Microversion Testing With Tempest¶

Tempest Scope For Microversion Testing¶

APIs microversions for any OpenStack service grow rapidly and testing each and every microversion in Tempest is not feasible and efficient way. Also not every API microversion changes the complete system behavior and many of them only change the API or DB layer to accept and return more data on API.

Tempest is an integration test suite, but not all API microversion testing fall under this category. As a result, Tempest mainly covers integration test cases for microversions, Other testing coverage for microversion should be hosted on project side as functional tests or via Tempest plugin as per project guidelines.

Tempest will cover only integration testing of applicable microversions with below exceptions:

1. Test covers a feature which is important for interoperability. This covers tests requirement from Defcore.

2. Test needed to fill Schema gaps. Tempest validates API responses with defined JSON schema. API responses can be different on each microversion and the JSON schemas need to be defined separately for the microversion. While implementing new integration tests for a specific microversion, there may be a gap in the JSON schemas (caused by previous microversions) implemented in Tempest. Filling that gap while implementing the new integration test cases is not efficient due to many reasons:

   • Hard to review
   • Sync between multiple integration tests patches which try to fill the same schema gap at same time
   • Might delay the microversion change on project side where project team wants Tempest tests to verify the results.

   Tempest will allow to fill the schema gaps at the end of each cycle, or more often if required. Schema gap can be filled with testing those with a minimal set of tests. Those tests might not be integration tests and might be already covered on project side also. This exception is needed because:

   • Allow to create microversion response schema in Tempest at the same time that projects are implementing their API microversions. This will make implementation easier for adding required tests before a new microversion change can be merged in the corresponding project and hence accelerate the development of microversions.
   • New schema must be verified by at least one test case which exercises such schema.

   For example:

   If any projects implemented 4 API microversion say- v2.3, v2.4, v2.5, v2.6 Assume microversion v2.3, v2.4, v2.6 change the API Response which means Tempest needs to add JSON schema for v2.3, v2.4, v2.6. In that case if only 1 or 2 tests can verify all new schemas then we do not need separate tests for each new schemas. In worst case, we have to add 3 separate tests.

3. Test covers service behavior at large scale with involvement of more deep layer like hypervisor etc not just API/DB layer. This type of tests will be added case by case basis and with project team consultation about why it cannot be covered on project side and worth to test in Tempest.

Project Scope For Microversion Testing¶

All microversions testing which are not covered under Tempest as per above section, should be tested on project side as functional tests or as Tempest plugin as per project decision.

Configuration options for Microversion¶

• Add configuration options for specifying test target Microversions. We need to specify test target Microversions because the supported Microversions may be different between OpenStack clouds. For operating multiple Microversion tests in a single Tempest operation, configuration options should represent the range of test target Microversions. New configuration options are:

  • min_microversion
  • max_microversion

  Those should be defined under respective section of each service. For example:

  [compute]
  min_microversion = None
  max_microversion = latest
  

How To Implement Microversion Tests¶

Tempest provides stable interfaces to test API Microversion. For Details, see: API Microversion testing Framework This document explains how to implement Microversion tests using those interfaces.

class BaseTestCase1(api_version_utils.BaseMicroversionTest):

    [..]
@classmethod
def skip_checks(cls):
    super(BaseTestCase1, cls).skip_checks()
    api_version_utils.check_skip_with_microversion(cls.min_microversion,
                                                   cls.max_microversion,
                                                   CONF.compute.min_microversion,
                                                   CONF.compute.max_microversion)

@classmethod
def resource_setup(cls):
    super(BaseTestCase1, cls).resource_setup()
    cls.request_microversion = (
        api_version_utils.select_request_microversion(
            cls.min_microversion,
            CONF.compute.min_microversion))

COMPUTE_MICROVERSION = None

class BaseClient1(rest_client.RestClient):
    api_microversion_header_name = 'X-OpenStack-Nova-API-Version'

def setUp(self):
    super(BaseTestCase1, self).setUp()
    self.useFixture(api_microversion_fixture.APIMicroversionFixture(
        self.request_microversion))

COMPUTE_MICROVERSION = None

class BaseClient1(rest_client.RestClient):
    api_microversion_header_name = 'X-OpenStack-Nova-API-Version'

    def get_headers(self):
        headers = super(BaseClient1, self).get_headers()
        if COMPUTE_MICROVERSION:
            headers[self.api_microversion_header_name] = COMPUTE_MICROVERSION
        return headers

class BaseTestCase1(api_version_utils.BaseMicroversionTest,
                    tempest.test.BaseTestCase):

    [..]


class Test1(BaseTestCase1):
    min_microversion = '2.2'
    max_microversion = '2.9'

    [..]

class Test2(BaseTestCase1):
    min_microversion = '2.10'
    max_microversion = 'latest'

    [..]

Notes about Compute Microversion Tests¶

Some of the compute Microversion tests have been already implemented with the Microversion testing framework. So for further tests only step 4 is needed.

Along with that JSON response schema might need versioning if needed.

Compute service clients strictly validate the response against defined JSON schema and does not allow additional elements in response. So if that Microversion changed the API response then schema needs to be versioned. New JSON schema file needs to be defined with new response attributes and service client methods will select the schema based on requested microversion.

If Microversion tests are implemented randomly meaning not in sequence order(v2.20 tests added and previous Microversion tests are not yet added) then, still schema might need to be version for older Microversion if they changed the response. This is because Nova Microversion includes all the previous Microversions behavior.

For Example:

Implementing the v2.20 Microversion tests before v2.9 and 2.19- v2.20 API request will respond as latest behavior of Nova till v2.20, and in v2.9 and 2.19, server response has been changed so response schema needs to be versioned accordingly.

That can be done by using the get_schema method in below module:

Microversion tests implemented in Tempest¶

• Compute

• 2.1

• 2.2

• 2.10

• 2.20

• 2.25

• 2.32

• 2.37

• 2.42

• 2.47

• 2.48

• Volume

• 3.3