--key="/etc/kubernetes/pki/etcd/server.key"  \--cacert="/etc/kubernetes/pki/etcd/ca.crt" \endpoint  health --cluster  -w table
https://192.168.26.101:2379 is healthy: successfully committed proposal: took = 3.753357ms
https://192.168.26.102:2379 is healthy: successfully committed proposal: took = 2.989943ms
https://192.168.26.100:2379 is unhealthy: failed to connect: dial tcp 192.168.26.100:2379: connect: connection refused
Error: unhealthy cluster

查看 etcd 的容器日志

┌──[root@vms100.liruilongs.github.io]-[~]
└─$docker ps -a | grep etcd
0f2f98ebf8c3   a8a176a5d5d6                                        "etcd --advertise-cl…"   4 minutes ago   Exited (2) 4 minutes ago                  k8s_etcd_etcd-vms100.liruilongs.github.io_kube-system_e8c17bb99f9bd8119cdd769556041e18_1252
a4b39d16a753   registry.aliyuncs.com/google_containers/pause:3.8   "/pause"                 4 hours ago     Up 4 hours                                k8s_POD_etcd-vms100.liruilongs.github.io_kube-system_e8c17bb99f9bd8119cdd769556041e18_54
┌──[root@vms100.liruilongs.github.io]-[~]
└─$docker logs 0f2f98ebf8c3
{"level":"info","ts":"2024-03-16T14:46:54.644Z","caller":"etcdmain/etcd.go:73","msg":"Running: ","args":["etcd","--advertise-client-urls=https://192.168.26.100:2379","--cert-file=/etc/kubernetes/pki/etcd/server.crt","--client-cert-auth=true","--data-dir=/var/lib/etcd","--experimental-initial-corrupt-check=true","--experimental-watch-progress-notify-interval=5s","--initial-advertise-peer-urls=https://192.168.26.100:2380","--initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380","--key-file=/etc/kubernetes/pki/etcd/server.key","--listen-client-urls=https://127.0.0.1:2379,https://192.168.26.100:2379","--listen-metrics-urls=http://127.0.0.1:2381","--listen-peer-urls=https://192.168.26.100:2380","--name=vms100.liruilongs.github.io","--peer-cert-file=/etc/kubernetes/pki/etcd/peer.crt","--peer-client-cert-auth=true","--peer-key-file=/etc/kubernetes/pki/etcd/peer.key","--peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt","--snapshot-count=10000","--trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt"]}
{"level":"info","ts":"2024-03-16T14:46:54.645Z","caller":"etcdmain/etcd.go:116","msg":"server has been already initialized","data-dir":"/var/lib/etcd","dir-type":"member"}
{"level":"info","ts":"2024-03-16T14:46:54.645Z","caller":"embed/etcd.go:131","msg":"configuring peer listeners","listen-peer-urls":["https://192.168.26.100:2380"]}
{"level":"info","ts":"2024-03-16T14:46:54.645Z","caller":"embed/etcd.go:479","msg":"starting with peer TLS","tls-info":"cert = /etc/kubernetes/pki/etcd/peer.crt, key = /etc/kubernetes/pki/etcd/peer.key, client-cert=, client-key=, trusted-ca = /etc/kubernetes/pki/etcd/ca.crt, client-cert-auth = true, crl-file = ","cipher-suites":[]}
{"level":"info","ts":"2024-03-16T14:46:54.645Z","caller":"embed/etcd.go:139","msg":"configuring client listeners","listen-client-urls":["https://127.0.0.1:2379","https://192.168.26.100:2379"]}
{"level":"info","ts":"2024-03-16T14:46:54.645Z","caller":"embed/etcd.go:308","msg":"starting an etcd server","etcd-version":"3.5.4","git-sha":"08407ff76","go-version":"go1.16.15","go-os":"linux","go-arch":"amd64","max-cpu-set":4,"max-cpu-available":4,"member-initialized":true,"name":"vms100.liruilongs.github.io","data-dir":"/var/lib/etcd","wal-dir":"","wal-dir-dedicated":"","member-dir":"/var/lib/etcd/member","force-new-cluster":false,"heartbeat-interval":"100ms","election-timeout":"1s","initial-election-tick-advance":true,"snapshot-count":10000,"snapshot-catchup-entries":5000,"initial-advertise-peer-urls":["https://192.168.26.100:2380"],"listen-peer-urls":["https://192.168.26.100:2380"],"advertise-client-urls":["https://192.168.26.100:2379"],"listen-client-urls":["https://127.0.0.1:2379","https://192.168.26.100:2379"],"listen-metrics-urls":["http://127.0.0.1:2381"],"cors":["*"],"host-whitelist":["*"],"initial-cluster":"","initial-cluster-state":"new","initial-cluster-token":"","quota-size-bytes":2147483648,"pre-vote":true,"initial-corrupt-check":true,"corrupt-check-time-interval":"0s","auto-compaction-mode":"periodic","auto-compaction-retention":"0s","auto-compaction-interval":"0s","discovery-url":"","discovery-proxy":"","downgrade-check-interval":"5s"}
panic: freepages: failed to get all reachable pages (page 7744: multiple references)goroutine 109 [running]:
go.etcd.io/bbolt.(*DB).freepages.func2(0xc00009c480)/go/pkg/mod/go.etcd.io/bbolt@v1.3.6/db.go:1056 +0xe9
created by go.etcd.io/bbolt.(*DB).freepages/go/pkg/mod/go.etcd.io/bbolt@v1.3.6/db.go:1054 +0x1cd
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

