Currently, the unified client does not have the ability to show the
"Forced Down" field of a GET /os-services response in microversion 2.11
even though the legacy client can.

This adds a "Forced Down" column to the 'openstack compute service list
--long' command output when microversion 2.11 is used.

Story: 2009115
Task: 43011

Change-Id: I10bc2fedbf0e867a990227962b2b6e60f5681f69

The compute APIs are unfortunately inconsistent with regard to the
response parameters for migrations.

* GET /servers/{server_id}/migrations returns server_uuid
* GET /os-migrations returns instance_uuid

Because the 'Server UUID' column is being specified for parsing the
response from GET /os-migrations, it is always showing as an empty
string to users.

There are a few other mismatches between the column names and the REST
API response fields [1]:

* 'Old Flavor' vs 'old_instance_type_id'
* 'New Flavor' vs 'new_instance_type_id'
* 'Type' vs 'migration_type'

This adds a new list containing the REST API response field names to
pass to utils.get_item_properties so that the responses are correctly
parsed and the client output contains the response data instead of
empty strings.

Story: 2009078
Task: 42890

[1] https://docs.openstack.org/api-ref/compute/?expanded=list-migrations-detail#list-migrations

Change-Id: I8aab60619e0225047f6a1c31e44917ca8fcc799e

Currently only the test_server_migration_list adds the 'Id' and 'Type'
columns to the expected output, so if the
test_server_migration_list_no_options test is run by itself, it fails
as the actual response contains 'Id' and 'Type' but the reference does
not. This example run fails:

  tox -epy38 test_server_migration_list_no_options

The reason the tests pass in the gate is because
test_server_migration_list (which adds the 'Id' and 'Type' columns to
self.MIGRATION_COLUMNS) appears to always run before
test_server_migration_list_no_options, so the latter test gets the
benefit of the former test's column additions.

This changes the test class to just include the 'Id' and 'Type' columns
all the time as they are always returned in microversion 2.23 anyway.

Story: 2009079
Task: 42891

Change-Id: I2c97e9f64790b5e978e4d04230d45b8e343b53d4

assertItemsEqual was removed from Python's unittest.TestCase in
Python 3.3 [1][2]. We have been able to use them since then, because
testtools required unittest2, which still included it. With testtools
removing Python 2.7 support [3][4], we will lose support for
assertItemsEqual, so we should switch to use assertCountEqual.

[1] - https://bugs.python.org/issue17866
[2] - https://hg.python.org/cpython/rev/d9921cb6e3cd
[3] - testing-cabal/testtools#286
[4] - testing-cabal/testtools#277

Change-Id: I0bbffbec8889b8b3067cfe17d258f5cb16624f38

There are a couple of other (networking-related) options which accept
paths, none of which insist on a URI-style path. Let's just drop this
bit of complexity before we release the feature.

Change-Id: Ia7f781d82f3f4695b49b55a39abbb6e582cd879c
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>

The syntax of the '--block-device' parameter is complex and easily
screwed up. Allow users to load a block device config from a file. For
example:

  $ openstack server create ... --block-device file:///tmp/bdm.json ...

This should alleviate the pain that is BDMv2 somewhat.

No functional tests are provided since we already have tests for the CSV
style of passing parameters and the unit tests show that the net result
is the same.

Change-Id: I3e3299bbdbbb343863b4c14fb4d9196ff3e1698d
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>

This mostly reuses the existing tests for '--block-device-mapping',
which can hopefully be removed at some point in the future.

This highlights two issues with the implementation of this option.
Firstly, the 'boot_index' parameter is not required so don't mandate it.
Secondly, and more significantly, we were defaulting the destination
type for the 'image' source type to 'local'. Nova only allows you to
attach a single image to local mapping [1], which means this default
would only make sense if you were expecting users to use the
'--block-device' option exclusively and omit the '--image' option. This
is the *less common* case so this is a bad default. Default instead to a
destination type of 'volume' like everything else, and require users
specifying '--block-device' alone to pass 'destination_type=local'
explicitly.

[1] https://github.com/openstack/nova/blob/c8a6f8d2e/nova/block_device.py#L193-L206

Change-Id: I1718be965f57c3bbdb8a14f3cfac967dd4c55b4d
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>

It's been long enough. Time to remove this.

Change-Id: I37ef09eca0db9286544a4b0bb33f845311baa9b2
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>

These options are not supported from Nova API microversion 2.25 and
above. This can be a source of confusion. Start warning, with an eye on
erroring out in the future.

Change-Id: I53f27eb3e3c1a84d0d77a1672c008d0e8bb8536f
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>