kubeadm v1.18.2 和 crio 版本 1.18.2 无法从 Centos7 / RH7 上的私有 repo 启动主节点

Question

说明

我对 Kubernetes 比较陌生。我可以在使用默认套接字 (/var/run/dockershim.sock) 时运行我的集群，但是当我尝试使用 crio 套接字从我的私人仓库中提取图像时，我注意到速度甚至无法与之相比。

我正在尝试将所有节点配置为使用 crio.socket，但我无法使用此套接字启动主节点。

我遵循了 kubernetes Configuring each kubelet in your cluster using kubeadm 和 git 文档 cri-o 的文档。

不幸的是，我无法让它工作，因为它似乎忽略了私人回购标志。

重现问题的步骤：

1. 使用以下初始化启动主节点（主节点）（使用私有存储库）：

kubeadm init \
        --upload-certs \
        --cri-socket=/var/run/crio/crio.sock \ 
        --node-name=my_node_name \
        --image-repository=my.private.repo \
        --pod-network-cidr=10.96.0.0/16 \
        --kubernetes-version=v1.18.2 \
        --control-plane-endpoint=ip:6443 \
        --apiserver-cert-extra-sans=ip \
        --apiserver-advertise-address=ip

1. 以 root 身份或使用 sudo 运行：journalctl -xeu crio -f
2. 在调试或信息模式下观察下面的日志示例

描述您收到的结果：

调试模式下来自 crio 的日志示例：

Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043499089+02:00" level=debug msg="Trying to access \"k8s.gcr.io/pause:3.2\"" file="docker/docker_image_src.go:68"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043547722+02:00" level=debug msg="Credentials not found" file="config/config.go:123"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043576124+02:00" level=debug msg="Using registries.d directory /etc/containers/registries.d for sigstore configuration" file="docker/lookaside.go:51"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043706369+02:00" level=debug msg=" Using \"default-docker\" configuration" file="docker/lookaside.go:169"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043736378+02:00" level=debug msg=" No signature storage configuration found for k8s.gcr.io/pause:3.2" file="docker/lookaside.go:174"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043769424+02:00" level=debug msg="Looking for TLS certificates and private keys in /etc/docker/certs.d/k8s.gcr.io" file="tlsclientconfig/tlsclientconfig.go:21"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.043858410+02:00" level=debug msg="GET https://k8s.gcr.io/v2/" file="docker/docker_client.go:516"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.046154250+02:00" level=debug msg="Ping https://k8s.gcr.io/v2/ err Get \"https://k8s.gcr.io/v2/\": dial tcp 10.254.3.15:443: connect: connection refused (&url.Error{Op:\"Get\", URL:\"https://k8s.gcr.io/v2/\", Err:(*net.OpError)(0xc00084d5e0)})" file="docker/docker_client.go:708"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.046239456+02:00" level=debug msg="GET https://k8s.gcr.io/v1/_ping" file="docker/docker_client.go:516"
Jun 30 20:03:45 hostname crio[6693]: time="2020-06-30 20:03:45.048653448+02:00" level=debug msg="Ping https://k8s.gcr.io/v1/_ping err Get \"https://k8s.gcr.io/v1/_ping\": dial tcp 10.254.3.15:443: connect: connection refused (&url.Error{Op:\"Get\", URL:\"https://k8s.gcr.io/v1/_ping\", Err:(*net.OpError)(0xc0006b0690)})" file="docker/docker_client.go:735"

