Rancher 2.x,K8S,使用问题记录（持续更新.)

1、ingress-nginx,修改默认的Nginx上传大小：

使用Rancher的负载均衡，上传1M以上的文件报错，查看ingress-nginx容器，发现配置

client_max_body_size 1m;

解决办法：

创建ingress的时候修改参数：

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  annotations:
    nginx.ingress.kubernetes.io/proxy-body-size: 50m

2、k8s使用cronjob自动备份gitlab

常规情况为使用定时任务运行脚本，运行命令即可

gitlab-rake gitlab:backup:create

k8s中运行的gitlab，使用cronjob定时运行脚本即可：

1、为了可以使用kubectl，需要使用kubeconfig，创建configmap 就叫kubeconfig：

#kubeconfig默认放在/root/.kube/config

kubectl create configmap kubeconfig -n gitlab --from-file=/root/.kube/config

2、创建备份脚本，configmap：

备份脚本如下：

#!/bin/sh

pod_name=$(kubectl get pods -l app=gitlab -o jsonpath={.items[*].metadata.name} -n gitlab --kubeconfig=/etc/kubeconfig/config)

kubectl --kubeconfig=/etc/kubeconfig/config exec $pod_name -n gitlab -- gitlab-rake gitlab:backup:create

导入到configmap

3、挂载两个configmap，需要放到对应路径，kubeconfig放到/etc/kubeconfig/config，备份脚本放到运行路径下即可

4、配置cronjob定时运行挂载的脚本即可

apiVersion: v1
items:

• apiVersion: batch/v1beta1
  kind: CronJob
  metadata:
  name: gitlab2-backup
  namespace: gitlab
  spec:
  concurrencyPolicy: Allow
  failedJobsHistoryLimit: 10
  jobTemplate:
  metadata:
  creationTimestamp: null
  spec:
  template:
  metadata:
  spec:
  containers:
  • command:
    • sh
    • /home/demo.sh
      image: lachlanevenson/k8s-kubectl:v1.17.0
      imagePullPolicy: IfNotPresent
      name: gitlab2-backup
      resources: {}
      stdin: true
      terminationMessagePath: /dev/termination-log
      terminationMessagePolicy: File
      tty: true
      volumeMounts:
    • mountPath: /etc/kubeconfig
      name: vol1
    • mountPath: /home
      name: vol2
      volumes:
  • configMap:
    defaultMode: 420
    name: kubeconfig
    optional: false
    name: vol1
  • configMap:
    defaultMode: 493
    name: backup-config
    optional: false
    name: vol2
    schedule: 25 8 *
    successfulJobsHistoryLimit: 10
    suspend: false
    kind: List
    metadata:
    resourceVersion: ""
    selfLink: ""

3、Rancher新增加节点报错

Runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:docker: network plugin is not ready: cni config uninitialized

查看发现新节点/etc/cni/net.d/下无配置文件，把其他的节点的配置文件拷贝过来即可

10-canal.conflist calico-kubeconfig

虽然节点显示正常了不过运行容器报错：

Failed create pod sandbox: rpc error: code = Unknown desc = [failed to set up sandbox container "75b6c96ee03bcdb754b01c126afb8f77016000a27e1ad7d55bd4d1e31c7889c4" network for pod "demo1111-645996f944-lrkwq": NetworkPlugin cni failed to set up pod "demo1111-645996f944-lrkwq_yj-test" network: failed to find plugin "loopback" in path [/opt/cni/bin], failed to clean up sandbox container "75b6c96ee03bcdb754b01c126afb8f77016000a27e1ad7d55bd4d1e31c7889c4" network for pod "demo1111-645996f944-lrkwq": NetworkPlugin cni failed to teardown pod "demo1111-645996f944-lrkwq_yj-test" network: failed to find plugin "portmap" in path [/opt/cni/bin]] 2 minutes ago

Warning FailedCreatePodSandBox Failed create pod sandbox: rpc error: code = Unknown desc = [failed to set up sandbox container

查看/opt/cni/bin，果然也没有相关文件，同上把其他节点的拷贝过来

可是感觉应该不会莫名的没有安装上，查看rancher system项目，发现kube-system空间下的canal在node3还未成功启动，在查看原来是相关镜像拉取较慢。。镜像拉取成功后（我是手动导的）即一切正常。