Before this change, it was not possible to use the SSH command to
connect to machines provisioned with v2 provisioning. After this change
(and including the changes to Rancher), the CLI will use the new CAPI
client to get the SSH key and config from Rancher for v2 provisioned
machines.
A side effect of this change is the addition of the new `rancher
machines ls` command that lists all machines for the current cluster
context.
Previously, both functions would return 404 with no additional information or recommended actions.
Now, both inform the user that either the current project or cluster is not available, respectively.
It is then suggested the user runs `rancher context switch`. Finally, the error is displayed.
This will aid the user in resolving the error.
Problem:
CLI asks to verify certs on login
No way to wait for a resource to be available
Solution:
Add a skip-verify flag to login
Add wait command to wait for a cluster to be in active state
Problem:
CLI asks to verify certs on login
No way to wait for a resource to be available
Solution:
Add a skip-verify flag to login
Add wait command to wait for a cluster to be in active state
Problem:
When creating the master client, three other clients are needed which
requires multiple calls to the rancher server. This is done synchronously
and causes large delays before the command is actually run
Solution:
Build the clients async to speed up the time it takes to build the
master client
Problem:
When logging in a previous config would be used and if the cluster was
removed an error would occur
Solution:
Previous configs are no longer used for login
