make more intelligent kubeconfig merge #953
Conversation
|
Yeah ----- Original Message -----
|
| glog.Fatalf("%v\n", err) | ||
| } | ||
| // check to see if we're already signed in. If so, simply make sure that .kubeconfig has that information | ||
| if userFullName, err := whoami(clientCfg); err == nil { |
There was a problem hiding this comment.
I think if the --username flag was provided we may want to make sure we are already signed in with the same username specified by the flag. That would allow users to simply switch between accounts by doing
openshift ex login -u stark
openshift ex login -u targaryen
It's already in my list btw the osc logout command that would also help in that flow.
|
@deads2k Very nice PR, thanks. |
|
[merge] |
|
continuous-integration/openshift-jenkins/merge SUCCESS (https://ci.openshift.redhat.com/jenkins/job/merge_pull_requests_openshift3/876/) (Image: devenv-fedora_749) |
|
Evaluated for origin up to 5d660bd |
Merged by openshift-bot
always install the jenkins sample template
@fabianofranz This is a better way to merge into .kubeconfig files. It looks for existing cluster and auth info stanzas that already specify the same information as we are using. Then it uses that information to avoid creating unnecessary cruft and maps the names to their appropriate values.
@smarterclayton do we want to push the merging logic upstream?