The repo is Python 3 now, so update hacking to version 3.0 which
supports Python 3.
Fix problems found.
Update local hacking checks for new flake8.
Remove hacking and friends from lower-constraints, they are not needed
to be installed at run-time.
Change-Id: Ia6af344ec8441dc98a0820176373dcff3a8c80d5
We have provided functions to get used and free resources in
class ModelRoot. So strategies can invoke the functions to
get used and free resources.
Change-Id: I3c74d56539ac6c6eb16b0d254a76260bc791567c
The new bp need to get audit type from audit,
so we need to add an audit parameter to do_execute
Partially Implements: blueprint node-resource-consolidation
Change-Id: Ia979781b32202c1821aa1cb91d24253fe6d7bd2d
For Compute node, when calculating resource(VCPU, memory and disk)
capacity, we need to consider reserved resource and allocation ratio.
Partially Implements: blueprint improve-compute-data-model
Depends-on: I3f9a3279a26f3df444117d9265e74cca57b38d6e
Change-Id: I70257dd5fb342a67a3ffda1055eddc54b8360ca3
Changes to the baseclass for datasources so strategies can be made
compatible with every datasource. Baseclass methods clearly describe
expected values and types for both parameters and for method returns.
query_retry has been added as base method since every current
datasource implements it.
Ceilometer is updated to work with the new baseclass. Several methods
which are not part of the baseclass and are not used by any strategies
are removed. The signature of these methods would have to be changed
to fit with the new base class while it would limit strategies to
only work with Ceilometer.
Gnocchi is updated to work with the new baseclass.
Gnocchi and Ceilometer will perform a transformation for the
host_airflow metric as it retrieves 1/10 th of the actual CFM
Monasca is updated to work with the new baseclass.
FakeMetrics for Gnocchi, Monasca and Ceilometer are updated to work
with the new method signatures of the baseclass.
FakeClusterAndMetrics for Ceilometer and Gnocchi are updated to work
with the new method signatures of the baseclass.
The strategies workload_balance, vm_workload_consolidation,
workload_stabilization, basic_consolidation, noisy_neighbour,
outlet_temp_control and uniform_airflow are updated to work with the
new datasource baseclass.
This patch will break compatibility with plugin strategies and
datasources due to the changes in signatures.
Depends-on: I7aa52a9b82f4aa849f2378d4d1c03453e45c0c78
Change-Id: Ie30ca3dbf01062cbb20d3be5d514ec6b5155cd7c
Implements: blueprint formal-datasource-interface
Allows to define a global preference for metric datasources with the
ability for strategy specific overrides. In addition, strategies which
do not require datasources have the config options removed this is
done to prevent confusion.
Some documentation that details the inner workings of selecting
datasources is updated.
Imports for some files in watcher/common have been changed to resolve
circular dependencies and now match the overall method to import
configuration.
Addtional datasources will be retrieved by the manager if the
datasource throws an error.
Implements: blueprint global-datasource-preference
Change-Id: I6fc455b288e338c20d2c4cfec5a0c95350bebc36
We have a bp to add resource_name in action input parameter field.
Before doing this, one of this method's parameter should be node
instead of resource_id.
Change-Id: I4ce5ae97efce98d80a460fd6003df3cc5cacab82
Many strategies execute very similar statements especially in
pre_execute and some might raise errors that others might not. This
same pattern of many similar statements can also be observed in
strategies their tests.
This patch addresses these issues, firstly; the BaseStrategy class gets
1 additional method _pre_execute which allows for general logic that
most strategies perform at that stage. This method can be executed
before the similarly named method of the superclass. A notable change
is that _pre_execute now handles common exception handling for
ClusterStateStale & ClusterStateNotDefined exceptions.
A similar pattern is applied to the test classes of the strategies
each of these classes now inherits from the TestBaseStrategy class.
This class provides the common attributes almost every test class for
the strategies requires such as: The mocked compute_model, mocked
audit_scope and an instance of FakerModelCollector.
Finally, some minor changes were required in test_strategy_context
& test_audit_handlers and exceptions around 0 nodes in cluster or
storage are removed.
Change-Id: Ia7154376b2448aac65cf17999cc8c3e1c8309b5b
Removed duplicates of strategies descriptions, added references to
that descriptions instead of module descriptions.
