When the nova-api services starts from its own standalone binary it
registers itself in the services table. The original wsgi script in
nova/wsgi/nova-api.py did not, leading to the bug referenced below.
The new wsgi script at nova.api.openstack.compute.wsgi, modelled on
a similar thing used for the placement API, provides the necessary
service registration.
If a ServiceTooOld exception happens while trying to register the
service then a very simple (currently very stubby) application is
loaded instead of the compute api. This application returns a 500
and a message.
Some caveats/todos:
* wsgi apps managed under mod-wsgi (and presumably other containers)
are not imported/compiled/run until the first request is made. In
this case that means the service handling does not happen until
that first request, somewhat defeating the purpose if the api is a
bit idle.
Change-Id: I7c4acfaa6c50ac0e4d6de69eb62ec5bbad72ff85
Closes-Bug: #1661360
This patch adds '/os-aggregates' related routes by a plain list, instead
of using stevedore. After all the Nova API endpoints moves to
the plain routes list, the usage of stevedore for API loading will be
removed from Nova.
Also note that the original 'ProjectMapper' use the 'routes.Mapper.resource'
to create a set of routes for a resource which comform to the Atom
publishing protocol. It includes some of URL mappings we didn't document
before. This patch will remove those URL mappings. For the detail,
please reference:
http://lists.openstack.org/pipermail/openstack-dev/2017-March/114736.html
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: Ia41c096a579eb1a7c5b99bd6e9267ab4b7f81776
This patch adds '/os-keypairs' related routes by a plain list, instead
of using stevedore. After all the Nova API endpoints moves to
the plain routes list, the usage of stevedore for API loading will be
removed from Nova.
Also note that the original 'ProjectMapper' use the 'routes.Mapper.resource'
to create a set of routes for a resource which comform to the Atom
publishing protocol. It includes some of URL mappings we didn't document
before. This patch will remove those URL mappings. For the detail,
please reference:
http://lists.openstack.org/pipermail/openstack-dev/2017-March/114736.html
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: I9b18e7acce37ba935f4302cc89c3aeccc4f50915
This patch adds flavors-access related routes by a plain list, instead
of using stevedore. After all the Nova API endpoints moves to
the plain routes list, the usage of stevedore for API loading will be
removed from Nova.
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: I284516a396e99a5a1c2113ad4ef3e4618628caa9
This patch adds flavors-extraspecs related routes by a plain list, instead
of using stevedore. After all the Nova API endpoints moves to
the plain routes list, the usage of stevedore for API loading will be
removed from Nova.
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: I7578ef265458365b44b366c4021d0566c39ea936
This patch adds flavor related routes by a plain list, instead
of using stevedore. After all the Nova API endpoints moves to
the plain routes list, the usage of stevedore for API loading will be
removed from Nova.
flavor access and extraspecs router will be added in part2.
Also note that the original 'ProjectMapper' use the 'routes.Mapper.resource'
to create a set of routes for a resource which comform to the Atom
publishing protocol. It includes some of URL mappings we didn't document
before. This patch will remove those URL mappings. For the detail,
please reference:
http://lists.openstack.org/pipermail/openstack-dev/2017-March/114736.html
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: I1c3db4e94ccc27925d800876dcf470614f563cd1
This patch add '/servers' related routes by a plain list, instead
of using stevedore. After all the Nova API endpoints moves to
the plain routes list, the usage of stevedore for loading the API
will be removed from Nova.
To remove the servers extension from stevedore, all the extensions
which depend on servers needs to be removed together. Those
extensions are about the servers API response extension and the action
extension.
Also note that the original 'ProjectMapper' use the 'routes.Mapper.resource'
to create a set of routes for a resource which comform to the Atom
publishing protocol. It includes some of URL mappings we didn't document
before. This patch will remove those URL mappings, also remove the
corresponding URL mappings for 'os-volumes_boot' endpoint. For the detail,
please reference:
http://lists.openstack.org/pipermail/openstack-dev/2017-March/114736.html
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: I76c384c10bd804fc2049aef305044149bb55d0dc
The os-pci API was never part of the v2.0 API and was added
to the v3 API, but when the v3 API turned into the v2.1 API
which is backward compatible with the v2.0 API, the os-pci
API was removed from v2.1. The original intent was to enable
it in a microversion but that never happened.
We should just delete this API since it has a number of issues
anyway:
1. It's not documented (which makes sense since it's not enabled).
2. The PciHypervisorController just takes the compute_nodes.pci_stats
dict and dumps it to json out of the REST API with no control over
the keys in the response. That means if we ever change the fields
in the PciDevicePool object, we implicitly introduce a backward
incompatible change in the REST API.
3. We don't want to be reporting host stats out of the API [1].
4. To make the os-hypervisors extension work in a multi-cell environment
we'd have to add uuids to the PciDevices model and change the API to
return and take in uuids to identify the devices for GET requests.
