Kubernetes version (use kubectl version):
Client Version: version.Info{Major:"1", Minor:"6", GitVersion:"v1.6.1", GitCommit:"b0b7a323cc5a4a2019b2e9520c21c7830b7f708e", GitTreeState:"clean", BuildDate:"2017-04-03T20:44:38Z", GoVersion:"go1.7.5", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"6", GitVersion:"v1.6.1", GitCommit:"b0b7a323cc5a4a2019b2e9520c21c7830b7f708e", GitTreeState:"clean", BuildDate:"2017-04-03T20:33:27Z", GoVersion:"go1.7.5", Compiler:"gc", Platform:"linux/amd64"}Environment: - Cloud provider or hardware configuration: - OS (e.g. from /etc/os-release): CentOS 7.2 - Kernel (e.g. uname -a): 4.9.0 - Install tools: bare metal
What happened: Add new node to exist cluster, and schedule the pod to new node. The IP of pod is not in range of PodCIDR, and use the docker bridge IP
dl.240.172.hadoop.sjz Ready,master 18d v1.6.1
dl.245.0.hadoop.sjz Ready 36m v1.6.1
dl.245.1.hadoop.sjz Ready 36m v1.6.1
dl.245.11.hadoop.sjz Ready 28m v1.6.1
dl.245.12.hadoop.sjz Ready 28m v1.6.1
dl.245.13.hadoop.sjz Ready 28m v1.6.1
dl.245.14.hadoop.sjz Ready 28m v1.6.1
dl.245.15.hadoop.sjz Ready 28m v1.6.1
dl.245.16.hadoop.sjz Ready 28m v1.6.1
dl.245.17.hadoop.sjz Ready 28m v1.6.1
dl.245.18.hadoop.sjz Ready 28m v1.6.1
dl.245.19.hadoop.sjz Ready 28m v1.6.1
dl.245.2.hadoop.sjz Ready 36m v1.6.1
dl.245.3.hadoop.sjz Ready 36m v1.6.1
dl.245.5.hadoop.sjz Ready 18d v1.6.1
dl.245.6.hadoop.sjz Ready 18d v1.6.1
dl.245.7.hadoop.sjz Ready 18d v1.6.1
dl.245.8.hadoop.sjz Ready 18d v1.6.1
dl.245.9.hadoop.sjz Ready 18d v1.6.1
l22-240-170 Ready 2h v1.6.1
l22-240-171 Ready,master 18d v1.6.1the node follow is new:
dl.245.0.hadoop.sjz Ready 36m v1.6.1
dl.245.1.hadoop.sjz Ready 36m v1.6.1
dl.245.11.hadoop.sjz Ready 28m v1.6.1
dl.245.12.hadoop.sjz Ready 28m v1.6.1
dl.245.13.hadoop.sjz Ready 28m v1.6.1
dl.245.14.hadoop.sjz Ready 28m v1.6.1
dl.245.15.hadoop.sjz Ready 28m v1.6.1
dl.245.16.hadoop.sjz Ready 28m v1.6.1
dl.245.17.hadoop.sjz Ready 28m v1.6.1
dl.245.18.hadoop.sjz Ready 28m v1.6.1
dl.245.19.hadoop.sjz Ready 28m v1.6.1
dl.245.2.hadoop.sjz Ready 36m v1.6.1
dl.245.3.hadoop.sjz Ready 36m v1.6.1and the pod
auto-discovery-4253124847-h1ln1 1/1 Running 0 2h 10.244.124.200 l22-240-171
busybox 1/1 Running 3 15d 10.244.71.2 dl.245.6.hadoop.sjz
gpu-test 1/1 Running 4 15d 10.244.71.7 dl.245.6.hadoop.sjz
gpu-test1 1/1 Running 3 13d 10.244.71.8 dl.245.6.hadoop.sjz
gpu-test1-1-2 1/1 Running 3 10d 10.244.203.12 dl.245.8.hadoop.sjz
gpu-test1-1-3 1/1 Running 2 9d 10.244.71.16 dl.245.6.hadoop.sjz
gpu-test12 1/1 Running 3 11d 10.244.239.73 dl.245.9.hadoop.sjz
nginx-2970154533-wwb3l 1/1 Running 0 17m 172.17.0.2 dl.245.19.hadoop.sjz
test-ssh-1-lxfgq 1/1 Running 0 9d 10.244.33.18 dl.245.5.hadoop.sjz
wanglinhong-mount-7jtv1 1/1 Running 0 2d 10.244.71.24 dl.245.6.hadoop.sjz
wanglinhong-test-wrvxv 1/1 Running 0 4d 10.244.239.94 dl.245.9.hadoop.sjz
wanglinhong-test3-8bnr1 1/1 Running 0 33m 172.17.0.2 dl.245.2.hadoop.sjz
wanglinhong-web-z37nb 1/1 Running 0 1d 10.244.203.31 dl.245.8.hadoop.sjzthe pod wanglinhong-test3-8bnr1 and nginx-2970154533-wwb3l has bean schedule to new nodes . and the ip is 172.17.0.x
kubelet at new node all have the log as follow
Apr 26 11:35:04 dl.245.11.hadoop.sjz kubelet[25363]: I0426 11:35:04.182974 25363 kuberuntime_manager.go:902] updating runtime config through cri with podcidr 10.244.48.0/24
Apr 26 11:35:04 dl.245.11.hadoop.sjz kubelet[25363]: I0426 11:35:04.183206 25363 docker_service.go:277] docker cri received runtime config &RuntimeConfig{NetworkConfig:&NetworkConfig{PodCidr:10.244.48.0/24,},}
Apr 26 11:35:04 dl.245.11.hadoop.sjz kubelet[25363]: I0426 11:35:04.183445 25363 kubelet_network.go:326] Setting Pod CIDR: -> 10.244.x.0/24Anyone who can help me? I try to solve cost two days. THX very much!
This problem has been solved, and I close this issue. The problem is I comment the cni config flag in kubelet.service at the nodes which I add to cluster at a later