描述您预期的结果：

使用 crio 套接字启动节点

您认为重要的其他信息（例如问题只是偶尔发生）：

如果我使用默认套接字启动节点，例如：

# kubeadm init \
        --upload-certs \
        --cri-socket=/var/run/dockershim.sock \ 
        --node-name=my_node_name \
        --image-repository=my.private.repo \
        --pod-network-cidr=10.96.0.0/16 \
        --kubernetes-version=v1.18.2 \
        --control-plane-endpoint=ip:6443 \
        --apiserver-cert-extra-sans=ip \
        --apiserver-advertise-address=ip
W0630 20:24:33.223266   29033 configset.go:202] WARNING: kubeadm cannot validate component configs for API groups [kubelet.config.k8s.io kubeproxy.config.k8s.io]
[init] Using Kubernetes version: v1.18.2
[preflight] Running pre-flight checks
[preflight] Pulling images required for setting up a Kubernetes cluster
[preflight] This might take a minute or two, depending on the speed of your internet connection
[preflight] You can also perform this action in beforehand using 'kubeadm config images pull'
[kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env"
[kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml"
[kubelet-start] Starting the kubelet
[certs] Using certificateDir folder "/etc/kubernetes/pki"
[certs] Using existing ca certificate authority
[certs] Using existing apiserver certificate and key on disk
[certs] Using existing apiserver-kubelet-client certificate and key on disk
[certs] Using existing front-proxy-ca certificate authority
[certs] Using existing front-proxy-client certificate and key on disk
[certs] Using existing etcd/ca certificate authority
[certs] Using existing etcd/server certificate and key on disk
[certs] Using existing etcd/peer certificate and key on disk
[certs] Using existing etcd/healthcheck-client certificate and key on disk
[certs] Using existing apiserver-etcd-client certificate and key on disk
[certs] Using the existing "sa" key
[kubeconfig] Using kubeconfig folder "/etc/kubernetes"
[kubeconfig] Using existing kubeconfig file: "/etc/kubernetes/admin.conf"
[kubeconfig] Using existing kubeconfig file: "/etc/kubernetes/kubelet.conf"
[kubeconfig] Using existing kubeconfig file: "/etc/kubernetes/controller-manager.conf"
[kubeconfig] Using existing kubeconfig file: "/etc/kubernetes/scheduler.conf"
[control-plane] Using manifest folder "/etc/kubernetes/manifests"
[control-plane] Creating static Pod manifest for "kube-apiserver"
[control-plane] Creating static Pod manifest for "kube-controller-manager"
W0630 20:24:35.839949   29033 manifests.go:225] the default kube-apiserver authorization-mode is "Node,RBAC"; using "Node,RBAC"
[control-plane] Creating static Pod manifest for "kube-scheduler"
W0630 20:24:35.841420   29033 manifests.go:225] the default kube-apiserver authorization-mode is "Node,RBAC"; using "Node,RBAC"
[etcd] Creating static Pod manifest for local etcd in "/etc/kubernetes/manifests"
[wait-control-plane] Waiting for the kubelet to boot up the control plane as static Pods from directory "/etc/kubernetes/manifests". This can take up to 4m0s
[apiclient] All control plane components are healthy after 11.003647 seconds
[upload-config] Storing the configuration used in ConfigMap "kubeadm-config" in the "kube-system" Namespace
[kubelet] Creating a ConfigMap "kubelet-config-1.18" in namespace kube-system with the configuration for the kubelets in the cluster
[upload-certs] Storing the certificates in Secret "kubeadm-certs" in the "kube-system" Namespace
[upload-certs] Using certificate key:
key
[mark-control-plane] Marking the node hostname as control-plane by adding the label "node-role.kubernetes.io/master=''"
[mark-control-plane] Marking the node hostname as control-plane by adding the taints [node-role.kubernetes.io/master:NoSchedule]
[bootstrap-token] Using token: token
[bootstrap-token] Configuring bootstrap tokens, cluster-info ConfigMap, RBAC Roles
[bootstrap-token] configured RBAC rules to allow Node Bootstrap tokens to get nodes
[bootstrap-token] configured RBAC rules to allow Node Bootstrap tokens to post CSRs in order for nodes to get long term certificate credentials
[bootstrap-token] configured RBAC rules to allow the csrapprover controller automatically approve CSRs from a Node Bootstrap Token
[bootstrap-token] configured RBAC rules to allow certificate rotation for all node client certificates in the cluster
[bootstrap-token] Creating the "cluster-info" ConfigMap in the "kube-public" namespace
[kubelet-finalize] Updating "/etc/kubernetes/kubelet.conf" to point to a rotatable kubelet client certificate and key
[addons] Applied essential addon: CoreDNS
[addons] Applied essential addon: kube-proxy

Your Kubernetes control-plane has initialized successfully!

To start using your cluster, you need to run the following as a regular user:

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config

You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
  https://kubernetes.io/docs/concepts/cluster-administration/addons/

You can now join any number of the control-plane node running the following command on each as root:

  kubeadm join ip:6443 --token token \
    --discovery-token-ca-cert-hash sha256:hash \
    --control-plane --certificate-key key

Please note that the certificate-key gives access to cluster sensitive data, keep it secret!
As a safeguard, uploaded-certs will be deleted in two hours; If necessary, you can use
"kubeadm init phase upload-certs --upload-certs" to reload certs afterward.

