k8s无法正常启动使用，排查：etcd损坏
k8s无法正常启动使用，排查：etcd损坏
问题：
在跑项目的时候，机器意外断电了，重启后发现kubectl无法使用，报错如下：
[root@k8s-master01 wal]# kubectl get pod
The connection to the server 192.168.18.101:6443 was refused - did you specify the right host or port?
排查：
1、
[root@k8s-master01 ~]# journalctl -fu kubelet
5月 06 12:05:34 k8s-master01 kubelet[18467]: I0506 12:05:34.798387 18467 kubelet.go:461] "Kubelet nodes not sync"
5月 06 12:05:35 k8s-master01 kubelet[18467]: I0506 12:05:35.120046 18467 kubelet.go:461] "Kubelet nodes not sync"
5月 06 12:05:35 k8s-master01 kubelet[18467]: I0506 12:05:35.368482 18467 kubelet.go:461] "Kubelet nodes not sync"
5月 06 12:05:35 k8s-master01 kubelet[18467]: I0506 12:05:35.432439 18467 kubelet.go:461] "Kubelet nodes not sync"
5月 06 12:05:35 k8s-master01 kubelet[18467]: I0506 12:05:35.626419 18467 kubelet_node_status.go:71] "Attempting to register node" node="k8s-master01"
5月 06 12:05:35 k8s-master01 kubelet[18467]: E0506 12:05:35.626808 18467 kubelet_node_status.go:93] "Unable to register node with API server" err="Post \"https://192.168.18.101:6443/api/v1/nodes\": dial tcp 192.168.18.101:6443: connect: connection refused" node="k8s-master01"
5月 06 12:05:35 k8s-master01 kubelet[18467]: I0506 12:05:35.797123 18467 kubelet.go:461] "Kubelet nodes not sync"
5月 06 12:05:35 k8s-master01 kubelet[18467]: I0506 12:05:35.797226 18467 kubelet.go:461] "Kubelet nodes not sync"
5月 06 12:05:35 k8s-master01 kubelet[18467]: E0506 12:05:35.797249 18467 kubelet.go:2298] "Error getting node" err="nodes have not yet been read at least once, cannot construct node object"
Unable to register node with API server" err=“Post “https://192.168.18.101:6443/api/v1/nodes”: dial tcp 192.168.18.101:6443: connect: connection refused” node="k8s-master01
发现连接不到apiserver
2、然后我到docker容器里，发现etcd和api都处于exitd状态
查看了apiserver相关的容器log
报错如下：
[root@k8s-master01 ~]# docker logs k8s_kube-apiserver_kube-apiserver-k8s-master01_kube-system_2dd80c82feea0886cf3a8a74b4b80db1_26
I0506 04:08:32.598696 1 client.go:360] parsed scheme: "endpoint"
I0506 04:08:32.598776 1 endpoint.go:68] ccResolverWrapper: sending new addresses to cc: [{https://127.0.0.1:2379 <nil> 0 <nil>}]
W0506 04:08:32.599383 1 clientconn.go:1223] grpc: addrConn.createTransport failed to connect to {https://127.0.0.1:2379 <nil> 0 <nil>}. Err :connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:2379: connect: connection refused". Reconnecting...
I0506 04:08:33.594958 1 client.go:360] parsed scheme: "endpoint"
I0506 04:08:33.595009 1 endpoint.go:68] ccResolverWrapper: sending new addresses to cc: [{https://127.0.0.1:2379 <nil> 0 <nil>}]
W0506 04:08:33.596996 1 clientconn.go:1223] grpc: addrConn.createTransport failed to connect to {https://127.0.0.1:2379 <nil> 0 <nil>}. Err :connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:2379: connect: connection refused". Reconnecting...
