Be More Tolerant of Network Timeouts

[ALRubinger]

[kind/Enhancement]

Which functionality do you think we should update/improve?

So I went to use odo to connect to an OpenShift cluster on another continent. Here's what happened:

[alr@tubescreamer ~]$ odo login %URL% -u user18 -p %PASSWORD%
Login successful.

You have one project on this server: "user18"

Using project "user18".

Cool! Now let's make an app.

[alr@tubescreamer ~]$ odo app create test
 ✗  Unable to connect to OpenShift cluster, is it down?

Allright; so we've hit a connection timeout. But the problem is in the messaging; this could be improved. I suggest the following fixes to the error message:

1. Change to say "✗ Unable to connect to OpenShift cluster, is it down? Perhaps you're hitting a network timeout?"
2. Add some context to suggest a workaround in case the timeout is not sufficient. For instance telling the user to run odo utils config set Timeout XX where XX is a new timeout value in seconds, say, 20, should do the trick.

Additionally, how about increasing the default timeout? I believe 1 second is the default? Surely we can be more permissive, yeah?

[cdrage]

More notes / suggestions:

• If the connection is taking more than 3 seconds / too longer to connect to OpenShift, we should use a spinner or at least indicate to the user that the connection is delayed and taking a while...
• Ex: If > 3 seconds, Do: Spinner + INFO, Else: Continue

[jorgemoralespou]

Additionally to this, --skip-connection-check prevented this error from happening.

[kadel]

/close

[openshift-ci-robot]

@kadel: Closing this issue.

In response to this:

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.