Then you can join any number of worker nodes by running the following on each as root:

kubeadm join ip:6443 --token token \
    --discovery-token-ca-cert-hash sha256:hash

如果我使用 crio 套接字启动节点：

# kubeadm init \
        --upload-certs \
        --cri-socket=/var/run/crio/crio.sock \ 
        --node-name=my_node_name \
        --image-repository=my.private.repo \
        --pod-network-cidr=10.96.0.0/16 \
        --kubernetes-version=v1.18.2 \
        --control-plane-endpoint=ip:6443 \
        --apiserver-cert-extra-sans=ip \
        --apiserver-advertise-address=ip
W0630 20:32:33.827957    2916 configset.go:202] WARNING: kubeadm cannot validate component configs for API groups [kubelet.config.k8s.io kubeproxy.config.k8s.io]
[init] Using Kubernetes version: v1.18.2
[preflight] Running pre-flight checks
[preflight] Pulling images required for setting up a Kubernetes cluster
[preflight] This might take a minute or two, depending on the speed of your internet connection
[preflight] You can also perform this action in beforehand using 'kubeadm config images pull'
[kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env"
[kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml"
[kubelet-start] Starting the kubelet
[certs] Using certificateDir folder "/etc/kubernetes/pki"
[certs] Generating "ca" certificate and key
[certs] Generating "apiserver" certificate and key
[certs] apiserver serving cert is signed for DNS names [hostname kubernetes kubernetes.default kubernetes.default.svc kubernetes.default.svc.cluster.local] and IPs [10.96.0.1 10.96.134.57 10.96.134.57 10.96.134.57]
[certs] Generating "apiserver-kubelet-client" certificate and key
[certs] Generating "front-proxy-ca" certificate and key
[certs] Generating "front-proxy-client" certificate and key
[certs] Generating "etcd/ca" certificate and key
[certs] Generating "etcd/server" certificate and key
[certs] etcd/server serving cert is signed for DNS names [hostname localhost] and IPs [10.96.134.57 127.0.0.1 ::1]
[certs] Generating "etcd/peer" certificate and key
[certs] etcd/peer serving cert is signed for DNS names [hostname localhost] and IPs [10.96.134.57 127.0.0.1 ::1]
[certs] Generating "etcd/healthcheck-client" certificate and key
[certs] Generating "apiserver-etcd-client" certificate and key
[certs] Generating "sa" key and public key
[kubeconfig] Using kubeconfig folder "/etc/kubernetes"
[kubeconfig] Writing "admin.conf" kubeconfig file
[kubeconfig] Writing "kubelet.conf" kubeconfig file
[kubeconfig] Writing "controller-manager.conf" kubeconfig file
[kubeconfig] Writing "scheduler.conf" kubeconfig file
[control-plane] Using manifest folder "/etc/kubernetes/manifests"
[control-plane] Creating static Pod manifest for "kube-apiserver"
[control-plane] Creating static Pod manifest for "kube-controller-manager"
W0630 20:32:37.829806    2916 manifests.go:225] the default kube-apiserver authorization-mode is "Node,RBAC"; using "Node,RBAC"
[control-plane] Creating static Pod manifest for "kube-scheduler"
W0630 20:32:37.830826    2916 manifests.go:225] the default kube-apiserver authorization-mode is "Node,RBAC"; using "Node,RBAC"
[etcd] Creating static Pod manifest for local etcd in "/etc/kubernetes/manifests"
[wait-control-plane] Waiting for the kubelet to boot up the control plane as static Pods from directory "/etc/kubernetes/manifests". This can take up to 4m0s
[kubelet-check] It seems like the kubelet isn't running or healthy.
[kubelet-check] The HTTP call equal to 'curl -sSL http://localhost:10248/healthz' failed with error: Get http://localhost:10248/healthz: dial tcp [::1]:10248: connect: connection refused.
[kubelet-check] It seems like the kubelet isn't running or healthy.
[kubelet-check] The HTTP call equal to 'curl -sSL http://localhost:10248/healthz' failed with error: Get http://localhost:10248/healthz: dial tcp [::1]:10248: connect: connection refused.
Unfortunately, an error has occurred:
                timed out waiting for the condition

        This error is likely caused by:
                - The kubelet is not running
                - The kubelet is unhealthy due to a misconfiguration of the node in some way (required cgroups disabled)

        If you are on a systemd-powered system, you can try to troubleshoot the error with the following commands:
                - 'systemctl status kubelet'
                - 'journalctl -xeu kubelet'

        Additionally, a control plane component may have crashed or exited when started by the container runtime.
        To troubleshoot, list all containers using your preferred container runtimes CLI.