如何解决

这里最快的办法是重新同步一下这个节点的数据，即把这个故障节点移出集群，清理完故障节点旧数据在重新添加，操作步骤

清理数据目录，移动静态Pod 的yaml 文件：停止故障节点服务，然后删除etcd数据目录。
移除故障节点：使用member remove命令剔除错误节点,可以在健康的节点执行命令。
添加节点：使用member add命令添加故障节点。
重新启动：移动故障节点yaml文件，进行启动

注: 静态Pod 通过加载指定目录的 yaml 文件来调度，kubelet 会定时扫描，删除移动 yaml 文件，静态 Pod 会自动停止,同理。添加 yaml 文件会自动创建静态 Pod

移动静态Pod 的yaml 文件

┌──[root@vms100.liruilongs.github.io]-[~]
└─$mv  /etc/kubernetes/manifests/{etcd.yaml,kube-apiserver.yaml}  /tmp/

删除etcd数据目录

┌──[root@vms100.liruilongs.github.io]-[~]
└─$rm -rf /var/lib/etcd/*

确认节点的 etcd 和 apiservier 都已经停止

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep apiserver
kube-system          kube-apiserver-vms101.liruilongs.github.io            1/1     Running   272 (4h15m ago)   415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms102.liruilongs.github.io            1/1     Running   246 (4h15m ago)   415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep etcd
kube-system          etcd-vms101.liruilongs.github.io                      1/1     Running   167 (4h15m ago)   415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms102.liruilongs.github.io                      1/1     Running   173 (4h15m ago)   415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

获取故障节点 ID,下面的操作我们在健康的 etcd 节点执行，或者可以修改 --endpoints

┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://192.168.26.101:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt" member list -w table
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
|        ID        | STATUS  |            NAME             |         PEER ADDRS          |        CLIENT ADDRS         |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
|  ee392e5273e89e2 | started | vms100.liruilongs.github.io | https://192.168.26.100:2380 | https://192.168.26.100:2379 |
| 70059e836d19883d | started | vms101.liruilongs.github.io | https://192.168.26.101:2380 | https://192.168.26.101:2379 |
| b8cb9f66c2e63b91 | started | vms102.liruilongs.github.io | https://192.168.26.102:2380 | https://192.168.26.102:2379 |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+

移除故障节点

┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt"  member remove ee392e5273e89e2
Member  ee392e5273e89e2 removed from cluster 4816f346663d82a7

重新添加

┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt"  member add vms100.liruilongs.github.io --peer-urls=https://192.168.26.100:2380
Member 456f71fdc1ad9917 added to cluster 4816f346663d82a7ETCD_NAME="vms100.liruilongs.github.io"
ETCD_INITIAL_CLUSTER="vms100.liruilongs.github.io=https://192.168.26.100:2380,vms101.liruilongs.github.io=https://192.168.26.101:2380,vms102.liruilongs.github.io=https://192.168.26.102:2380"
ETCD_INITIAL_ADVERTISE_PEER_URLS="https://192.168.26.100:2380"
ETCD_INITIAL_CLUSTER_STATE="existing"

回到 100 节点机器,移动 Yaml 文件，恢复节点

┌──[root@vms100.liruilongs.github.io]-[~]
└─$mv /tmp/{etcd.yaml,kube-apiserver.yaml} /etc/kubernetes/manifests/

确认 Pod 状态

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep etcd
kube-system          etcd-vms100.liruilongs.github.io                      1/1     Running   0                 16s    192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms101.liruilongs.github.io                      1/1     Running   167 (4h32m ago)   415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms102.liruilongs.github.io                      1/1     Running   173 (4h32m ago)   415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep apiserver
kube-system          kube-apiserver-vms100.liruilongs.github.io            1/1     Running   0                 24s    192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms101.liruilongs.github.io            1/1     Running   272 (4h32m ago)   415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms102.liruilongs.github.io            1/1     Running   246 (4h32m ago)   415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

