9b6b8d2762
This fix updates .travis.yml and Makefile for several places: - Remove unneeded `docker pull gcr.io/google_containers/hyperkube-amd64:v1.2.4` (only v1.3.7 was used) - Use docker to deploy etcd (insteadof downloading etcd binary). - Merge `make testk8s` and `make testk8s-setup` (no need to have multiple targets for a couple of tests) - Set version of etcd and kubernetes in .travis.yml (so that it is easy to update new version in the future) Signed-off-by: Yong Tang <yong.tang.github@outlook.com> |
||
|---|---|---|
| .. | ||
| 00_run_k8s.sh | ||
| 10_setup_kubectl.sh | ||
| 15_run_skydns.sh | ||
| 20_setup_k8s_services.sh | ||
| kill_all_containers.sh | ||
| nginx-rc.yml | ||
| README.md | ||
| start_k8s_with_services.sh | ||
Test scripts to automate kubernetes startup
Requirements: docker curl
The scripts in this directory startup kubernetes with docker as the container runtime.
After starting kubernetes, a couple of kubernetes services are started to allow automatic
testing of CoreDNS with kubernetes. The kubernetes integration tests in test/kubernetes_test.go depend on having some sample services running. The scripts in this folder
automate the launch of kubernetes and the creation of the expected sample services.
To start up kubernetes and launch some sample services,
run the script start_k8s_with_services.sh.
$ ./start_k8s_with_services.sh
Alternatively, the individual scripts may be run independently as needed:
$ ./00_run_k8s.sh && ./10_setup_kubectl.sh && ./20_setup_k8s_services.sh
After running the above scripts, kubernetes will be running on the localhost with the following services exposed:
~~ NAMESPACE NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE default kubernetes 10.0.0.1 443/TCP 48m demo mynginx 10.0.0.168 80/TCP 9m demo webserver 10.0.0.28 80/TCP 2m test mynginx 10.0.0.4 80/TCP 2m test webserver 10.0.0.39 80/TCP 2m ~~
Kubernetes and all running containers can be uncerimoniously stopped by
running the kill_all_containers.sh script.
$ ./kill_all_containers.sh