k8s集群部署时etcd容器不停重启问题以及处理详解

目录

• 问题现象
• 解决问题
• 总结

问题现象

在安装部署Kubernetes 1.26版本时，通过kubeadm初始化集群后，发现执行kubectl命令报以下错误：

The connection to the server localhost:8080 was refused - did you specify the right host or port?

查看kubelet状态是否正常，发现无法连接apiserver的6443端口。

Dec 21 09:36:03 k8s-master kubelet[7127]: E1221 09:36:03.015089    7127 kubelet_node_status.go:540] "Error updating node status, will retry" err="error getting node \"k8s-master\": Get \"https://192.168.2.200:6443/api/v1/nodes/k8s-master?timeout=10s\": dial tcp 192.168.2.200:6443: connect: connection refused"
Dec 21 09:36:03 k8s-master kubelet[7127]: E1221 09:36:03.015445    7127 kubelet_node_status.go:540] "Error updating node status, will retry" err="error getting node \"k8s-master\": Get \"https://192.168.2.200:6443/api/v1/nodes/k8s-master?timeout=10s\": dial tcp 192.168.2.200:6443: connect: connection refused"
Dec 21 09:36:03 k8s-master kubelet[7127]: E1221 09:36:03.015654    7127 kubelet_node_status.go:540] "Error updating node status, will retry" err="error getting node \"k8s-master\": Get \"https://192.168.2.200:6443/api/v1/nodes/k8s-master?timeout=10s\": dial tcp 192.168.2.200:6443: connect: connection refused"
Dec 21 09:36:03 k8s-master kubelet[7127]: E1221 09:36:03.015818    7127 kubelet_node_status.go:540] "Error updating node status, will retry" err="error getting node \"k8s-master\": Get \"https://192.168.2.200:6443/api/v1/nodes/k8s-master?timeout=10s\": dial tcp 192.168.2.200:6443: connect: connection refused"

进而查看apiserver容器的状态，由于是基于containerd作为容器运行时，此时kubectl不可用的情况下，使用crictl ps -a命令可以查看所有容器的情况。

root@k8s-master:~/k8s/calico# crictl ps -a
CONTAINER           IMAGE               CREATED             STATE               NAME                      ATTEMPT             POD ID              POD
395b45b1cb733       a31e1d84401e6       50 seconds ago      Exited              kube-apiserver            28                  e87800ae06ff5       kube-apiserver-k8s-master
b5c7e2a07bf1b       5d7c5dfd3ba18       3 minutes ago       Running             kube-controller-manager   32                  6b7cc9dd07f1d       kube-controller-manager-k8s-master
944aa31862613       556768f31eb1d       4 minutes ago       Exited              kube-proxy                27                  ccb6557c6f629       kube-proxy-ctjjq
c097332b6f416       fce326961ae2d       4 minutes ago       Exited              etcd                      30                  079d491eb9925       etcd-k8s-master
b8103090322c4       dafd8ad70b156       6 minutes ago       Exited              kube-scheduler            32                  48f9544c9798c       kube-scheduler-k8s-master
a14b969e8ad05       5d7c5dfd3ba18       12 minutes ago      Exited              kube-controller-manager   31                  5576806b4e142       kube-controller-manager-k8s-master

发现此时kube-apiserver容器已经退出，查看容器日志是否有异常信息。通过日志信息发现是kube-apiserver无法连接etcd的2379端口，那么问题应该是出在etcd了。

W1221 07:00:20.392868       1 logging.go:59] [core] [Channel #1 SubChannel #2] grpc: addrConn.createTransport failed to connect to {
  "Addr": "127.0.0.1:2379",
  "ServerName": "127.0.0.1",
  "Attributes": null,
  "BalancerAttributes": null,
  "Type": 0,
  "Metadata": null
}. Err: connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:2379: connect: connection refused"
W1221 07:00:21.391330       1 logging.go:59] [core] [Channel #4 SubChannel #6] grpc: addrConn.createTransport failed to connect to {
  "Addr": "127.0.0.1:2379",
  "ServerName": "127.0.0.1",
  "Attributes": null,
  "BalancerAttributes": null,
  "Type": 0,
  "Metadata": null
}. Err: connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:2379: connect: connection refused"

