RDP console was only for HyperV driver so removing the
API. As API url stay same (because same used for other
console types API), RDP console API will return 400.
Cleaning up the related config options as well as moving its
API ref to obsolete seciton.
Keeping RPC method to avoid error when old controller is used
with new compute. It can be removed in next RPC version bump.
Change-Id: I8f5755009da4af0d12bda096d7a8e85fd41e1a8c
The RDP console was only available for HyperV driver, therefore its
connection information via API ``os-console-auth-tokens`` will now return
HTTP ``400 (BadRequest)`` error.
Starting from 2.31 microversion, this API return connection info
for all other console type.
Change-Id: I94e590eb4cbe3b2d8eff7fe881f7b98af8979be2
With HyperV driver going away, HyperVLiveMigrateData
is not needed but we will keep it for a couple of cycle.
Adding note about removing it in later releases.
Change-Id: I3574488faffb5a6f5993097496be4de146d61d5c
Awhile back change I52046e6f7acdfb20eeba67dda59cbb5169e5d17e disabled
cinder in the nova-ovs-hybrid-plug job and added checks for cinder
before attempting to run evacuate BFV tests.
Resource setup for BFV was however not bypassed and the attempt to
setup a BFV server resource fails with:
keystoneauth1.exceptions.catalog.EndpointNotFound: publicURL endpoint
for volumev3 service not found
This adds a bypass to avoid attempting to create a BFV server when
cinder is not available.
Change-Id: I52c7e5ce268bb38cee16c18c5523fe0e224970aa
Nova Hyper-V driver is not tested in OpenStack upstream and no maintianers.
This driver has been marked as deprecated in Antelope release. It has dependency
on the OpenStack Winstacker project which has been retired[1].
As discussed in vPTG[2], removing the HyperV driver, tests, and its config.
[1] https://review.opendev.org/c/openstack/governance/+/886880
[2] https://etherpad.opendev.org/p/nova-caracal-ptg#L301
Change-Id: I568c79bae9b9736a20c367096d748c730ed59f0e
Adds the fields to the object so we can use them for passing the values
of the mediated devices betwen the target and the source.
Change-Id: I70d69c2b9c58f34db7e0d57c438db337d52a9a75
Partially-Implements: blueprint libvirt-mdev-live-migrate
Since blockdiag seems a bit unmaintenained, let's just statically
generate the SVGs but let's keep the source files in tree so we can
modify the diagrams whenever we want, provided blockdiag exists in
a foreseenable future :-)
Closes-Bug: #2026345
Change-Id: I1cc078554ab149a9849c895e08c878180b7510b0
This patch introduce the ShareMapping and ShareMappingList top level objects
and associated attach and detach methods that interact with the DB
It also introduce ShareMappingLibvirt and ShareMappingLibvirtNFS
children objects that will be used by the driver part.
Manila is the OpenStack Shared Filesystems service.
These series of patches implement changes required in Nova to allow the shares
provided by Manila to be associated with and attached to instances using
virtiofs.
Implements: blueprint libvirt-virtiofs-attach-manila-shares
Change-Id: Ibacfeaf7daa726bf0ee7802bacd3f70329fec624
we only need to verify if bdm has attachment id and it should be present in both nova and cinde DB.
For tests coverage, added tests for bfv server to test different bdm source type.
Closes-Bug: 2048154
Closes-Bug: 2048184
Change-Id: Icffcbad27d99a800e3f285565c0b823f697e388c
glance client takes image_min_disk as string and return as int,
while glance fixture is returning as it is.
Change-Id: I47df4d147d9dd4c901af2c55c97e111e603cd806
This reverts commit ed2ac71a46b6caf620ddc08219ebae31f5af197d.
This is unnecessary as 2c4421568ea62e66257b55c08092de3e0303fb0a already
fixed this bug on master
Change-Id: Id2c253a6e223bd5ba22512d9e5a40a9d12680da2
The oslo.limit 2.3.0 release contains the validation to ensure
the [oslo_limit] endpoint_id option is set[1]. This fixes the test
failures caused by that validation by overriding the option in all
test cases where unified quota implementation is used.
[1] https://review.opendev.org/c/openstack/oslo.limit/+/897509
Closes-Bug: #2049064
Change-Id: Iffc7a66033a9c09e850890bb380b6667412f0d98
Add mypy so we can run these checks as part of our linting step. Note
that mypy is the last item in the list of checks since flake8 is going
to catch e.g. syntax issues which should be addressed first.
This also allows us to drop use of mypywrap: rather than wrapping mypy,
we can make use of mypy's configuration to specify the files we wish to
run checks on.
Change-Id: Ie51401ead4e31ccd70ab833c59cb397f802ff188
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
list_instances and list_instance_uuids, as written in the Ironic driver,
do not currently respect conductor_group paritioning. Given a nova
compute is intended to limit it's scope of work to the conductor group
it is configured to work with; this is a bug.
Additionally, this should be a significant performance boost for a
couple of reasons; firstly, instead of calling the Ironic API and
getting all nodes, instead of the subset (when using conductor group),
we're now properly getting the subset of nodes -- this is the optimized
path in the Ironic DB and API code. Secondly, we're now using the
driver's node cache to respond to these requests. Since list_instances
and list_instance_uuids is used by periodic tasks, these operating with
data that may be slightly stale should have minimal impact compared to
the performance benefits.
Closes-bug: #2043036
Change-Id: If31158e3269e5e06848c29294fdaa147beedb5a5
pre-commit is an effective linter runner. No need to duplicate things.
We are able to remove our homegrown "is this valid JSON" check by using
the one built into pre-commit. Beautiful!
Change-Id: I89398d4a1c5d4e7205f5f9c499cba0b4ff90fa46
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
This is causing lint failures in the requirements job.
Change-Id: I7afc10130b77bdfb419bb68b670ec7ebecb4026b
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