查看 etcd 集群状态

┌──[root@vms101.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt" member list -w table
+------------------+-----------+-----------------------------+-----------------------------+-----------------------------+
|        ID        |  STATUS   |            NAME             |         PEER ADDRS          |        CLIENT ADDRS         |
+------------------+-----------+-----------------------------+-----------------------------+-----------------------------+
| 54952f3b494c0286 | unstarted |                             | https://192.168.26.100:2380 |                             |
| 70059e836d19883d |   started | vms101.liruilongs.github.io | https://192.168.26.101:2380 | https://192.168.26.101:2379 |
| b8cb9f66c2e63b91 |   started | vms102.liruilongs.github.io | https://192.168.26.102:2380 | https://192.168.26.102:2379 |
+------------------+-----------+-----------------------------+-----------------------------+-----------------------------+

这里我们发现新添加的节点状态不正常，一直是 unstarted

我们在故障节点执行 etcd 命令。发现故障节点并没有添加到集群，而是作为一个单节点运行。

┌──[root@vms100.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt" member list -w table
+-----------------+---------+-----------------------------+-----------------------------+-----------------------------+
|       ID        | STATUS  |            NAME             |         PEER ADDRS          |        CLIENT ADDRS         |
+-----------------+---------+-----------------------------+-----------------------------+-----------------------------+
| ee392e5273e89e2 | started | vms100.liruilongs.github.io | https://192.168.26.100:2380 | https://192.168.26.100:2379 |
+-----------------+---------+-----------------------------+-----------------------------+-----------------------------+
┌──[root@vms100.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt" endpoint status --cluster  -w table
+-----------------------------+-----------------+---------+---------+-----------+-----------+------------+
|          ENDPOINT           |       ID        | VERSION | DB SIZE | IS LEADER | RAFT TERM | RAFT INDEX |
+-----------------------------+-----------------+---------+---------+-----------+-----------+------------+
| https://192.168.26.100:2379 | ee392e5273e89e2 |   3.5.4 |  815 kB |      true |         2 |       2261 |
+-----------------------------+-----------------+---------+---------+-----------+-----------+------------+
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

也没有同步当前集群的数据

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide  --server=https://vms100.liruilongs.github.io:6443
No resources found

遇到这种情况，大部分原因是某个节点的 etcd配置文件的问题，我的这个问题是 故障节点的 etcd 配置文件，没有集群信息相关配置，所以这里把集群相关配置写入配置

原本的配置文件

┌──[root@vms100.liruilongs.github.io]-[~]
└─$cat /etc/kubernetes/manifests/etcd.yaml
apiVersion: v1
kind: Pod
metadata:annotations:kubeadm.kubernetes.io/etcd.advertise-client-urls: https://192.168.26.100:2379creationTimestamp: nulllabels:component: etcdtier: control-planename: etcdnamespace: kube-system
spec:containers:- command:- etcd- --advertise-client-urls=https://192.168.26.100:2379- --cert-file=/etc/kubernetes/pki/etcd/server.crt- --client-cert-auth=true- --data-dir=/var/lib/etcd- --experimental-initial-corrupt-check=true- --experimental-watch-progress-notify-interval=5s- --initial-advertise-peer-urls=https://192.168.26.100:2380- --initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380- --key-file=/etc/kubernetes/pki/etcd/server.key- --listen-client-urls=https://127.0.0.1:2379,https://192.168.26.100:2379- --listen-metrics-urls=http://127.0.0.1:2381- --listen-peer-urls=https://192.168.26.100:2380- --name=vms100.liruilongs.github.io- --peer-cert-file=/etc/kubernetes/pki/etcd/peer.crt- --peer-client-cert-auth=true- --peer-key-file=/etc/kubernetes/pki/etcd/peer.key- --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt- --snapshot-count=10000- --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crtimage: registry.aliyuncs.com/google_containers/etcd:3.5.4-0
。。。。。。。。。。。。。。。。

