_create_subnet mehod in scenario manager fails if cidr is included in **kwargs

[Xavi León]

We have an internal test which passes a specific cdir to the private method tempest.scenario.manager._create_subnet(**kwargs). Internally, this method creates a dict with some fields already populated and includes the argument kwargs:

subnet = dict(
                name=data_utils.rand_name(namestart),
                network_id=network.id,
                tenant_id=network.tenant_id,
                cidr=str_cidr,
                ip_version=ip_version,
                **kwargs
            )

If kwargs contains the key cidr, this dict creation fails with:

Traceback (most recent call last):
TypeError: type object got multiple values for keyword argument 'cidr'

Changing the above subnet dict creation by a two-step create/update solves the problem:

subnet = dict(
                name=data_utils.rand_name(namestart),
                network_id=network.id,
                tenant_id=network.tenant_id,
                cidr=str_cidr,
                ip_version=ip_version
            )
subnet.update(**kwargs)

I have a patch ready, but wanted to share it in case I missed something.

[Mauro S M Rodrigues (maurorodrigues)]

Hey Xavi, thanks for reporting this! Yeah the bug exists, you're not missing anything.
Please assign yourself and submit the patch if you're still willing to work on the fix

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/151692

[OpenStack Infra (hudson-openstack)]

Change abandoned by Ken'ichi Ohmichi (<email address hidden>) on branch: master
Review: https://review.openstack.org/151692
Reason: As http://lists.openstack.org/pipermail/openstack-dev/2016-May/096204.html , we start abandoning patches have gotten old without any updating after negative feedback. Please restore if necessary to restart this again.

[Ken'ichi Ohmichi (oomichi)]

The method is not public/stable interface, and it is difficult to apply requirements of companies' internal tests to upstream dev.
This needs to be marked as "Won't fix" then.

[Xavi León (xleon)]

Sounds good to me. However, if internal tests uncover upstream bugs/not reliable code, I don't see the problem in merging the fix (even more if it doesn't impact other upstream tests). Anyway, I'm ok with abandoning it. Thanks for taking the time to review and update the issue.