In Pike OpenStack documentation is changed in a way that the project repos contain most of the documentation material [1]. This affcects how the doc folder shall be strucutred inside the project repo. With this change I align the install guide to the new structure and correct the warnings during the document generation. The overall work is tracked in [2]. [1]: http://specs.openstack.org/openstack/docs-specs/specs/pike/os-manuals-migration.html [2]: https://etherpad.openstack.org/p/doc-migration-tracking Change-Id: Ib5400247d2ba5f8dac80ef68a43d06868bd42f1c Depends-On: Ia750cb049c0f53a234ea70ce1f2bbbb7a2aa9454 Signed-off-by: csatari <gergely.csatari@nokia.com>
4.3 KiB
Static Datasource Configuration
Overview
The static datasource allows users to integrate the unmanaged resources and topology into Vitrage. Unmanaged means the resource, relationship or property can not be retrieved from any API or database, except static configuration file.
This datasource is static - pre-configured in a file. This is sufficient in many cases where the resources and relationship is relatively unchanging.
Static datasource suppresses the legacy static physical datasource. Theoretically both physical and virtual resources and relationship between them can be configured in it.
Configure Access to Static
The following should be set in
/etc/vitrage/vitrage.conf, under [static]
section:
Name | Description | Default Value |
---|---|---|
directory | Directory path from where to load the configurations | /etc/vitrage/static_datasources/ |
changes_interval | Interval of checking changes in the configuration files | 30 seconds |
Configure Static Mapping
Static configuration is made for configuring statically managed resources, and their relationships to other resources in the topology. Some physical resources, such as switches, can not be retrieved from OpenStack, and so are defined here.
Static datasource use the same semantics as Vitrage template, except for the following extension
- Static resources are identified in Vitrage by
static_id
instead oftemplate_id
- Entity
id
must be specified to map the actual resource - All entities configured in static datasource are considered
RESOURCE
scenarios
section is not applicable
There may be more than one configuration file. All files will be read
from /etc/vitrage/static_datasources/
. See previous section
on how to configure this location.
- Notes:
-
- Static datasource shares the same configuration folder as legacy static physical datasource.
- Both static configuration and legacy static physical configuration will be loaded in Ocata release.
- The format is distinguished by checking existence of
metadata
key which is only available in static datasource.
Example
metadata:
name: # configuration name
description: # configuration description
definitions:
entities: # list of resources, note that alarms can not be defined for contrast with Vitrage template
- static_id: s1 # unique ID in static datasource, it will be referred in relationship definition
type: switch # resource type, could be any string, not limited to the type from existing datasource.
id: 12345 # resource ID, used together with ``type`` to refer a resource in real world
name: switch-1 # name, state and other properties are considered as metadata of the resource
state: active # the state of the resource
...
- static_id: h1
type: nova.host # resource type could be from existing datasource
id: 1 # resource ID, used together with ``type`` to refer a resource in corresponding datasource
state: active # the state of the resource
purpose: CI # additional properties could be defined - if from existing datasource, it could be updated
...
relationships:
- source: s1 # static ID of source entity
target: h1 # static ID of target entity
relationship_type: attached # relationship type, it will be used in scenario condition check
...
The example above defines a switch with ID 12345
attached to nova host with ID 1
. The user also noted that
this host will be used for CI.