集群信息不全的，添加后的配置文件

┌──[root@vms100.liruilongs.github.io]-[~]
└─$cat /etc/kubernetes/manifests/etcd.yaml
apiVersion: v1
kind: Pod
metadata:annotations:kubeadm.kubernetes.io/etcd.advertise-client-urls: https://192.168.26.100:2379creationTimestamp: nulllabels:component: etcdtier: control-planename: etcdnamespace: kube-system
spec:containers:- command:- etcd- --advertise-client-urls=https://192.168.26.100:2379- --cert-file=/etc/kubernetes/pki/etcd/server.crt- --client-cert-auth=true- --data-dir=/var/lib/etcd- --experimental-initial-corrupt-check=true- --experimental-watch-progress-notify-interval=5s- --initial-advertise-peer-urls=https://192.168.26.100:2380- --initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380,vms101.liruilongs.github.io=https://192.168.26.101:2380- --initial-cluster-state=existing- --key-file=/etc/kubernetes/pki/etcd/server.key- --listen-client-urls=https://127.0.0.1:2379,https://192.168.26.100:2379- --listen-metrics-urls=http://127.0.0.1:2381- --listen-peer-urls=https://192.168.26.100:2380- --name=vms100.liruilongs.github.io- --peer-cert-file=/etc/kubernetes/pki/etcd/peer.crt- --peer-client-cert-auth=true- --peer-key-file=/etc/kubernetes/pki/etcd/peer.key- --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt- --snapshot-count=10000- --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt

然后我们以上面相同的方式从新恢复一次，发现节点直接没有起来

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep apiserver
kube-system          kube-apiserver-vms100.liruilongs.github.io            0/1     CrashLoopBackOff   1 (18s ago)       39s    192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms101.liruilongs.github.io            1/1     Running            272 (5h29m ago)   415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          kube-apiserver-vms102.liruilongs.github.io            1/1     Running            246 (5h29m ago)   415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>
┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl get pod -A -o wide | grep etcd
kube-system          etcd-vms100.liruilongs.github.io                      0/1     CrashLoopBackOff   3 (21s ago)       53s    192.168.26.100   vms100.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms101.liruilongs.github.io                      1/1     Running            167 (5h29m ago)   415d   192.168.26.101   vms101.liruilongs.github.io   <none>           <none>
kube-system          etcd-vms102.liruilongs.github.io                      1/1     Running            173 (5h29m ago)   415d   192.168.26.102   vms102.liruilongs.github.io   <none>           <none>

查看日志

┌──[root@vms100.liruilongs.github.io]-[~]
└─$kubectl logs etcd-vms100.liruilongs.github.io -n kube-system
.............................
{"level":"fatal","ts":"2024-03-16T16:25:19.981Z","caller":"etcdmain/etcd.go:204","msg":"discovery failed","error":"error validating peerURLs {ClusterID:4816f346663d82a7 Members:[&{ID:b8cb9f66c2e63b91 RaftAttributes:{PeerURLs:[https://192.168.26.102:2380] IsLearner:false} Attributes:{Name:vms102.liruilongs.github.io ClientURLs:[https://192.168.26.102:2379]}} &{ID:3fbbbed942c51f7b RaftAttributes:{PeerURLs:[https://192.168.26.100:2380] IsLearner:false} Attributes:{Name: ClientURLs:[]}} &{ID:70059e836d19883d RaftAttributes:{PeerURLs:[https://192.168.26.101:2380] IsLearner:false} Attributes:{Name:vms101.liruilongs.github.io ClientURLs:[https://192.168.26.101:2379]}}] RemovedMemberIDs:[]}: member count is unequal","stacktrace":"go.etcd.io/etcd/server/v3/etcdmain.startEtcdOrProxyV2\n\t/go/src/go.etcd.io/etcd/release/etcd/server/etcdmain/etcd.go:204\ngo.etcd.io/etcd/server/v3/etcdmain.Main\n\t/go/src/go.etcd.io/etcd/release/etcd/server/etcdmain/main.go:40\nmain.main\n\t/go/src/go.etcd.io/etcd/release/etcd/server/main.go:32\nruntime.main\n\t/go/gos/go1.16.15/src/runtime/proc.go:225"}