此时etcd容器也在不断地重启，查看其日志发现没有错误级别的信息。

{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"d975d9ebc69964b3 is starting a new election at term 2"}
{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"d975d9ebc69964b3 became pre-candidate at term 2"}
{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"d975d9ebc69964b3 received MsgPreVoteResp from d975d9ebc69964b3 at term 2"}
{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"d975d9ebc69964b3 became candidate at term 3"}
{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"d975d9ebc69964b3 received MsgVoteResp from d975d9ebc69964b3 at term 3"}
{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"d975d9ebc69964b3 became leader at term 3"}
{"level":"info","ts":"2022-12-21T10:29:00.740Z","logger":"raft","caller":"etcdserver/zap_raft.go:77","msg":"raft.node: d975d9ebc69964b3 elected leader d975d9ebc69964b3 at term 3"}
{"level":"info","ts":"2022-12-21T10:29:00.742Z","caller":"etcdserver/server.go:2054","msg":"published local member to cluster through raft","local-member-id":"d975d9ebc69964b3","local-member-attributes":"{Name:k8s-master ClientURLs:[https://192.168.2.200:2379]}","request-path":"/0/members/d975d9ebc69964b3/attributes","cluster-id":"f88ac1c8c4bab6","publish-timeout":"7s"}
{"level":"info","ts":"2022-12-21T10:29:00.742Z","caller":"embed/serve.go:100","msg":"ready to serve client requests"}
{"level":"info","ts":"2022-12-21T10:29:00.742Z","caller":"embed/serve.go:100","msg":"ready to serve client requests"}
{"level":"info","ts":"2022-12-21T10:29:00.743Z","caller":"etcdmain/main.go:44","msg":"notifying init daemon"}
{"level":"info","ts":"2022-12-21T10:29:00.743Z","caller":"etcdmain/main.go:50","msg":"successfully notified init daemon"}
{"level":"info","ts":"2022-12-21T10:29:00.744Z","caller":"embed/serve.go:198","msg":"serving client traffic securely","address":"192.168.2.200:2379"}
{"level":"info","ts":"2022-12-21T10:29:00.745Z","caller":"embed/serve.go:198","msg":"serving client traffic securely","address":"127.0.0.1:2379"}
{"level":"info","ts":"2022-12-21T10:30:20.624Z","caller":"osutil/interrupt_unix.go:64","msg":"received signal; shutting down","signal":"terminated"}
{"level":"info","ts":"2022-12-21T10:30:20.624Z","caller":"embed/etcd.go:373","msg":"closing etcd server","name":"k8s-master","data-dir":"/var/lib/etcd","advertise-peer-urls":["https://192.168.2.200:2380"],"advertise-client-urls":["https://192.168.2.200:2379"]}
{"level":"info","ts":"2022-12-21T10:30:20.636Z","caller":"etcdserver/server.go:1465","msg":"skipped leadership transfer for single voting member cluster","local-member-id":"d975d9ebc69964b3","current-leader-member-id":"d975d9ebc69964b3"}
{"level":"info","ts":"2022-12-21T10:30:20.637Z","caller":"embed/etcd.go:568","msg":"stopping serving peer traffic","address":"192.168.2.200:2380"}
{"level":"info","ts":"2022-12-21T10:30:20.639Z","caller":"embed/etcd.go:573","msg":"stopped serving peer traffic","address":"192.168.2.200:2380"}
{"level":"info","ts":"2022-12-21T10:30:20.639Z","caller":"embed/etcd.go:375","msg":"closed etcd server","name":"k8s-master","data-dir":"/var/lib/etcd","advertise-peer-urls":["https://192.168.2.200:2380"],"advertise-client-urls":["https://192.168.2.200:2379"]}

但是，其中一行日志信息表示etcd收到了关闭的信号，并不是异常退出的。

{"level":"info","ts":"2022-12-21T10:30:20.624Z","caller":"osutil/interrupt_unix.go:64","msg":"received signal; shutting down","signal":"terminated"}

解决问题

该问题为未正确设置cgroups导致，在containerd的配置文件/etc/containerd/config.toml中，修改SystemdCgroup配置为true。

[plugins."io.containerd.grpc.v1.cri".containerd.runtimes.runc.options]
  BinaryName = ""
  CriuImagePath = ""
  CriuPath = ""
  CriuWorkPath = ""
  IoGid = 0
  IoUid = 0
  NoNewKeyring = false
  NoPivotRoot = false
  Root = ""
  ShimCgroup = ""
  SystemdCgroup = true

重启containerd服务

systemctl restart containerd

etcd容器不再重启，其他容器也恢复正常，问题解决。