Change-Id: Ife396ddce5c3cc926cc111f1ff1abd3a42c22561
This patch set adds /state resource to strategy API
which allows to retrieve strategy requirements.
Partially-Implements: blueprint check-strategy-requirements
Change-Id: I177b443648301eb50da0da63271ecbfd9408bd4f
This patch set follows deprecation steps
and changes default datasource to gnocchi
Closes-Bug: #1735180
Change-Id: I3b7ec47af7707359373a424a6469d28cbf9ce177
At various places in watcher code, we are using property getters
to set property, in this way the property setters defined are
never used, this patch fixes to use property setters to set
property.
Change-Id: Idb274887f383523cea39277b166ec9b46ebcda85
add 'disabled_reason' filed into 'ComputeNode' resource, to distinguish
which nodes are disabled by Watcher and which are not by Watcher.
Implements:blueprint extend-node-status
Change-Id: I7175f14870834a4582e45309529d7e8d9fbb2e6f
Diffrent stratege has diffrent default scope, restrict them to their
default scope will avoid usage problems.
1)workload_balancing/thermal_optimization/airflow_optimization goals
react on enabled nodes, so restrict default scope to compute nodes
with up state and enabled status.
2)server_consolidation goal react on enabled or disabled nodes, So
restrict default scope to compute nodes with up state and
enabled/disabled status.
Change-Id: I7437dee699ee2d3dd227a047196d4d8db811b81e
Closes-Bug: #1714002
This patch adds the functionality to filter out VMs which have
metadata field 'optimize' set to False. This patch implements the
functionality for basic_consolidation strategy.
Change-Id: Iaf7b63e09534e4a67406e7f092242558b78c0bde
Partially-Implements: BP audit-tag-vm-metadata
This patch adds gnocchi support in basic_consolidation strategy
and adds unit tests corresponding to that change.
Change-Id: Ia1ee55fca8eadffbd244c0247577805b6856369d
Partiallly-Implements: bp gnocchi-watcher
This patch set adds new period strategy input parameter
which allows allows to specify the time length of
statistic aggregation.
Partial-Bug: #1614021
Change-Id: I1a276206e5b2c05d8f94acdeb866c8822fa84f35
In this changeset, I use https://review.openstack.org/#/c/362730/
as an example to make the existing ModelRoot fully graph-based.
Change-Id: I3a1ec8674b885d75221035459233722c18972f67
Implements: blueprint graph-based-cluster-model
As instance migration cost is petty compared to the cost of
compute node release, I update the way to compute the global
efficacy for a server consolidation goal. The new formula is simplest
and it's only based on compute node.
Change-Id: Ibcce31a85af70429f412c96c584a761d681366a2
In this changeset, I added the support for both Monasca and
Ceilometer for the basic_consolidation strategy.
Partially Implements: blueprint monasca-support
Change-Id: Ide98550fbf4a29954e46650190a05be1b8800317
In this changeset, I implemented a Helper class to deal with
Monasca requests.
Change-Id: I14cfab2c45451b8bb2ea5f1f48254b41fa5abae8
Partially-Implements: blueprint monasca-support
Specific exception should be thrown when cluster state
is stale. Current usage is to raise this exception if
compute_model.state is True.
Bug was describeid by Jean-Emile DARTOIS.
Change-Id: Iaddb4cc8007c51bb14759c9da829751e834499d0
Closes-Bug: #1621855
Developer should provide a detailled documentation about his strategy
algorithm to make it even easier to use by a Watcher end user.
I propose in this changeset a template for strategy documentation.
you will find also a example with basic consolidation strategy.
Change-Id: I66da1a33b87a94b508dd23ac7dce4cae6f4e068b
In this changeset, I fixed the issue with the basic server
consolidation strategy to now loop over all compute nodes
as expected instead of stopping after the first one.
Change-Id: If594f0df41e39dfb0ef8f0fce41822018490c4ec
Closes-bug: #1548874
In this changeset, I added debug logs to dump the cluster data model
copy structure (as XML) before and after the execution of the strategy.
By doing so, we can see the cluster data model structure that is
expected after the execution of the corresponding action plan.
Change-Id: I81c23e148a78d9b176154f7620087a322a5bce28