根据日志信息，可以看到有用的信息 RemovedMemberIDs:[]}: member count is unequal ,成员数量不相等,在分析日志

{"level": "info","ts": "2024-03-16T16:25:19.961Z","caller": "etcdmain/etcd.go:73","msg": "Running: ","args": ["etcd","--advertise-client-urls=https://192.168.26.100:2379","--cert-file=/etc/kubernetes/pki/etcd/server.crt","--client-cert-auth=true","--data-dir=/var/lib/etcd","--experimental-initial-corrupt-check=true","--experimental-watch-progress-notify-interval=5s","--initial-advertise-peer-urls=https://192.168.26.100:2380","--initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380,vms101.liruilongs.github.io=https://192.168.26.101:2380","--initial-cluster-state=existing","--key-file=/etc/kubernetes/pki/etcd/server.key","--listen-client-urls=https://127.0.0.1:2379,https://192.168.26.100:2379","--listen-metrics-urls=http://127.0.0.1:2381","--listen-peer-urls=https://192.168.26.100:2380","--name=vms100.liruilongs.github.io","--peer-cert-file=/etc/kubernetes/pki/etcd/peer.crt","--peer-client-cert-auth=true","--peer-key-file=/etc/kubernetes/pki/etcd/peer.key","--peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt","--snapshot-count=10000","--trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt"]
}
..............................................................................
{"level": "warn","ts": "2024-03-16T16:25:19.981Z","caller": "etcdmain/etcd.go:146","msg": "failed to start etcd","error": "error validating peerURLs {ClusterID:4816f346663d82a7 Members:[&{ID:b8cb9f66c2e63b91 RaftAttributes:{PeerURLs:[https://192.168.26.102:2380] IsLearner:false} Attributes:{Name:vms102.liruilongs.github.io ClientURLs:[https://192.168.26.102:2379]}} &{ID:3fbbbed942c51f7b RaftAttributes:{PeerURLs:[https://192.168.26.100:2380] IsLearner:false} Attributes:{Name: ClientURLs:[]}} &{ID:70059e836d19883d RaftAttributes:{PeerURLs:[https://192.168.26.101:2380] IsLearner:false} Attributes:{Name:vms101.liruilongs.github.io ClientURLs:[https://192.168.26.101:2379]}}] RemovedMemberIDs:[]}: member count is unequal"
}
{"level": "fatal","ts": "2024-03-16T16:25:19.981Z","caller": "etcdmain/etcd.go:204","msg": "discovery failed","error": "error validating peerURLs {ClusterID:4816f346663d82a7 Members:[&{ID:b8cb9f66c2e63b91 RaftAttributes:{PeerURLs:[https://192.168.26.102:2380] IsLearner:false} Attributes:{Name:vms102.liruilongs.github.io ClientURLs:[https://192.168.26.102:2379]}} &{ID:3fbbbed942c51f7b RaftAttributes:{PeerURLs:[https://192.168.26.100:2380] IsLearner:false} Attributes:{Name: ClientURLs:[]}} &{ID:70059e836d19883d RaftAttributes:{PeerURLs:[https://192.168.26.101:2380] IsLearner:false} Attributes:{Name:vms101.liruilongs.github.io ClientURLs:[https://192.168.26.101:2379]}}] RemovedMemberIDs:[]}: member count is unequal","stacktrace": "go.etcd.io/etcd/server/v3/etcdmain.startEtcdOrProxyV2\n\t/go/src/go.etcd.io/etcd/release/etcd/server/etcdmain/etcd.go:204\ngo.etcd.io/etcd/server/v3/etcdmain.Main\n\t/go/src/go.etcd.io/etcd/release/etcd/server/etcdmain/main.go:40\nmain.main\n\t/go/src/go.etcd.io/etcd/release/etcd/server/main.go:32\nruntime.main\n\t/go/gos/go1.16.15/src/runtime/proc.go:225"
}

可以看到它提示可能错误与 vms102.liruilongs.github.io 节点相关

然后我们看一下 vms102.liruilongs.github.io 的配置文件

┌──[root@vms102.liruilongs.github.io]-[~]
└─$cat /etc/kubernetes/manifests/etcd.yaml
apiVersion: v1
kind: Pod
metadata:annotations:kubeadm.kubernetes.io/etcd.advertise-client-urls: https://192.168.26.102:2379creationTimestamp: nulllabels:component: etcdtier: control-planename: etcdnamespace: kube-system
spec:containers:- command:- etcd- --advertise-client-urls=https://192.168.26.102:2379- --cert-file=/etc/kubernetes/pki/etcd/server.crt- --client-cert-auth=true- --data-dir=/var/lib/etcd- --experimental-initial-corrupt-check=true- --experimental-watch-progress-notify-interval=5s- --initial-advertise-peer-urls=https://192.168.26.102:2380- --initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380,vms102.liruilongs.github.io=https://192.168.26.102:2380,vms101.liruilongs.github.io=https://192.168.26.101:2380- --initial-cluster-state=existing- --key-file=/etc/kubernetes/pki/etcd/server.key- --listen-client-urls=https://127.0.0.1:2379,https://192.168.26.102:2379- --listen-metrics-urls=http://127.0.0.1:2381- --listen-peer-urls=https://192.168.26.102:2380- --name=vms102.liruilongs.github.io- --peer-cert-file=/etc/kubernetes/pki/etcd/peer.crt- --peer-client-cert-auth=true- --peer-key-file=/etc/kubernetes/pki/etcd/peer.key- --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt- --snapshot-count=10000- --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.crt

通过配置文件比对，可以发现，之前配置的故障节点的配置任然有问题，少了一个vms102.liruilongs.github.io=https://192.168.26.102:2380节点信息。