        Here is one example how you may list all Kubernetes containers running in cri-o/containerd using crictl:
                - 'crictl --runtime-endpoint /var/run/crio/crio.sock ps -a | grep kube | grep -v pause'
                Once you have found the failing container, you can inspect its logs with:
                - 'crictl --runtime-endpoint /var/run/crio/crio.sock logs CONTAINERID'

error execution phase wait-control-plane: couldn't initialize a Kubernetes cluster
To see the stack trace of this error execute with --v=5 or higher

我可以看到 localhost 正在监听 10248 端口：

# curl -sSL http://localhost:10248/healthz
ok

crio 套接字示例（如文档中所述）：

# curl -v --unix-socket /var/run/crio/crio.sock http://localhost/info | jq
* About to connect() to localhost port 80 (#0)
*   Trying /var/run/crio/crio.sock...
* Failed to set TCP_KEEPIDLE on fd 3
* Failed to set TCP_KEEPINTVL on fd 3
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0* Connected to localhost (/var/run/crio/crio.sock) port 80 (#0)
> GET /info HTTP/1.1
> User-Agent: curl/7.29.0
> Host: localhost
> Accept: */*
>
< HTTP/1.1 200 OK
< Content-Type: application/json
< Date: Tue, 30 Jun 2020 18:36:35 GMT
< Content-Length: 240
<
{ [data not shown]
100   240  100   240    0     0   144k      0 --:--:-- --:--:-- --:--:--  234k
* Connection #0 to host localhost left intact
{
  "storage_driver": "overlay2",
  "storage_root": "/var/lib/containers/storage",
  "cgroup_driver": "systemd",
  "default_id_mappings": {
    "uids": [
      {
        "container_id": 0,
        "host_id": 0,
        "size": 4294967295
      }
    ],
    "gids": [
      {
        "container_id": 0,
        "host_id": 0,
        "size": 4294967295
      }
    ]
  }
}

kubelet status的输出

# systemctl status kubelet -l
● kubelet.service - kubelet: The Kubernetes Node Agent
   Loaded: loaded (/usr/lib/systemd/system/kubelet.service; enabled; vendor preset: disabled)
  Drop-In: /usr/lib/systemd/system/kubelet.service.d
           └─10-kubeadm.conf
   Active: active (running) since Tue 2020-06-30 20:39:49 CEST; 6s ago
     Docs: https://kubernetes.io/docs/
 Main PID: 8502 (kubelet)
    Tasks: 15
   Memory: 20.1M
   CGroup: /system.slice/kubelet.service
           └─8502 /usr/bin/kubelet --bootstrap-kubeconfig=/etc/kubernetes/bootstrap-kubelet.conf --kubeconfig=/etc/kubernetes/kubelet.conf --config=/var/lib/kubelet/config.yaml --container-runtime=remote --container-runtime-endpoint=/var/run/crio/crio.sock --hostname-override=hostname

Jun 30 20:39:55 hostname kubelet[8502]: I0630 20:39:55.369441    8502 kubelet_node_status.go:294] Setting node annotation to enable volume controller attach/detach
Jun 30 20:39:55 hostname kubelet[8502]: I0630 20:39:55.399015    8502 kubelet_node_status.go:70] Attempting to register node hostname
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.403707    8502 kubelet.go:2267] node "hostname" not found
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.503871    8502 kubelet.go:2267] node "hostname" not found
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.604115    8502 kubelet.go:2267] node "hostname" not found
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.704324    8502 kubelet.go:2267] node "hostname" not found
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.769448    8502 kubelet_node_status.go:92] Unable to register node "hostname" with API server: Post https://ip:6443/api/v1/nodes: dial tcp ip:6443: connect: connection refused
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.805779    8502 kubelet.go:2267] node "hostname" not found
Jun 30 20:39:55 hostname kubelet[8502]: E0630 20:39:55.906014    8502 kubelet.go:2267] node "hostname" not found
Jun 30 20:39:56 hostname kubelet[8502]: E0630 20:39:56.007272    8502 kubelet.go:2267] node "hostname" not found