transport: Error while dialing dial tcp 127.0.0.1:2379: connect: connection refused
好你个拒绝连接，来看看2379端口是干嘛的
[root@k8s-master01 ~]# cat /etc/kubernetes/manifests/kube-apiserver.yaml
原来是etcd连接失败，怪不得容器状态也是exitd。
3、查看etcd容器的log，发现报错如下：
[root@k8s-master01 ~]# docker logs k8s_etcd_etcd-k8s-master01_kube-system_7a98aec14f1d812921d8b2f6c35fbbcc_24
[WARNING] Deprecated '--logger=capnslog' flag is set; use '--logger=zap' flag instead
2023-05-06 04:10:42.638389 I | etcdmain: etcd Version: 3.4.13
2023-05-06 04:10:42.638426 I | etcdmain: Git SHA: ae9734ed2
2023-05-06 04:10:42.638429 I | etcdmain: Go Version: go1.12.17
2023-05-06 04:10:42.638431 I | etcdmain: Go OS/Arch: linux/amd64
2023-05-06 04:10:42.638434 I | etcdmain: setting maximum number of CPUs to 4, total number of available CPUs is 4
2023-05-06 04:10:42.638485 N | etcdmain: the server is already initialized as member before, starting as etcd member...
[WARNING] Deprecated '--logger=capnslog' flag is set; use '--logger=zap' flag instead
2023-05-06 04:10:42.638516 I | embed: peerTLS: cert = /etc/kubernetes/pki/etcd/peer.crt, key = /etc/kubernetes/pki/etcd/peer.key, trusted-ca = /etc/kubernetes/pki/etcd/ca.crt, client-cert-auth = true, crl-file =
2023-05-06 04:10:42.639017 I | embed: name = k8s-master01
2023-05-06 04:10:42.639042 I | embed: data dir = /var/lib/etcd
2023-05-06 04:10:42.639047 I | embed: member dir = /var/lib/etcd/member
2023-05-06 04:10:42.639050 I | embed: heartbeat = 100ms
2023-05-06 04:10:42.639053 I | embed: election = 1000ms
2023-05-06 04:10:42.639056 I | embed: snapshot count = 10000
2023-05-06 04:10:42.639079 I | embed: advertise client URLs = https://192.168.18.101:2379
2023-05-06 04:10:42.639084 I | embed: initial advertise peer URLs = https://192.168.18.101:2380
2023-05-06 04:10:42.639087 I | embed: initial cluster =
2023-05-06 04:10:42.641786 W | wal: ignored file 0000000000000000-0000000000000000.wal.broken in wal
2023-05-06 04:10:42.641832 W | wal: ignored file 0000000000000004-000000000005dd9e.wal.broken in wal
2023-05-06 04:10:43.208339 I | etcdserver: recovered store from snapshot at index 600062
2023-05-06 04:10:43.212163 C | etcdserver: recovering backend from snapshot error: failed to find database snapshot file (snap: snapshot file doesn't exist)
panic: recovering backend from snapshot error: failed to find database snapshot file (snap: snapshot file doesn't exist)
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x20 pc=0xc0587e]
goroutine 1 [running]:
go.etcd.io/etcd/etcdserver.NewServer.func1(0xc0002cae30, 0xc0002c8d80)
/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/server.go:334 +0x3e
panic(0xee6840, 0xc0001120a0)
/usr/local/go/src/runtime/panic.go:522 +0x1b5
github.com/coreos/pkg/capnslog.(*PackageLogger).Panicf(0xc000203a60, 0x10c13b2, 0x2a, 0xc0002c8e50, 0x1, 0x1)
/home/ANT.AMAZON.COM/leegyuho/go/pkg/mod/github.com/coreos/[email protected]/capnslog/pkg_logger.go:75 +0x135
go.etcd.io/etcd/etcdserver.NewServer(0x7ffead4fde41, 0xc, 0x0, 0x0, 0x0, 0x0, 0xc0001bb080, 0x1, 0x1, 0xc0001bb200, ...)