"--initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380,vms101.liruilongs.github.io=https://192.168.26.101:2380",
"--initial-cluster=vms100.liruilongs.github.io=https://192.168.26.100:2380,vms102.liruilongs.github.io=https://192.168.26.102:2380,vms101.liruilongs.github.io=https://192.168.26.101:2380"

修改完配置，按照上面相同的流程重新恢复节点，节点恢复

通过 etcdctl 命令检查

┌──[root@vms100.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt" member list -w table
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
|        ID        | STATUS  |            NAME             |         PEER ADDRS          |        CLIENT ADDRS         |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
| 70059e836d19883d | started | vms101.liruilongs.github.io | https://192.168.26.101:2380 | https://192.168.26.101:2379 |
| ac5f6045dbe477b3 | started | vms100.liruilongs.github.io | https://192.168.26.100:2380 | https://192.168.26.100:2379 |
| b8cb9f66c2e63b91 | started | vms102.liruilongs.github.io | https://192.168.26.102:2380 | https://192.168.26.102:2379 |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+

┌──[root@vms100.liruilongs.github.io]-[~]
└─$ETCDCTL_API=3 etcdctl  --endpoints https://127.0.0.1:2379  --cert="/etc/kubernetes/pki/etcd/server.crt"  --key="/etc/kubernetes/pki/etcd/server.key"  --cacert="/etc/kubernetes/pki/etcd/ca.crt" endpoint status --cluster  -w table
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
|          ENDPOINT           |        ID        | VERSION | DB SIZE | IS LEADER | RAFT TERM | RAFT INDEX |
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
| https://192.168.26.101:2379 | 70059e836d19883d |   3.5.4 |   88 MB |     false |       603 |   22227327 |
| https://192.168.26.100:2379 | ac5f6045dbe477b3 |   3.5.4 |   88 MB |     false |       603 |   22227327 |
| https://192.168.26.102:2379 | b8cb9f66c2e63b91 |   3.5.4 |   88 MB |      true |       603 |   22227327 |
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
┌──[root@vms100.liruilongs.github.io]-[~]
└─$

故障节点恢复，在实际的操作中，添加完节点，我们需要确认故障节点的配置文件是否是正确的配置文件

K8s 集群高可用master节点ETCD挂掉如何恢复?

写在前面很常见的集群运维场景，整理分享博文内容为 K8s 集群高可用 master 节点故障如何恢复的过程理解不足小伙伴帮忙指正不必太纠结于当下，也不必太忧虑未来，当你经历过一些事情的时候，眼前的风景已经和从前不一样了。——村上…...

编程日记 2024/3/18 5:48:09

【Godot 4.2】常见几何图形、网格、刻度线点求取函数及原理总结

概述本篇为ShapePoints静态函数库的补充和辅助文档。ShapePoints函数库是一个用于生成常见几何图形顶点数据（PackedVector2Array）的静态函数库。生成的数据可用于_draw和Line2D、Polygon2D等进行绘制和显示。因为不断地持续扩展，ShapePoint…...

编程日记 2024/3/18 5:47:08

如何利用POI导出报表

一、报表格式二、依赖坐标 <dependency><groupId>org.apache.poi</groupId><artifactId>poi</artifactId><version>3.16</version> </dependency> <dependency><groupId>org.apache.poi</groupId><art…...

编程日记 2024/3/18 5:46:07

自动部署SSL证书到阿里云腾讯云CDN

项目地址：https://github.com/yxzlwz/ssl_update 项目简介目前，自动申请和管理免费SSL证书的项目有很多，如个人正在使用的 acme.sh。然而在申请后，如果我们的需求不仅限于服务器本地的使用，证书的部署也是一件麻烦事…...

编程日记 2024/3/18 5:43:05

【系统性】循序渐进学C++

循序渐进学C 第一阶段：基础一、环境配置 1.1.第一个程序（基本格式） #include <iosteam> using namespace std;int main(){cout<<"hello world"<<endl;system("pause"); } 模板 #include &…...

编程日记 2024/3/18 5:42:04

rust - 一个日志缓存记录的通用实现

本文给出了一个通用的设计模式，通过建造者模式实例化记录对象，可自定义格式化器将实例化后的记录对象写入到指定的缓存对象中。定义记录对象 use chrono::prelude::*; use std::{cell::RefCell, ffi::OsStr, fmt, io, io::Write, path::Path, rc::Rc,…...

