Don't die if the "docker rmi" fails

Slightly neuters #8955, but we haven't had a build succeed in a while
for whatever reason. (I checked on Jenkins and the images in the build
log where deletion was attempted were actually deleted, so I think
this is primarily an exit code / API issue of some sort.)

Fixes nfs-server security issue in image

make the Container.Image field optional in v1

Fix permissions in e2e volume tests.

Disable --patch for kubectl update

Fix proxying of URLs that end in "/" in the pod proxy subresource

Stop generating v1beta1,2 swagger spec

Release 0.18.0

Add event about Node when cannot assign CIDR.

remove ReplicationControllerSpec.TemplateRef

Update ContainerState.Termination to ContainerState.Terminated

Move SSH code into a common util package.  Add support for SSH tunnels.

Add explicitly setting the runtime and option to set rkt version

Adjust the Jenkins E2E_REPORT_DIR to something less silly

Start of a fix for complaints on #8852.

Add event generation when create/delete of balancers fails.

Do not set container requests in limit ranger for Kube 1.0

fix TestGetEtcdVersion_NotListening

Optionally use a github api token when compiling relase-notes

Support glob wildcards for gcr.io credentials

Increase timeout on biggest contributor to hack/test-integration flakes

Fix the race when monit is used to monitor services which are also

expose: Use separate parameters for default and custom name

TTL is not preserved automatically during edit

Revert "Revert "Wake up rcs when pods get DeletionFinalStateUnknown t…

[Harmless] K8petstore shell update, separated from the larger PR

Create a system container

Added e2e test case which triggers kernel panic on a node.

Use spew for more informative debug output

let the coreos bare_metal_offline.md point to examples/ rather than having its own examples

update fedora_ansible_config.md to v1beta3