Currently, when there is a volume_migrate action and migration_type is
`retype`, watcher assumes that the retype always triggers a migration
and checks the result of the retype based on the fields related to
the migration action (actually, it uses the same function to check the
result when `migration_type` is `retype` or `migrate`. This creates
problem in different scenarios:
- Actions keep in ONGOING status forever for volumes which have never
being migrated as the migration fields of the volume are empty.
- Actions which were migrated anytime before, still have the old values
so it may report the status of te retype actions wrongly.
This patch is implementing an entirely new function to check the result
of a retype action based on the final type and the status field of the
volume. This should be valid for any kind of retype action, with or
without migration. The criteria for successfull retype is that the type
for the volume is the destination one in the action and the status is
available or in-use.
Closes-Bug: #2112100
Change-Id: I76e91ed99e7a814a43a6dd906b6bcc150d471624
Signed-off-by: jgilaber <jgilaber@redhat.com>
Monasca is deprecated for removal. This change makes the Monasca client
an optional dependency and ensures it is only imported and instantiated
when the Monasca datasource is explicitly selected. This reduces the
default footprint while preserving functionality for deployments that
still rely on Monasca.
What changed
============
- requirements.txt: remove python-monascaclient from hard deps
- setup.cfg: add [options.extras_require] monasca extra
- watcher/common/clients.py: lazy import with clear UnsupportedError
- watcher/decision_engine/datasources/monasca.py: lazy client property
and deferred import of monascaclient.exc; reset on Unauthorized
- watcher/decision_engine/datasources/manager.py: unconditionally
import Monasca helper and include in metric_map; helper is lazy
- tests: conditionally include Monasca based on availability; adjust
expectations instead of skipping by default; avoid over-mocking
- tox.ini: enable optional extras via WATCHER_EXTRAS env var
- docs: datasources index notes Monasca is deprecated and optional
- releasenotes: upgrade note with install example and behavior
Why
===
- Allow deployments not using Monasca to run without the client
- Keep Monasca functional when explicitly installed via extras
- Provide clear operator guidance and smooth upgrades
Compatibility
=============
- No change for deployments that do not use Monasca
- Deployments using Monasca must install the optional extra:
pip install watcher[monasca]
Testing
=======
- Default: tox -e py3
- With Monasca: WATCHER_EXTRAS=monasca tox -e py3
Assisted-By: GPT-5 (Cursor)
Closes-Bug: #2120192
Change-Id: I7c02b74e83d656083ce612727e6da58761200ae4
Signed-off-by: Sean Mooney <work@seanmooney.info>
Fixed action status_message update restrictions to allow updates when
action is already in SKIPPED state. Previously, users could only update
the status_message when initially transitioning to SKIPPED state.
Changes include:
- Modified validation logic to allow status_message updates for SKIPPED actions
- Changed exception type from PatchError to Conflict for better semantics
- Added comprehensive test coverage for the new behavior
- Updated API documentation and samples
- Added release note documenting the fix
This enables administrators to fix typos, provide more detailed
explanations, or expand on reasons in action status messages after
the action has been skipped.
Generated-By: claude-code
Closes-Bug: #2121601
Change-Id: I64def708389a8ecd32080fba1638a4499ead349d
Signed-off-by: Sean Mooney <work@seanmooney.info>
Job watcher-aetos-integration is failing because of
having real metrics enabled coming from ceilometer.
We need to disable ceilometer-acompute and node_exporter so only
injected data will be considered when asking prometheus to take
decisions
Change-Id: If4f2c3f6f89527d768c48f1ca4967339837bb994
Signed-off-by: morenod <dsanzmor@redhat.com>
These do not actually define timeout but interval. Rename the options
to reflect what they actually define. The existing deprecated options
in the [gnocchi_client] are also removed, because these have been kept
for 6 years.
In addition, fix inconsistent name (query vs call).
Change-Id: Ib29115746a25b45bdff1c3da8df9d7167c2db662
Signed-off-by: Takashi Kajinami <kajinamit@oss.nttdata.com>
This patch extends compute model attributes by
adding new fields to Instance element. Values are
populated by nova the collector, using the same
nova list call, but requires a more recent compute
API microversion.
A new config option was added to allow users to
enable or disable the extended attributes and it is
disable by default.
Configure prometheus-based jobs to run on newer version
of nova api (2.96) and enables the extended attributes
collection.
Implements: bp/extend-compute-model-attributes
Assisted-By: Cursor (claude-4-sonnet)
Change-Id: Ibf31105d780dce510a59fc74241fa04e28529ade
Signed-off-by: Douglas Viroel <viroel@gmail.com>
These are some of the requested changes from reviews
in the series of patches for add-skip-action blueprint.
Some of them may required another specific patch since
would touch in more files that are not related to
this feature.
Change-Id: I9e30ca385e7b184ab19449a60db6f6d0f3c0e1b9
Signed-off-by: Douglas Viroel <viroel@gmail.com>
... to avoid the following warning.