据我所知，网络错误无关紧要，因为我尚未启动网络容器，因此此时会出现错误。

crio --version的输出：

# crio --version
crio version 1.18.2
Version:       1.18.2
GitCommit:     7f261aeebffed079b4475dde8b9d602b01973d33
GitTreeState:  clean
BuildDate:     2020-06-18T21:05:27Z
GoVersion:     go1.14
Compiler:      gc
Platform:      linux/amd64
Linkmode:      static

kubelet --version的输出：

# kubelet --version
Kubernetes v1.18.2

LinuxOS version的输出：

# cat /etc/redhat-release
Red Hat Enterprise Linux Server release 7.8 (Maipo)

其他环境详细信息（AWS、VirtualBox、物理等）：

安装应用在准系统节点上。

kubelet 文件示例

# cat /etc/default/kubelet
KUBELET_EXTRA_ARGS=--feature-gates="AllAlpha=false,RunAsGroup=true" --container-runtime=remote --cgroup-driver=systemd --container-runtime-endpoint='unix:///var/run/crio/crio.sock' --runtime-request-timeout=5m

更新：我在 github Kubernetes v1.18.2 with crio version 1.18.2 failing to sync with kubelet on RH7 #3915 提出了一张票。看起来有一个错误，因为 cri-o 无法处理远程存储库，并且它正在尝试提取默认存储库 k8s.io。我会在获得更多信息后立即更新票证。

Answer 1

所以问题并不完全是我们最初认为的CRI-O 上的错误（也是CRI-O 开发团队），但如果用户希望使用@987654326，似乎需要应用很多配置@ 作为kubernetes 的CRI 并且还希望使用私人回购。

所以我不会把 CRI-O 的配置放在这里，因为它已经记录在我向团队 Kubernetes v1.18.2 with crio version 1.18.2 failing to sync with kubelet on RH7 #3915 提出的票证上。

应该应用的第一个配置是配置将提取图像的容器的注册表：

$ cat /etc/containers/registries.conf
[[registry]]
prefix = "k8s.gcr.io"
insecure = false
blocked = false
location = "k8s.gcr.io"

[[registry.mirror]]
location = "my.private.repo"

CRI-O 建议将此配置作为标志传递给 kubelet (haircommander/cri-o-kubeadm)，但对我而言，它并不能仅使用此配置。

我回到了 kubernetes 手册，建议不要将标志传递给 kubelet，而是在运行时传递给文件 /var/lib/kubelet/config.yaml。对我来说这是不可能的，因为节点需要从 CRI-O 套接字而不是任何其他套接字开始（参考 Configure cgroup driver used by kubelet on control-plane node）。

所以我设法通过在下面的配置文件示例中传递这个标志来启动并运行它：

$ cat /tmp/config.yaml
apiVersion: kubeadm.k8s.io/v1beta2
kind: InitConfiguration
localAPIEndpoint:
  advertiseAddress: 1.2.3.4
  bindPort: 6443
nodeRegistration:
  criSocket: unix:///var/run/crio/crio.sock
  name: node.name
  taints:
  - effect: NoSchedule
    key: node-role.kubernetes.io/master
---
apiServer:
  timeoutForControlPlane: 4m0s
apiVersion: kubeadm.k8s.io/v1beta2
certificatesDir: /etc/kubernetes/pki
clusterName: kubernetes
controllerManager: {}
dns:
  type: CoreDNS
etcd:
  local:
    dataDir: /var/lib/etcd
controlPlaneEndpoint: 1.2.3.4:6443
imageRepository: my.private.repo
kind: ClusterConfiguration
kubernetesVersion: v1.18.2
networking:
  dnsDomain: cluster.local
  podSubnet: 10.85.0.0/16
  serviceSubnet: 10.96.0.0/12
scheduler: {}
---
apiVersion: kubelet.config.k8s.io/v1beta1
kind: KubeletConfiguration
cgroupDriver: systemd

然后用户只需使用标志--config <file.yml>启动主/工作节点，节点就会成功启动。

希望这里的所有信息对其他人有所帮助。