编程日记 2024/3/18 5:41:03

elasticsearch(RestHighLevelClient API操作)(黑马)

操作全是换汤不换药，创建一个request，然后使用client发送就可以了一、增加索引库数据 Testvoid testAddDocument() throws IOException {//从数据库查出数据Writer writer writerService.getById(199);//将查出来的数据处理成json字符串String json …...

编程日记 2024/3/18 5:40:00

用尾插的思想实现移除链表中的元素

目录一、介绍尾插 1.链表为空 2.链表不为空二、题目介绍三、思路四、代码五、代码解析 1. 2. 3. 4. 5. 6. 六、注意点 1. 2. 一、介绍尾插整体思路为 1.链表为空 void SLPushBack(SLTNode** pphead, SLTDataType x) {SLTNode* newnode BuyLTNode(x); …...

编程日记 2024/3/18 5:38:59

【Kubernetes】k8s删除master节点后重新加入集群

目录前言一、思路二、实战1.安装etcdctl指令2.重置旧节点的k8s3.旧节点的的 etcd 从 etcd 集群删除4.在 master03 上，创建存放证书目录5.把其他控制节点的证书拷贝到 master01 上6.把 master03 加入到集群7.验证 master03 是否加入到 k8s 集群，检查业务…...

编程日记 2024/3/18 5:36:57

HCIP—OSPF虚链路实验

OSPF虚链路—Vlink 作用：专门解决OSPF不规则区域所诞生的技术，是一种虚拟的，逻辑的链路。实现非骨干区域和骨干区域在逻辑上直接连接。注意虚链路条件：只能穿越一个区域，通常对虚链路进行认证功能的配置。虚链路认证也…...

编程日记 2024/3/18 5:35:57

RAxML-NG安装与使用-raxml-ng-v1.2.0（bioinfomatics tools-013）

01 背景 1.1 ML树 ML树，或最大似然树，是一种在进化生物学中用来推断物种之间进化关系的方法。最大似然（Maximum Likelihood, ML）是一种统计框架，用于估计模型参数，使得观察到的数据在该模型参数下的概率最…...

编程日记 2024/3/18 5:34:56

Tomcat内存马

Tomcat内存马前言描述Servlet3.0后允许动态注册组件这一技术的实现有赖于官方对Servlet3.0的升级，Servlet在3.0版本之后能够支持动态注册组件。而Tomcat直到7.x才支持Servlet3.0，因此通过动态添加恶意组件注入内存马的方式适合Tomcat7.x及以上。…...

编程日记 2024/3/18 5:33:55

pytorch之诗词生成3--utils

先上代码： import numpy as np import settingsdef generate_random_poetry(tokenizer, model, s):"""随机生成一首诗:param tokenizer: 分词器:param model: 用于生成古诗的模型:param s: 用于生成古诗的起始字符串，默认为空串:return: …...

编程日记 2024/3/18 5:32:54

OpenAI的ChatGPT企业版专注于安全性、可扩展性和定制化。

OpenAI的ChatGPT企业版：安全、可扩展性和定制化的重点 OpenAI的ChatGPT在商业世界引起了巨大反响，而最近推出的ChatGPT企业版更是证明了其在企业界的日益重要地位。企业版ChatGPT拥有企业级安全、无限GPT-4访问、更长的上下文窗口以及一系列定制选项等增…...

编程日记 2024/3/18 5:31:53

JS06-class对象

class对象 className 修改样式 <!DOCTYPE html> <html lang"en"> <head><meta charset"UTF-8"><meta http-equiv"X-UA-Compatible" content"IEedge"><meta name"viewport" content&quo…...

编程日记 2024/3/18 5:30:52

深度学习1650ti在win10安装pytorch复盘

深度学习1650ti在win10安装pytorch复盘前言1. 安装anaconda2. 检查更新显卡驱动3. 根据pytorch选择CUDA版本4. 安装CUDA5. 安装cuDNN6. conda安装pytorch结语前言建议有条件的，可以在安装过程中，开启梯子。例如cuDNN安装时登录 or 注册，会…...

编程日记 2024/3/18 5:29:51

Node.js与webpack（三）

上一节：Node.js与Webpack笔记（二）-CSDN博客从0来一遍（webpack项目） 将之前的webpack 的纯开发配置，重新创建空白项目，重新做一遍，捋一遍思路防止加入生产模式时候弄混 1.创建文件夹…...

编程日记 2024/3/18 5:28:50

测试覆盖率那些事

在测试过程中，会出现测试覆盖不全的情况，特别是工期紧张的情况下，测试的时间被项目的周期一压再压，测试覆盖概率不全就会伴随而来。网上冲浪，了解一下覆盖率的文章，其中一篇感觉写的很不错，将…...

