Luka Peschke f3ec0d24dd Add an API migration status table to the roadmap

This adds an API migration table to the roadmap. Patches porting v1 API
endpoints to v2 must update this table.

Change-Id: Ia3e74b4377639ce505b1dfc06aff7faab40c86c8
Story: 2006578
Task: 36690

2019-09-19 10:19:42 +02:00

13 KiB

Raw Permalink Blame History

Roadmap

This is the roadmap for planned changes in CloudKitty. Changes are split into:

Continuous
Short-term (planned for the next release)
Mid-term (ideally for the next release, else for release R+2)
Long-term (for changes that will definitely not happen during the next release).

Note

This document must be kept up-to-date. Any newly planned feature should be added. The statuses of the existing features should be updated regularly. At each release, it is the CloudKitty's PTL's responsibility to remove the changes that have been merged during the previous release.

How to edit this document

The first two columns should not need to be modified. If there are several assignees to a change, you can either specify each person individually or write the word multiple in the Assignees column.

Status columns can be in four states:

Not started
Started
Review
Done

See the source file of this document for highlighting syntax (doc/source/developer/roadmap.rst).

Continuous effort

Some points deserve continuous effort. These are not tied to a specific release, but are some of the most important aspects of the project. Some of these can be good first contributions.

Welcoming and mentoring new contributors. Reviewers should be especially kind when reviewing a person's first contribution. Don't assume that they know the "developer workflow" document and OpenStack guidelines by heart, and point them to the right resources if needed.
Improving the documentation. This includes migrating documentation to the new format (adopt a user-profile and component-based layout), but also adding information you figured out by yourself and couldn't find in the existing documentation (for example: notes for specific configuration options, some examples, additional explanations on some notions that may be difficult to grasp for newcomers...).
Improving the troubleshooting documentation. The creation of this documentation is part of the mid-term effort (see below). It will be especially useful for new users. Good first contribution
Adding tests. There are never enough tests, so don't be shy and feel free to improve the current unit tests or add some scenarios to the tempest plugin. Good first contribution

Short-term effort

Planned Change	Assignees	Spec status	Implementation status	Short summary
Adding the v2 API	peschk_l	Done	Done	The new API of CloudKitty, to which all new endpoints will be added.
Support local timezones	peschk_l	Done	Done	Currently, CloudKitty converts all dates to UTC and is not timezone-aware. This must be changed in order to get a better user experience.
Add a Prometheus scope fetcher	jferrieu	Done	Done	A scope fetcher that will work in a similar way to the Gnocchi fetcher (retrieving all values for a given metadata field on a set of metrics).
Add support for the v2 API to the client	peschk_l	Done	Done	Add the necessary base to the client to start supporting v2 API endpoints.
Add a v2 API endpoint allowing to reset the state of a scope	jferrieu	Done	Done	This will allow to delete all the data for a specific scope after a given date, and reset the state of this scope to that date.
Add a V2 API endpoint allowing to retrieve rating information	Multiple	Done	Done	This will be an improved version of the `/summary` endpoint available in the v1 API. It will allow grouping of data on any groupby attribute.
Add a v2 API endpoint allowing to generate reports	jferrieu	Started	Not started	This will be a replacement for `cloudkitty-writer`.

Mid-term effort

Planned Change	Assignees	Spec status	Implementation status	Short summary
Creating a new rating module	Multiple	Started	Not started	This module will add support for validity periods on rating rules, rulesets and will allow rule creation in a declarative way.
Add a second v2 storage backend	peschk_l	Review: https://review.opendev.org/#/c/673461/	Not started	An alternative to InfluxDB, with support for clustering. For now, Elasticsearch has been retained.
Add a troubleshooting documentation	Multiple	Not started	Not started	A documentation providing responses, checklists and tutorials to the most frequently asked questions on the `#cloudkitty` IRC channel.

Long-term effort

Planned Change	Assignees	Spec status	Implementation status	Short summary
Complete migration of the v1 API into v2	Multiple	Started	Not started	Making every (if not deprecated) endpoint of the v1 API available in the v2 API.
Adding authentication middlewares to the API in case it is used without keystone.	Undefined	Not started	Not started	This would allow support for RBAC outside of an openstack context.

API Migration status