This patch enables a white list for query parameter.
The parameters, which are out of white list, will be ignored.
The sort_key still accepts open value. The later patch will add white
list to it.
Co-Authored-By: Zhenyu Zheng <zhengzhenyu@huawei.com>
partial implement of bp add-whitelist-for-server-list-filter-sort-parameters
Change-Id: I7141eef6a1c85ec6d6e8ee170911572535652978
The expected VIFHostUser object in this test is setting the
vif_name field on the object which didn't actually exist until
version 1.1 of that object which is being released in os-vif 1.4.0.
The test passes against os-vif 1.3.0 and VIFHostUser 1.0 today
because the obj_to_primitive() routine will not include anything
that's not a field on the object, which is vif_name in this case.
But when moving to os-vif 1.4.0, we're hitting a failure because
the expected object has vif_name set but the actual object doesn't
because though the vif_name field is defined, it's not actually
used yet in the code, so it's not set in the primitive and our
object comparison fails.
Change-Id: I1c27726d583a41ab69d9eab23e8484e7e047942d
Closes-Bug: #1656025
Live migration for Libvirt vz driver works the same
way as for qemu. However some migration capabilities
are supported differently:
* Virtuozzo containers doesn't use backing file, so
we shouldn't ensure it exists in pre live migration.
* migrate_disks parameter doesn't make sense for
Virtuozzo, as backend determines whether specific
block device should be migrated or not.
* Passing new domain xml to migration method is not
supported.
Implements: blueprint live-migration-vz-driver
Change-Id: I89ecdef13ad47800abc8a5158f8834e46750b9ea
Signed-off-by: Pavel Gluschak <pglushchak@virtuozzo.com>
Minor edits to clarify some points and grammar in the placement
doc. Since I was giving it a close read anyway, I thought it
best to commit changes for the small number of issues that were
apparent.
Change-Id: I8eb94a94a72e3a30fd17479d5168cd6a10f62ed3
gate-nova-tox-db-functional-py35-ubuntu-xenial job has a 100%
failure rate, let's fixup the blacklist and then start working
on the failing tests.
Change-Id: I2b34abc432010b52784f03eff8af669e9c629506
vNUMA can improve the performance of workloads running on virtual machines
that are configured with large amounts of memory. This feature is useful
for high-performance NUMA-aware applications, such as database or web
servers.
Returns Hyper-V host NUMA node information during get_available_resource
Adds validation for instances requiring NUMA topology (no asymmetric
topology and no CPU pinning supported).
Creates NUMA aware instances, if necessary.
The compute-cpu-topologies page in the admin-guide will have to be
updated to include Hyper-V NUMA topologies usage and configuration.
DocImpact
Change-Id: Iba2110e95e80b9511698cb7df2963fd218264c8e
Implements: blueprint hyper-v-vnuma-enable
When we start creating the cellsv2 mappings in the cellsv1 devstack
job, we need to avoid that causing us to circumvent the top-level
merged replica of the cell databases. Doing this causes us to return
stale information that has not yet been updated by cells replication.
Change-Id: Icfbb17cce8ce8b03dc8b7b4ffb202db01e5218a6
This branch adds a new ViF driver to the Nova Hyper-V compute driver.
It's responsible for creating the OVS port and the related association
with the Hyper-V virtual switch port. Aside from the openvswitch binaries
it has no other dependencies. Flow rules are added by an external agent
such as OpenDaylight or by a Neutron OVS agent.
The new VIF driver is using os-vif in order to create the OVS ports.
Co-Authored-By: Adelina Tuvenie <atuvenie@cloudbasesolutions.com>
Co-Authored-By: Claudiu Belu <cbelu@cloudbasesolutions.com>
Depends-On: Ia07ade945ca7469b9c23c801a6bfe08135d39466
Depends-On: Ia33e76a28f96fb94e6fdeedc4947b8a353e3d5bc
Implements: blueprint hyper-v-ovs-vif
Change-Id: I4cee52f65968fa503f2c8a2c9a206489719fb15a
Update the placement docs to reflect the fact that the compute service
will keep attempting to connect to the placement service rather than
just aborting like it used to.
Depends-On: Ie6387afeb239a20d39c00f519e8288f3b3a5e9cb
Change-Id: I6246f6c7318f53c2a917dc320b328fec536506b5
Nova already has support for hotplugging neutorn ports in the libvirt
driver. This extends the support to the XenAPI driver, implement
attaching/detaching VIFs
I have made a patch to run releated testcase in xenserver CI
https://review.openstack.org/#/c/416797/
Change-Id: I22f3fe52d07100592015007653c7f8c47c25d22c
Implements: blueprint xenapi-vif-hotplug
Service token will be passed along with user token to communicate with
services when dealing with long running tasks like live migration.
This change addresses adding service_token to the request when nova
requests cinder session to interact with cinder.
Change-Id: I51eb0a8937fa39a2e5dafb1ad915e7113ea61f72
Implements: blueprint use-service-tokens
Currently the neutron api has different arguments to all the other
network apis. Before we consider adding another argument to this method,
we should first make them all consistent.
blueprint prep-for-network-aware-scheduling-ocata
Change-Id: I42d26f4199d44c7f2e68021014e321a2610a7ee6
XenServer has released os-xenapi lib on pypi, I have made a patch
https://review.openstack.org/#/c/406059/ use os-xenapi in nova
project. In this patch, we just delete those useless files and
folders.
Change-Id: I7cd83a492bae38fd9249f25b95663fa4385b262c
Partially-Implements: blueprint add-os-xenapi-library
XenServer has released os-xenapi lib on pypi, this patch is to
use os-xenapi in nova project. In this patch, we only change
the usage of XenAPI to os-xenapi and fix unit tests. In the next
patch, we will delete those unused files.
Partially-Implements: blueprint add-os-xenapi-library
Depends-On: Ic327135b893a77672fd42af919f47f181e932773
Change-Id: I424dfcd507c4b7fbeea5245cc1c234dec01d5781
In Newton, placement is optional and computes will stop even trying
to connect to placement when they encounter an error or missing
configuration. We really want them to keep trying so that enabling
the service pre-upgrade does not require restarting all computes
to start filling data.
This patch removes the auto-disable logic and replaces it with a
limited, but persistent warning to the logs about the required
nature of placement for the upgrade. If we had messaged the upcoming
requirement better, I think we could have been less chatty here.
However, we know that it's not been received, so this patch periodically
re-emits the warning and mentions the upgrade specifically.
Closes-Bug: #1655494
Change-Id: Ie6387afeb239a20d39c00f519e8288f3b3a5e9cb
He Jie Xu