编程日记 2024/3/18 5:27:49

Etcd 介绍与使用（入门篇）

etcd 介绍 etcd 简介 etc （基于 Go 语言实现）在 Linux 系统中是配置文件目录名；etcd 就是配置服务； etcd 诞生于 CoreOS 公司，最初用于解决集群管理系统中 os 升级时的分布式并发控制、配置文件的存储与分发等问题。基…...

编程日记 2024/3/18 5:26:48

Docker 安装 LogStash

关于LogStash Logstash，作为Elastic Stack家族中的核心成员之一，是一个功能强大的开源数据收集引擎。它专长于从各种来源动态地获取、解析、转换和丰富数据，并将这些结构化或非结构化的数据高效地传输到诸如Elasticsearch等存储系统中进行集…...

编程日记 2024/3/18 5:25:47

（十）学生端搭建

本次旨在将之前的已完成的部分功能进行拼装到学生端，同时完善学生端的构建。本次工作主要包括： 1.学生端整体界面布局 2.模拟考场与部分个人画像流程的串联 3.整体学生端逻辑一、学生端在主界面可以选择自己的用户角色选择学生则进入学生登录界面…...

编程新知 2025/10/30 4:27:23

基于uniapp+WebSocket实现聊天对话、消息监听、消息推送、聊天室等功能，多端兼容

基于 UniApp + WebSocket实现多端兼容的实时通讯系统，涵盖WebSocket连接建立、消息收发机制、多端兼容性配置、消息实时监听等功能，适配微信小程序、H5、Android、iOS等终端目录技术选型分析WebSocket协议优势UniApp跨平台特性WebSocket 基础实现连接管理消息收发连接…...

编程新知 2025/10/19 17:15:36

1688商品列表API与其他数据源的对接思路

将1688商品列表API与其他数据源对接时，需结合业务场景设计数据流转链路，重点关注数据格式兼容性、接口调用频率控制及数据一致性维护。以下是具体对接思路及关键技术点： 一、核心对接场景与目标商品数据同步场景：将1688商品信息…...

编程新知 2025/9/11 12:55:04

定时器任务——若依源码分析

分析util包下面的工具类schedule utils： ScheduleUtils 是若依中用于与 Quartz 框架交互的工具类，封装了定时任务的创建、更新、暂停、删除等核心逻辑。 createScheduleJob createScheduleJob 用于将任务注册到 Quartz，先构建任务的 JobD…...

编程新知 2025/10/19 9:46:38

2025 后端自学UNIAPP【项目实战：旅游项目】6、我的收藏页面

代码框架视图 1、先添加一个获取收藏景点的列表请求【在文件my_api.js文件中添加】 // 引入公共的请求封装 import http from ./my_http.js// 登录接口（适配服务端返回 Token） export const login async (code, avatar) > {const res await http…...

编程新知 2025/10/27 13:40:48

基于Docker Compose部署Java微服务项目

一. 创建根项目根项目（父项目）主要用于依赖管理一些需要注意的点： 打包方式需要为 pom<modules>里需要注册子模块不要引入maven的打包插件，否则打包时会出问题 <?xml version"1.0" encoding"UTF-8…...

编程新知 2025/10/27 21:54:32

反射获取方法和属性

Java反射获取方法在Java中，反射（Reflection）是一种强大的机制，允许程序在运行时访问和操作类的内部属性和方法。通过反射，可以动态地创建对象、调用方法、改变属性值，这在很多Java框架中如Spring和Hiberna…...

编程新知 2025/10/11 22:46:21

C++ 求圆面积的程序（Program to find area of a circle）

给定半径r，求圆的面积。圆的面积应精确到小数点后5位。例子： 输入：r 5 输出：78.53982 解释：由于面积 PI * r * r 3.14159265358979323846 * 5 * 5 78.53982，因为我们只保留小数点后 5 位数字。输…...

编程新知 2025/10/29 6:13:18

CMake 从 GitHub 下载第三方库并使用

有时我们希望直接使用 GitHub 上的开源库，而不想手动下载、编译和安装。可以利用 CMake 提供的 FetchContent 模块来实现自动下载、构建和链接第三方库。 FetchContent 命令官方文档✅ 示例代码我们将以 fmt 这个流行的格式化库为例，演示如何：使用 FetchContent 从 GitH…...

编程新知 2025/10/30 5:11:46

有限自动机到正规文法转换器v1.0

1 项目简介这是一个功能强大的有限自动机（Finite Automaton, FA）到正规文法（Regular Grammar）转换器，它配备了一个直观且完整的图形用户界面，使用户能够轻松地进行操作和观察。该程序基于编译原理中的经典…...

编程新知 2025/10/18 18:23:21

写在前面

遇到了什么问题

如何排查

如何解决

相关文章：