```
UserWarning: converting '1' to a string
warnings.warn('converting \'%s\' to a string' % str_val)
```
Change-Id: I852d63523d3582f00d4d7953199181e3d2b6a885
Signed-off-by: Takashi Kajinami <kajinamit@oss.nttdata.com>
Bug #2113862 details a number of suggested
corrections and additions to the Workload
Stabilization doc. This patch adds those
suggested changes.
Closes-Bug: #2113862
Assisted-By: Cursor (claude-3.5-sonnet)
Change-Id: I4131a304c064d2ea397b2447025c7edf69a56e2a
Signed-off-by: Ronelle Landy <rlandy@redhat.com>
This change enhances the Host Maintenance strategy by introducing
two new input parameters: `disable_live_migration` and
`disable_cold_migration`. These parameters allow cloud
administrators to control whether live or cold migration should be
considered during host maintenance operations.
If `disable_live_migration` is set, active instances will be cold
migrated if `disable_cold_migration` is not set, otherwise
active instances will be stopped. If `disable_cold_migration` is set,
inactive instances will not be cold migrated.
If both are set, only stop actions will be performed on instances.
The strategy logic and action plan generation have been updated to
reflect these behaviors. A new "stop" action is introduced and
registered, and the weight planner is updated to handle new action.
Documentation for the Host Maintenance strategy is updated to
describe the new parameters and their effects.
Test Plan:
- Unit tests for HostMaintenance strategy with new parameters
- Integration tests for action plan generation with stop action
This implements the specification:
Spec: https://review.opendev.org/c/openstack/watcher-specs/+/943873
Change-Id: I201b8e5c52e1bc1a74f3886a0e301e3c0fa5d351
Signed-off-by: Quang Ngo <quang.ngo@canonical.com>
Fixes the microversion comparison in both enable and
disable nova-compute service methods in NovaHelper.
The previous implementation was incorrect and started to
fail for microversion greather than 2.99.
Closes-Bug: #2120586
Assisted-By: Cursor (claude-4-sonnet)
Change-Id: I69da7f10cd5b42f7d4613d8947bca3e382815c3f
Signed-off-by: Douglas Viroel <viroel@gmail.com>
This patch implements the changes in the API required for the
skipped action blueprint. It includes:
- New field `status_message` is visible in API get calls for Audits,
ActionPlans and Audits.
- New Patch call is added to `/actions/{action_id}` which allows to
manually move actions in PENDING state to SKIPPED for ActionPlans
which have not been started.
- A new API microversion 1.5 is added for these changes.
It also adds requried tests and documentation.
Implements: blueprint add-skip-actions
Assisted-By: Cursor (claude-4-sonnet)
Change-Id: I71fb9af76085e5941a7fd3e9e4c89d6f3a3ada47
Signed-off-by: Alfredo Moralejo <amoralej@redhat.com>
This patch is implementing skipping automatically actions based on the
result of action pre_condition method. This will allow to manage
properly situations as migration actions for vms which does not longer
exist. This patch includes:
- Adding a new state SKIPPED to the Action objects.
- Add a new Exception ActionSkipped. An action which raises it from the
pre_condition execution is moved to SKIPPED state.
- pre_condition will not be executed for any action in SKIPPED state.
- execute will not be executed for any action in SKIPPED or FAILED state.
- post_condition will not be executed for any action in SKIPPED state.
- moving transition to ONGOING from pre_condition to execute. That means
that actions raising ActionSkipped will move from PENDING to SKIPPED
while actions raising any other Exception will move from PENDING to
FAILED.
- Adding information on action failed or skipped state to the
`status_message` field.
- Adding a new option to the testing action nop to simulate skipping on
pre_condition, so that we can easily test it.
Implements: blueprint add-skip-actions
Assisted-By: Cursor (claude-4-sonnet)
Change-Id: I59cb4c7006c7c3bcc5ff2071886d3e2929800f9e
Signed-off-by: Alfredo Moralejo <amoralej@redhat.com>
... instead of documenting the supported values, so that more explicit
error is presented to users.
Also drop redundant description about the default values. The default
values are added to sample config files generated, so don't have to
be explained in help texts.
Change-Id: I12b201da3e742b55f6cfcf71bdd4413cbf3ee4e5
Signed-off-by: Takashi Kajinami <kajinamit@oss.nttdata.com>
This patch is part of the skipped action blueprint. It adds the
`status_message` field to the Audit, ActionPlan and Action objects and
all related notifications.
It bumps the versions of all the affected objects and notifications and
update the tests to include the new fields.
Change-Id: I3b9467e7e37188e647379cd9c4cbbda8ed75383f
Signed-off-by: Alfredo Moralejo <amoralej@redhat.com>
This patch implements the changes in the database required for the
skipped action blueprint.
It just adds a new nullable column to the required tables and add tests
for it.
Note that I am also introducing a fix in a previous tables tests which
will be affected by the changes in the objects.
Implements: blueprint add-skip-actions
Change-Id: I027bc3861b589bd281a7216583a8c5c351a53c57
Signed-off-by: Alfredo Moralejo <amoralej@redhat.com>
In order to test the different code paths for action execution
it is very useful to be able to make the actions fail in the different
execution stages.