/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/server.go:464 +0x433c
go.etcd.io/etcd/embed.StartEtcd(0xc000292000, 0xc000197080, 0x0, 0x0)
/tmp/etcd-release-3.4.13/etcd/release/etcd/embed/etcd.go:214 +0x988
go.etcd.io/etcd/etcdmain.startEtcd(0xc000292000, 0x10963d6, 0x6, 0x1, 0xc000221140)
/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/etcd.go:302 +0x40
go.etcd.io/etcd/etcdmain.startEtcdOrProxyV2()
/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/etcd.go:144 +0x2ef9
go.etcd.io/etcd/etcdmain.Main()
/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/main.go:46 +0x38
main.main()
/tmp/etcd-release-3.4.13/etcd/release/etcd/main.go:28 +0x20
snap: snapshot file doesn’t exist
发现snap文件不存在，估计是意外断电导致的文件系统损坏或者遗失了
将备份的文件覆盖该目录文件即可，因为我没有备份，是在学习测试阶段，没啥重要东西，就直接删除了
[root@k8s-master01 ~]# rm -rf /var/lib/etcd/member/snap/.*
重启kubelet服务，发现还是报错
[root@k8s-master01 wal]# systemctl restart kubelet.service
[root@k8s-master01 wal]# kubectl get pod
The connection to the server 192.168.18.101:6443 was refused - did you specify the right host or port?
4、继续查看etcd容器日志，发现又有报错：
[root@k8s-master01 ~]# docker logs k8s_etcd_etcd-k8s-master01_kube-system_7a98aec14f1d812921d8b2f6c35fbbcc_27
[WARNING] Deprecated '--logger=capnslog' flag is set; use '--logger=zap' flag instead
2023-05-06 04:20:51.236303 I | etcdmain: etcd Version: 3.4.13
2023-05-06 04:20:51.236348 I | etcdmain: Git SHA: ae9734ed2
2023-05-06 04:20:51.236351 I | etcdmain: Go Version: go1.12.17
2023-05-06 04:20:51.236354 I | etcdmain: Go OS/Arch: linux/amd64
2023-05-06 04:20:51.236358 I | etcdmain: setting maximum number of CPUs to 4, total number of available CPUs is 4
2023-05-06 04:20:51.236409 N | etcdmain: the server is already initialized as member before, starting as etcd member...
[WARNING] Deprecated '--logger=capnslog' flag is set; use '--logger=zap' flag instead
2023-05-06 04:20:51.236456 I | embed: peerTLS: cert = /etc/kubernetes/pki/etcd/peer.crt, key = /etc/kubernetes/pki/etcd/peer.key, trusted-ca = /etc/kubernetes/pki/etcd/ca.crt, client-cert-auth = true, crl-file =
2023-05-06 04:20:51.236926 I | embed: name = k8s-master01
2023-05-06 04:20:51.236947 I | embed: data dir = /var/lib/etcd
2023-05-06 04:20:51.236951 I | embed: member dir = /var/lib/etcd/member
2023-05-06 04:20:51.236953 I | embed: heartbeat = 100ms
2023-05-06 04:20:51.236954 I | embed: election = 1000ms
2023-05-06 04:20:51.236956 I | embed: snapshot count = 10000
2023-05-06 04:20:51.236962 I | embed: advertise client URLs = https://192.168.18.101:2379
2023-05-06 04:20:51.236964 I | embed: initial advertise peer URLs = https://192.168.18.101:2380
2023-05-06 04:20:51.236967 I | embed: initial cluster =
2023-05-06 04:20:51.238833 W | wal: ignored file 0000000000000000-0000000000000000.wal.broken in wal
2023-05-06 04:20:51.238879 W | wal: ignored file 0000000000000004-000000000005dd9e.wal.broken in wal
2023-05-06 04:20:51.801126 W | wal: ignored file 0000000000000000-0000000000000000.wal.broken in wal
2023-05-06 04:20:51.801201 W | wal: ignored file 0000000000000004-000000000005dd9e.wal.broken in wal
2023-05-06 04:20:51.801254 C | etcdserver: open wal error: wal: file not found
etcdserver: open wal error: wal: file not found
发现找不到该文件walr
建议将备份的文件覆盖该目录文件，我就直接删除了
[root@k8s-master01 ~]# rm -rf /var/lib/etcd/member/wal/*
5、重启kubelet，过一会发现可以正常使用了，容器服务也正常启动了