5. And last but not least, no one has asked for this in over two years.
As a result of removing this API we can also remove the join on the
pci_devices table when showing details about an instance or listing
instances, which were added years ago because of the PciServerController:
Id3c8a0b187e399ce2acecd4aaa37ac95e731d46c
Id3e60c3c56c2eb4209e8aca8a2c26881ca86b435
[1] https://docs.openstack.org/developer/nova/policies.html?#metrics-gathering
Closes-Bug: #1426241
Closes-Bug: #1673869
Change-Id: I9099744264eeec175672d10d04da69648dec1a9d
This patch remove the stevedore extension point for server
create.
This patch move all the extension points into ServersController
class attribute 'server_create_func_list'. This is for
backward-compatible with the stevedore extension interface.
The final goal is that merging all of those extended code into
the ServerController. So this is a middle step for refactor.
Partial-implement-blueprint api-no-more-extensions-pike
Change-Id: I9a8f56037b35e89543626922221b910ab8b1917e
With pbr 2.0 and Sphinx 1.5, the setting for treat sphinx warnings as
errors is setting warning-is-error in build_sphinx section. Migrate the
setting from the old warnerror one.
Change-Id: Icd5cf4c279c566e1eddeede8df11a9d1c4d4c83f
Fix the following warnings.
- A warning in config sample generation
- Warnings about nova-status.rst
Change-Id: Ifcc3b4a89eeea9d0dd62e2a8b560c5e6a9ff3d1a
Closes-Bug: #1659485
* Add osprofiler wsgi middleware. This middleware is used for 2 things:
1) It checks that person who want to trace is trusted and knows
secret HMAC key.
2) It starts tracing in case of proper trace headers
and adds the first wsgi trace point with info about the HTTP request
* Add initialization of osprofiler on start of a service
Currently that includes oslo.messaging notifier instance creation
to send Ceilometer backend notifications.
oslo-spec: https://review.openstack.org/#/c/103825/
python-novaclient change: https://review.openstack.org/#/c/254699/
based on: https://review.openstack.org/#/c/105096/
Co-Authored-By: Boris Pavlovic <boris@pavlovic.me>
Co-Authored-By: Munoz, Obed N <obed.n.munoz@intel.com>
Co-Authored-By: Roman Podoliaka <rpodolyaka@mirantis.com>
Co-Authored-By: Tovin Seven <vinhnt@vn.fujitsu.com>
Implements: blueprint osprofiler-support-in-nova
Change-Id: I82d2badc8c1fcec27c3fce7c3c20e0f3b76414f1
This adds the basic framework for the nova-status
upgrade check commands. Follow up patches will flesh
this out to perform cells v2 and placement API related
upgrade status checks.
A man page is added for the new CLI and as part of that
the man page index is sorted.
Part of blueprint cells-scheduling-interaction
Part of blueprint resource-providers-scheduler-db-filters
Change-Id: I687dd7317703a1bb76c197ebba849ca368c0872e
This was deprecated and to set to be removed in Ocata. It was only
intended for testing purposes and not for production use. Individual
binaries will be used instead.
Change-Id: I260962ec7c4167b07e9480a04454a9cb5666f8ed
We want a wsgi_script as the entry point for our placement API
actually getting run, this is wrapping in the smooth and mellow pbr
patterns that make it sensible to also run just by starting it on the
command line.
This also actually initializes the logging subsystem for the placement
API, and does the standard pattern of dumping the configuration if
DEBUG is enabled. Pieces of this were cribbed/inspired by equivalent
keystone code.
The config directory is now setable via environment, which may be be
needed by folks with venvs.
Change-Id: I00d032554de273d7493cfb467f81687c08fd5389
This patch adds new command line 'nova-policy'. This is administrator
command line, it is tool for experience the policy discovery which
is still under-development.
Co-Authered-by: Alex Xu <hejie.xu@intel.com>
Change-Id: If4abf0fca043de2cb5c719a53505526f3c0b871b
Implements: blueprint discoverable-policy-cli
There are two helper scripts in oslo.policy to help deployers understand
their policy configuration better. With the setup.cfg entry these can be
called directly from oslo.policy.
Change-Id: I08dc33367401ec1f98e1795a52d4e981f09a07de
Implements: bp policy-in-code
All remaining list_opts methods come from nova.conf.opts
now so remove the last remaining unused (nonexistent)
list_opts registrations.
Change-Id: Idfc9c464bae828dd1d024369cb972401e52ac101
There was never a list_opts method in nova.cache_utils,
and the options registeration was moved to nova.conf.cache
in 466bce576a so we can
cleanup the entrypoint.
Change-Id: If02620752c1bb018117f2c6755d9b9661185e4d6
list_opts was moved from neutronv2.api to nova.conf.neutron
in 1b8b0834a0 so we can cleanup
this broken entry point.
Change-Id: Ie3e1b25b6a0117aed2f87d1fd6ecaab35201349f
This adds the entry point and config file necessary for using the
oslo.policy sample generation script. It also adds a tox target to
simplify the usage of it.
The command is also run in the docs tox target since the sample file
can be considered documentation.
Change-Id: I18a1fa567fc04b0e0a37c02f20024b2ce483f060
Partially-Implements: bp policy-in-code
This removes personality extension, as well as the rebuild extension
point for servers.
We add an operation flag to translate extension because personality
extension uses different kwarg names depending on the operation. This
is done with a set of constants which should prevent typos silently
causing failures.
