The virt driver FakeDriver used in both the functional and in the unit
test used a global state to configure the host and node names the driver
reports. This was hard to use when more then one compute service is started.
Also global state is dangerous.
It turned out that only a set of unit tests are using multiple nodes per
compute the rest of the tests can simply use host=<hostname>,
nodes=[<hostname>] setup.
So this removes the global state.
Change-Id: I2cf2fcbaebc706f897ce5dfbff47d32117064f9c
This hasn't been needed since we moved console authentication to the
database in Rocky.
Part of blueprint remove-consoleauth
Change-Id: Ie5e7d70f9d6af77edd22756128937f06d0b961a9
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>
Previously, if _ensure_resource_provider encountered any error from the
placement REST API, it would (sometimes log a message and) return None.
Furthermore, a name conflict while creating the provider was treated the
same as a UUID conflict, which would actually result in None being
returned.
With this change set, the error paths that previously returned None now
raise one of the new ResourceProviderRetrievalFailed or
ResourceProviderCreationFailed exceptions; and the name conflict path is
detected and treated as an error condition.
Note: This change set only touches the SchedulerReportClient side of
these error conditions - it makes no attempt to add error handling to
its callers. Case in point, the API samples tests needed fixing because
they were previously running into the name conflict error condition, but
not noticing. As currently implemented, the new exceptions will
percolate up to ComputeManager.update_available_resource_for_node like
any others coming from SchedulerReportClient, where they will be logged
and ignored.
Change-Id: I0c4ca6a81f213277fe7219cb905a805712f81e36
Closes-Bug: #1735430
FilterScheduler is the only scheduler driver that properly integrated
with placement api so this patch changes the functional test environment
to use the FilterScheduler instead of the ChanceScheduler.
The amount of vcpu the SmallFakeDriver reports need to be bumped to 2
as during resize to same host the filter scheduler doubles the
allocation on the host. Simply relying on the 16.0 default allocation
ratio does not work as the max_unit of the vcpu does not use the
allocation ratio to avoid that a single server overallocates the cpu.
The only change in unit test is
test_create_instance_with_oversubscribed_cpu which also loads the
SmallFakeDriver and asserts available resources on it.
Change-Id: I12de2e195022593ea2a3e2894f2c3b5226930d4f
There are quite a few changes here as this is not only handling
uuids for the hypervisor id but it's also a refactor in several
APIs for consistency.
The main changes are detailed in the REST API Version History
doc in this change, but to summarize the changes:
* Hypervisor and service IDs are handled as the UUIDs for those
resources; this is necessary for accurately working with these
resources across multiple cells.
* The 'servers' and 'search' routes are deprecated and folded into
the index and detail methods as query parameters, validated using
json schema.
* The show method will also be able to return the list of servers
hosted on the given hypervisor using the with_servers query
parameter.
* The marker used when paging over lists of hypervisors is the
compute node UUID.
* Using the hypervisor_hostname_pattern query parameter will not
work with paging parameters.
* API reference docs are updated for the detailed changes.
* Functional and unit tests are provided for all changes.
Part of blueprint service-hyper-uuid-in-api
Change-Id: I828350c179df8bcfa4739910abeafaba2f96982b
When there are thousands of compute nodes, it would be slow to get the
whole hypervisor list, and it is bad for user experience to display
thousands of items in a table in horizon. This patch is proposed to
support pagination for hypervisor by adding `limit` and `marker` to
the list api.
Implements blueprint: pagination-for-hypervisor
Change-Id: Ie7f8b5c733b383f3e69fa23188e56257e503b5f7
After this microversion Nova API the GET request to
/v2.1/os-hypervisors/{hypervisor_id} responses cpu_info
as regular JSON object instead of string.
Implements blueprint nova-api-hypervsor-cpu-info
Change-Id: I14db9017f69e70e759cc80d031f76af8be179e45
This patch move the all v2.1 api sample tests under
'functional/api_sample_tests'. Also move sample files under
'doc/api-samples'.
Co-Authored-By: Ed Leafe <ed@leafe.com>
Co-Authored-By: Alex Xu <hejie.xu@intel.com>
Partial-Bug: #1462901
Change-Id: I2b924f2ad7687a23a018a9b658e8acd9e04d7963
Currently v2 and v2.1 have separate functional tests and their
corresponding sample files. As v2 and v2.1 are supposed to be identical,
there is overhead to maintain two set of functional tests and sample files.
We can have one set of tests which can run for both v2 and v2.1.
This commit merges hypervisor functional tests.
Also adding test for hypervisor servers.
In V2 hypervsior API has following extensions
- os-extended-hypervisors
- os-hypervisor-status
In V2.1 above extensions have been merged together in hypervisor plugins
Change-Id: I5ddf3c54dd80a67f71762769d6130db41e772b01
Currently the API sample get '?' for the cpu_info, this is really
confusing, it even does not tell the type of the return.
Change it to be more meaningful.
Change-Id: I791beb7502f47dff646a46c08e818c29f9e6cd79
In I5a580fc323c3809790b4a68a9f8f8129ecdc2cf0 we switched off XML support. In
this review we entirely remove all the testcases and supporting files.
Change-Id: I83827d438753fd3899053dd6e09bc77c997c7406
This is a follow up to 9e770e6213, which
missed the change in a duplicate copy of host_status.
Add regression test to test_virt_drivers.
This requires changes to the API samples, because they were wrong. virt
drivers use convert_version_to_int which converts a version string to a
4 digit number ("1.0" becomes to 1000)
Change-Id: I28ce23509e3c9feae183a49a8fc5bf3c7c601295
Closes-Bug: #1285035
get_available_resources is actually used by the scheduler while
get_host_stats was previously used, and is still used for host
capabilities. This patch makes get_available_resources and
get_host_stats use the same logic to clean up the code. As part of
making them use the same logic, some of the unused data returned from
get_host_stats is changed to be what get_available_resources expects.
This is also in preparation for removing the periodic RPC fanout from
compute nodes to the scheduler.
This patch cleans up libvirt and fake drivers only. This is not needed
for other virt backends as this is a cleanup only. Further cleanup of
libvirt and other drivers will happen after get_host_stats isn't used
for the compute fanout to schedulers.
Because this makes a change to the fake driver, several api samples
needed to be changed as well. The fake driver is changed so the
test_virt_driver tests can continue to be used
Part of bp no-compute-fanout-to-scheduler
Change-Id: I1eec5c117a1cb0490e9f9c09e731909bc31698a9
Balazs Gibizer