Add v3 endpoint for identity in catalog for sample data

[Haneef Ali]

This is a wish list.

Since we are moving to v3, it is better to add v3 endpoint in sample_data.sh. We still have only v2.0 endpoint.I don't think keystoenclient will be affected since it doesn't use the endpoint from catalog, but relies on version discovery

[Morgan Fainberg (mdrnstm)]

sample_data.sh update is a reasonable request (or figure out a better way to convey the sample data for Keystone)

[Dolph Mathews (dolph)]

Instead of putting versioned endpoints in the catalog, drop the versioned endpoints in sample_data.sh in favor of unversioned ones (for identity, at least).

[Haneef Ali (haneef)]

Yes. that should work. I'm not sure about any 3rd partly libraries such as Jcloud. If they complain we should ask them to use discovery to find out the version

[Kenjiro Kosaka (k-kosaka)]

I try to replace unversioned one from versioned one.

I have two questions:

1. It seems no test for the script. Is there any rule about shell script testing?
2. Is there a need to consider about 3rd party software?

[Dolph Mathews (dolph)]

1. sample_data.sh serves as documentation, so it's only tested as such.
2. I'm not sure what your concern as, as sample_data.sh doesn't preclude the use of anything else.

[Kenjiro Kosaka (k-kosaka)]

1. I'll run the script and check created correct endpoint as test.
2. I understood that changing sample_data.sh is not affects to 3rd party libraries directly.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/130669

[Steve Martinelli (stevemar)]

Patch https://review.openstack.org/#/c/290897/ fixed this issue

[OpenStack Infra (hudson-openstack)]

Change abandoned by Steve Martinelli (<email address hidden>) on branch: master
Review: https://review.openstack.org/130669
Reason: Abandoning since no change in many months. Also https://review.openstack.org/#/c/290897/ fixed the issue.