A few variables are renamed in servers.py to avoid multiline
statements.
Part of bp:api-no-more-extensions
Change-Id: I47deae0279f85d9a355d1248b6e90af732406514
This folds this back into the main rebuild flow. As there were no unit
tests for this, doing so was pretty simple.
Part of bp:api-no-more-extensions
Change-Id: I19f791ed5af917b5509940765fedc1b944fcf315
There are no test changes as all new code is tested by the
extension_info functional tests, and should we want to eliminate the
extension_info portion of our API later, new tests would just make
this much harder to tear down.
This modifies default policy to drop os-disk-config as well as
modifying extension tests to pass even though we're injecting
additional hardcoded items into the list for compatibility.
Part of bp:api-no-more-extensions
Change-Id: I0192ccef7e858fd0ff15c339f683abaabbdc09b7
This was the entrypoint namespace for
compute resources which was deprecated in mitaka
under 649decbaad
and removed in newton 49d9433c62
so we should also remove the entrypoint.
Change-Id: Idaed7291ab6c3ab122edabfd3d614f42d469a34f
This removes the portions of the os-disk-config extension that handles
additional attributes being passed in during
create/update/resize/rebuild. These are folded directly into
servers.py with a new module wide function for translating REST in =>
attr dict data structures.
Tests which use disk_config as the example for loading and using these
extensions are removed.
As this is the only in tree extension that hooks the resize entry
point, that is fully removed from setup.cfg and well as any extension
manager processing.
Part of bp:api-no-more-extensions
Change-Id: I691a3a917a3dbaa86072d689e9ebf42ba356ca60
Some comments on `Improve the help text for cells options`
reviews are implemented with this patch.
- nova.cells.opts:list_ops function removed.
Implements blueprint centralize-config-options-newton
Change-Id: Idcaf66f0db1695d4fff34a656df2934c1b6d2ede
Two years ago the translation files have been split into several
files, separating the log messages of different log levels from each
other, like X.pot, X-log-warning.pot, X-log-info.pot, and so on.
However, the setup.py command `compile_catalogs`, that comes from the
babel package and compiles the corresponding .po files into .mo
files, only supported one file per python package. This means that
during packaging `compile_catalogs` never compiled the X-log-*.po
files, so the corresponding translations were always missing.
Since babel 2.3 the domain can be set to a space separated list of
domains. This change adds the the additional log level files to the
domain list.
The obsolete check that .po and .pot files are valid is removed from
tox.ini.
Change-Id: I0d9cb16ac50caae570c57a1db6bed656bf13c3dc
Closes-Bug: #1536226
This is the fifth and final patch moving config options from the
nova/api directory. In this patch, the deprecated options from the
legacy_v2 directory have been moved to nova/conf/legacy_v2.py. A
subsequent patch will enhance the help text for these options.
Blueprint centralize-config-options-newton
Change-Id: I3fdff0d3c0c2f36155ad4f7060395e788b78d150
Moving the config options, used in nova/network/rpcapi.py
for nova network, to the new central location
"nova/conf/network.py".
Implements: blueprint centralize-config-options-newton
Change-Id: I62f019632e72359711430ad2f017911337f2e30b
Added new API microversion which allows the following:
- add tag to the server
- replace set of server tags with new set of tags
- get information about server, including list of tags for server
- get just list of tags for server
- check if tag exists on a server
- remove specified tag from server
- remove all tags from server
- search servers by tags
DocImpact
APIImpact
Implements: blueprint tag-instances
Change-Id: I9573aa52aae9f49945d8806ca5e52ada29fb087a
This disables the generation of the raw module api documentation in
our docs target. It is mostly not useful, as it builds a giant tree of
module documentation that are 98% boiler plate lists of methods with
no real content.
We leave the sphinx.ext.autodoc in the conf, which allows you to
specifically pull in modules for documentation when you want
to. doc/source/services.rst is an example of doing this in tree, which
still works after this change.
Change-Id: I4c10a8e45756cdcf612faca574e2fb3b7ffa2bdb
No config generator hooks should ever be registered with a name that
belongs to another project. In this case, using oslo.middleware.cors
means that *every other project* that loads the middleware gets this
application's defaults when the generator is run on a system with
everything installed (such as a dev box with devstack). Use the name
of the app instead, to ensure that the defaults are only set when this
app's sample config and documentation are being generated.
Change-Id: I6a8c7d44b9db9325003ff2fdb667b0ced7739e96
Signed-off-by: Doug Hellmann <doug@doughellmann.com>
This patch makes use of oslo.config's genconfig hooks to add
nova-specific default config values to the generated configuration.
It also adds this same method to nova's configuration parsing logic,
to ensure that these default values are also the ones used at runtime.
Closes-bug: 1551836
Change-Id: Ic799aea89f681ef8626e996f0357ade388f76ca9
This change adds manual knob to force ongoing live migration to
complete. It is implemented as a new server-migrations API.
DocImpact
ApiImpact
Implements: blueprint pause-vm-during-live-migration
Change-Id: I034b4041414a797f65ede52db2963107f2ef7456
Jenkins