1、Kubernetes: requesting flag for "kubectl logs" to avoid 5-minute timeout if no stdout/stderr
When running kubectl logs --follow on a pod, after 5 minutes of no stdout/stderr, we received:
$ kubectl --kubeconfig=config --namespace=foo logs --follow foo-oneoff-w8npn --container bar ###################################### # # # /system_tests/test_derp.py (1/4) # # # ###################################### RESULTS: [/system_tests/test_derp/TestDerp] Ran 4 tests in 130.044 s, ALL TESTS PASSED ###################################### # # # /system_tests/test_fuzz.py (2/4) # # # ###################################### error: unexpected EOF
解決方案：
For posterity, the problem here is that we're using haproxy to serve a VIP to balance HA Kubernetes masters.
Specifically we're using HA-Proxy v1.4.21, and we have this in our haproxy cfg:
defaults timeout client 500000 timeout server 500000
2、The connection to the server localhost:8080 was refused - did you specify the right host or port? 錯誤
K8s集群初始化成功后，kubectl get nodes 查看節點信息時報錯
報錯信息：The connection to the server localhost:8080 was refused - did you specify the right host or port?
解決方法：
mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudo chown $(id -u):$(id -g) $HOME/.kube/config
3、master節點的calico組件是0/1 Running狀態，其他worker節點的calico組件是1/1 Running狀態，describe pod發現是
Readiness probe failed: calico/node is not ready: BIRD is not ready: BGP not established with 10.244.0.1,10.244.2.12020-04-13 06:29:59.582 [INFO][682] health.go 156: Number of node(s) with BGP peering established = 0
解決辦法：
修改的calico.yaml，新增兩行：
- name: IP_AUTODETECTION_METHOD
value: "interface=eth0"
value指向從ip a看到的實際網卡名。結果如下：
# Cluster type to identify the deployment type
- name: CLUSTER_TYPE
value: "k8s,bgp"
- name: IP_AUTODETECTION_METHOD
value: "interface=eth0"
# Auto-detect the BGP IP address.
- name: IP
value: "autodetect"
# Enable IPIP
- name: CALICO_IPV4POOL_IPIP
value: "Always"
等一會就正常了。
參考：https://blog.csdn.net/majixiang1996/article/details/105438506/
4、使用kubeadm安裝kubernetes集群etcd一直卡在starting狀態，查看etcd日志報2020-12-08 17:11:10.741954 I | embed: rejected connection from "192.168.100.179:47288" (error "tls: failed to verify client‘s certificate: x509: certificate has expired or is not yet valid", ServerName "") 錯誤
錯誤分析：原因是生成證書的機器時間要比服務器時間快，導致服務器驗證時，證書超出了時間使用范圍。
解決辦法：
1）、服務器和生成證書機器進行時間同步更新（高可用k8s集群需要master節點之間時間同步）。
2）、或者直接調整生成證書的機器時間，小於服務器的時間。
參考：http://www.bubuko.com/infodetail-3670080.html
5、如果使用nfs作為存儲方案所有節點都需要保證安裝好nfs-client
檢查服務器是否已安裝好nfs-utils、rpcbind、libtirpc包。
參考：http://www.bubuko.com/infodetail-3546759.html
6、報錯 cannot allocate memory 或者 no space left on device ，修復K8S內存泄露問題
鏈接：https://www.cnblogs.com/zhangmingcheng/p/14309962.html
7、執行kubectl命令時報錯 error: You must be logged in to the server (Unauthorized)
鏈接：https://www.cnblogs.com/zhangmingcheng/p/14317551.html
8、kubectl delete kube-scheduler-xxxx -n=kube-system 后，這個pod沒有重啟直接沒了，重啟kubelet服務調度pod重啟