This patch adds three new options `fail_pre_condition`, `fail_execute`
and `fail_post_condition`. Setting any of them to True makes the action
to fail in the specified step.
Change-Id: Ied8c0bb767d9bb6bdfb9209365857a3b4d606b40
Signed-off-by: Alfredo Moralejo <amoralej@redhat.com>
Currently, patch call field validations are done based on exclussion,
all the fields can be patched unless included in a list
`internal_attrs`.
This patch is adding a new validation rule based on fields inclussion
in a list `allowed_attrs`. When that list is non-empty, only the fields
included on it can be patched. in order to keep the existing behavior
for the existing patch calls, I am defining the list as empty, so that
the rest of validation rules are applied and it is not affecting the
current behavior.
Change-Id: I22010649332c8fb872446a9d0483a0303a4eba3b
Signed-off-by: Alfredo Moralejo <amoralej@redhat.com>
This change removes watchers in tree functionality
for swapping instance volumes and defines swap as an alias
of cinder volume migrate.
The watcher native implementation was missing error handling
which could lead to irretrievable data loss.
The removed code also forged project user credentials to
perform admin request as if it was done by a member of a project.
this was unsafe an posses a security risk due to how it was
implemented. This code has been removed without replacement.
While some effort has been made to allow existing
audits that were defined to work, any reduction of functionality
as a result of this security hardening is intentional.
Closes-Bug: #2112187
Change-Id: Ic3b6bfd164e272d70fe86d7b182478dd962f8ac0
Signed-off-by: Sean Mooney <work@seanmooney.info>
Implement the spec for multi-tenancy support for metrics. This adds
a new 'Aetos' datasource very similar to the current Prometheus
datasource. Because of that, the original PrometheusHelper class
was split into two classes and the base class is used for
PrometheusHelper and for AetosHelper. Except for the split, there
is one more change to the original PrometheusHelper class code, which
is the addition and use of the _get_fqdn_label() and
_get_instance_uuid_label() methods.
As part of the change, I refactored the current prometheus datasource
unit tests. Most of them are now used to test the PrometheusBase class
with minimal changes. Changes I've made to the original tests:
- the ones that can be be used to test the base class are moved into the
TestPrometheusBase class
- the _setup_prometheus_client, _get_instance_uuid_label and
_get_fqdn_label functions are mocked in the base class tests.
Their concrete implementations are tested in each datasource tests
separately.
- a self._create_helper() is used to instantiate the helper class with
correct mocking.
- all config value modification is the original tests got moved out and
instead of modifying the config values, the _get_* methods are mocked
to return the wanted values
- to keep similar test coverage, config retrieval is tested for each
concrete class by testing the _get_* methods.
New watcher-aetos-integration and watcher-aetos-integration-realdata
zuul jobs are added to test the new datasource. These use the same set
of tempest tests as the current watcher-prometheus-integration jobs.
The only difference is the environment setup and the Watcher config,
so that the job deploys Aetos and Watcher uses it instead of accessing
Prometheus directly.
At first this was generated by asking cursor to implement the linked spec
with some additional prompts for some smaller changes. Afterwards I manually
went through the code doing some cleanups, ensuring it complies with
PEP8 and hacking and so on. Later on I manually adjusted the code to use
the latest observabilityclient changes.
The zuul job was also mostly generated by cursor.
Implements: https://blueprints.launchpad.net/watcher/+spec/prometheus-multitenancy-support
Generated-By: Cursor with claude-4-sonnet model
Change-Id: I72c2171f72819bbde6c9cbbf565ee895e5d2bd53
Signed-off-by: Jaromir Wysoglad <jwysogla@redhat.com>
The data_model list API response comes from the model to_list()
method, which generates both server_* and node_* attributes from
Instance and Node classes fields[1]. Any change on these classes
can break the data_model list API and require a new microversion.
These tests validate the current expected fields.
[1] 5ba086095c/watcher/decision_engine/model/model_root.py (L250-L270)
Change-Id: I77fac162101013aa923272aa99c7c6695cc5fdca
Signed-off-by: Douglas Viroel <viroel@gmail.com>
Some response parameters from GET /infra-optim/v1/data_model
endpoint are missing from api-ref documentation. This patch
updates the doc to include them.
For more details see, LP #2117726
Closes-Bug: #2117726
Change-Id: Iaa775f56bb8167d9c6b458cd07f1ec3cefaf70fe
Signed-off-by: Douglas Viroel <viroel@gmail.com>
It is done by disabling the eventlet patching and configuring
oslo.service backend to threading. Once oslo.service backend is
configured, it can't be reverted to eventlet. This needs to be
done before including other modules, which may include oslo.service
library.
Adds a job that run a subset of tests with eventlet patching disabled.
Change-Id: I9f8c2c5bbcf3192313cc3b309e8f2719a3bea18f
Signed-off-by: Douglas Viroel <viroel@gmail.com>
