Description of problem: kube-controller-manager ignores --master argument Version-Release number of selected component (if applicable): [root@f23 kubernetes]# rpm -qa | grep kubernetes sort kubernetes-1.2.0-0.13.alpha6.gitf0cd09a.fc23.x86_64 kubernetes-client-1.2.0-0.13.alpha6.gitf0cd09a.fc23.x86_64 kubernetes-master-1.2.0-0.13.alpha6.gitf0cd09a.fc23.x86_64 kubernetes-node-1.2.0-0.13.alpha6.gitf0cd09a.fc23.x86_64 How reproducible: Always Steps to Reproduce: After setting up system like you normally would do (create cert etc..) observe the kube-controller manager service ignores the '--master' arg on the command line. You can repro with this command: ``` /usr/bin/kube-controller-manager --logtostderr=true --v=0 --master=http://127.0.0.1:8080 --service_account_private_key_file=/etc/pki/kube-apiserver/serviceaccount.key ``` I set up my systems using the lines in this ansible playbook: https://github.com/dustymabe/vagrantdirs/blob/master/f23/playbook.yml#L62 Actual results: The following is the message you receive. Note the "nor --master was specified" log message, which indicates it didn't recognize the --master=http://127.0.0.1:8080 argument we provided: ``` [root@f23 kubernetes]# /usr/bin/kube-controller-manager --logtostderr=true --v=0 --master=http://127.0.0.1:8080 --service_account_private_key_file=/etc/pki/kube-apiserver/serviceaccount.key W0307 20:07:44.007893 13698 client_config.go:352] Neither --kubeconfig nor --master was specified. Using the inClusterConfig. This might not work. W0307 20:07:44.008076 13698 client_config.go:357] error creating inClusterConfig, falling back to default config: %vunable to load in-cluster configuration, KUBERNETES_SERVICE_HOST and KUBERNETES_SERVICE_PORT must be defined I0307 20:07:44.008618 13698 plugins.go:71] No cloud provider specified. I0307 20:07:44.010789 13698 replication_controller.go:185] Starting RC Manager I0307 20:07:44.012329 13698 nodecontroller.go:134] Sending events to api server. E0307 20:07:44.012490 13698 controllermanager.go:212] Failed to start service controller: ServiceController should not be run without a cloudprovider. I0307 20:07:44.012519 13698 controllermanager.go:225] allocate-node-cidrs set to false, node controller not creating routes I0307 20:07:44.023177 13698 controllermanager.go:258] Starting extensions/v1beta1 apis I0307 20:07:44.023323 13698 controllermanager.go:260] Starting horizontal pod controller. I0307 20:07:44.023500 13698 controllermanager.go:274] Starting daemon set controller I0307 20:07:44.023730 13698 controllermanager.go:280] Starting job controller I0307 20:07:44.023833 13698 controller.go:180] Starting Daemon Sets controller manager ``` As a result we see this when trying to create a pod: ``` [root@f23 ~]# kubectl create -f /tmp/busybox.yaml Error from server: error when creating "/tmp/busybox.yaml": pods "busybox" is forbidden: no API token found for service account default/default, retry after the token is automatically created and added to the service account ```
I have the same result just by starting kube-controller-manager.service from installed rpm without any configuration file modification.
kubernetes-1.2.0-0.15.alpha6.gitf0cd09a.fc23 has been submitted as an update to Fedora 23. https://bodhi.fedoraproject.org/updates/FEDORA-2016-a89f5ce5f4
kubernetes-1.2.0-0.15.alpha6.gitf0cd09a.fc23 has been pushed to the Fedora 23 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2016-a89f5ce5f4
kubernetes-1.2.0-0.15.alpha6.gitf0cd09a.fc23 has been pushed to the Fedora 23 stable repository. If problems still persist, please make note of it in this bug report.
ult config: unable to load in-cluster configuration, KUBERNETES_SERVICE_HOST and KUBERNETES_SERVICE_PORT must be defined Dec 11 10:35:19 master01 kube-controller-manager: invalid configuration: no configuration has been provided Dec 11 10:35:19 master01 systemd: kube-controller-manager.service: main process exited, code=exited, status=1/FAILURE Dec 11 10:35:19 master01 systemd: Unit kube-controller-manager.service entered failed state. Dec 11 10:35:19 master01 systemd: kube-controller-manager.service failed. Dec 11 10:35:19 master01 systemd: kube-controller-manager.service holdoff time over, scheduling restart. Dec 11 10:35:19 master01 systemd: --experimental-cluster-signing-duration=87600h0m0s': /opt/k8s/conf/kube-controller-manager.env Dec 11 10:35:19 master01 kube-controller-manager: I1211 10:35:19.911712 124830 serving.go:319] Generated self-signed cert in-memory Dec 11 10:35:19 master01 kube-controller-manager: W1211 10:35:19.911767 124830 client_config.go:541] Neither --kubeconfig nor --master was specified. Using the inClusterConfig. This might not work. Dec 11 10:35:19 master01 kube-controller-manager: W1211 10:35:19.911774 124830 client_config.go:546] error creating inClusterConfig, falling back to default config: unable to load in-cluster configuration, KUBERNETES_SERVICE_HOST and KUBERNETES_SERVICE_PORT must be defined Dec 11 10:35:19 master01 kube-controller-manager: invalid configuration: no configuration has been provided Dec 11 10:35:19 master01 systemd: kube-controller-manager.service: main process exited, code=exited, status=1/FAILURE Dec 11 10:35:19 master01 systemd: Unit kube-controller-manager.service entered failed state. Dec 11 10:35:19 master01 systemd: kube-controller-manager.service failed. Dec 11 10:35:20 master01 systemd: kube-controller-manager.service holdoff time over, scheduling restart. Dec 11 10:35:20 master01 systemd: --experimental-cluster-signing-duration=87600h0m0s': /opt/k8s/conf/kube-controller-manager.env Dec 11 10:35:20 master01 kube-controller-manager: I1211 10:35:20.527572 124838 serving.go:319] Generated self-signed cert in-memory Dec 11 10:35:20 master01 kube-controller-manager: W1211 10:35:20.527650 124838 client_config.go:541] Neither --kubeconfig nor --master was specified. Using the inClusterConfig. This might not work. Dec 11 10:35:20 master01 kube-controller-manager: W1211 10:35:20.527660 124838 client_config.go:546] error creating inClusterConfig, falling back to default config: unable to load in-cluster configuration, KUBERNETES_SERVICE_HOST and KUBERNETES_SERVICE_PORT must be defined Dec 11 10:35:20 master01 kube-controller-manager: invalid configuration: no configuration has been provided Dec 11 10:35:20 master01 systemd: kube-controller-manager.service: main process exited, code=exited, status=1/FAILURE Dec 11 10:35:20 master01 systemd: Unit kube-controller-manager.service entered failed state. Dec 11 10:35:20 master01 systemd: kube-controller-manager.service failed. Dec 11 10:35:20 master01 systemd: kube-controller-manager.service holdoff time over, scheduling restart. Dec 11 10:35:20 master01 systemd: start request repeated too quickly for kube-controller-manager.service Dec 11 10:35:20 master01 systemd: Unit kube-controller-manager.service entered failed state. Dec 11 10:35:20 master01 systemd: kube-controller-manager.service failed. [root@master01 ansible]# cat /etc/redhat-release CentOS Linux release 7.6.1810 (Core) [root@master01 ansible]# hi, when i install the kubernetes via binary,i got the same bug. the install version is kubernetes-server-linux-amd64-1.16.tar.gz and my system environment is CentOS Linux release 7.6.1810 (Core).