[others]: 部署在k8s后，无法exec -it登入huatuo pod的shell

[Dongrc]

你的问题

报错如下

Internal error occurred: 
	error executing command in container: 
		failed to exec in container: 
			failed to start exec "720c6ece92ef26568130c69c3283171f77ea5a2c469281e52e4b6b2cb663d446": 
				OCI runtime exec failed: 
					exec failed: 
						unable to start container process: 
							error adding pid 300656 to cgroups: 
								failed to write 300656: 
									openat2 /sys/fs/cgroup/blkio/kubepods.slice/kubepods-besteffort.slice/kubepods-besteffort-pod7ab5c773_63c2_44e8_8a7b_b10e17dc6fa6.slice/cri-containerd-9b23430f3f530e13da69c14de3ec474f7d5a3abad4e5cdc3a619ad1ef591642d.scope/cgroup.procs:
										no such file or directory: unknown
Internal error occurred: 
	error executing command in container: 
		failed to exec in container: 
			failed to start exec "e9929e03fa844acf65d5ee5eb05f3a886d315a33731dd53fdead42f50fce0f62": 
				OCI runtime exec failed: 
					exec failed: 
						unable to start container process: 
							error adding pid 300725 to cgroups: 
								failed to write 300725: 
									openat2 /sys/fs/cgroup/blkio/kubepods.slice/kubepods-besteffort.slice/kubepods-besteffort-pod7ab5c773_63c2_44e8_8a7b_b10e17dc6fa6.slice/cri-containerd-9b23430f3f530e13da69c14de3ec474f7d5a3abad4e5cdc3a619ad1ef591642d.scope/cgroup.procs:
										no such file or directory: unknown
Unable to open a shell to the container (none of the shell commmands succeeded)

部署yaml如下

kind: DaemonSet
metadata:
  name: huatuo-bamai
  namespace: sre-common
spec:
  selector:
    matchLabels:
      app: huatuo-bamai
  template:
    metadata:
      labels:
        app: huatuo-bamai
    spec:
      nodeSelector:
        #huatuo: "true"
        kubernetes.io/hostname: ip-xxx.ap-southeast-1.compute.internal
      hostNetwork: true
      hostPID: true
      containers:
        - name: huatuo-bamai
          image: huatuo/huatuo-bamai:latest
          securityContext:
            privileged: true
          env:
            # - name: ELASTICSEARCH_HOST
            #   value: "elasticsearch.monitoring"
            - name: ELASTIC_PASSWORD
              value: "huatuo-bamai"
            - name: RUN_PATH
              value: "/home/huatuo-bamai"
          command:  ["/bin/sh", "/run/huatuo/run.sh"]
          volumeMounts:
            - name: sys
              mountPath: /sys
            - name: proc
              mountPath: /proc
            - name: run
              mountPath: /run
            - name: etc
              mountPath: /etc
              readOnly: true
            - name: var
              mountPath: /var
              readOnly: true
            - name: huatuo-conf
              mountPath: /home/huatuo-bamai/conf/huatuo-bamai.conf
              subPath: huatuo-bamai.conf
            - name: huatuo-script
              mountPath: /run/huatuo/run.sh
              subPath: run.sh
      resources:
        limits:
          cpu: 200m
          memory: 300Mi
        requests:
          cpu: 100m
          memory: 200Mi
      volumes:
        - name: sys
          hostPath:
            path: /sys
        - name: proc
          hostPath:
            path: /proc
        - name: run
          hostPath:
            path: /run
        - name: etc
          hostPath:
            path: /etc
        - name: var
          hostPath:
            path: /var
        - name: huatuo-conf
          configMap:
            name: huatuo-bamai-config
            items:
              - key: huatuo-bamai.conf
                path: huatuo-bamai.conf
        - name: huatuo-script
          configMap:
            name: huatuo-bamai-scripts
            items:
              - key: run.sh
                path: run.sh
      tolerations:
        - operator: Exists
      restartPolicy: Always
      dnsPolicy: ClusterFirstWithHostNet```

似乎和挂载/sys目录有关，如果去掉挂载/sys的话是可以登入shell的，但是huatuo启动后会有类似这样的报错
level="error" msg="start tracing memory_free: attach: attach tracepoint: can't attach tracepoint vmscan/mm_vmscan_direct_reclaim_end in TracePoint(tracepoint_try_to_free_pages_end)#62: neither debugfs nor tracefs are mounted" func="huatuo-bamai/pkg/tracing.(*EventTracing).doStart" file="tracing.go:88"

[fanzu8]

我本地环境用的cgroup v2，测试了执行exec是可以正常进容器的，应该跟cgroup v1 下挂/sys有关，我看看怎么修复