Feature Flags
Feature flags are declared in Sentry's codebase (look for SENTRY_FEATURES
in
server.py
).
For self-hosted users, those flags are then configured via sentry.conf.py
.
For Sentry's SaaS deployment, options automator is used to configure flags in production.
You can find a list of features available by looking at two files:
sentry/features/permanent.py
- Features typically managed by subscriptions for sentry.iosentry/features/temporary.py
- Features used during development intended to be removed
They're declared on the FeatureManager
like so:
# pass FeatureHandlerStrategy.FLAGPOLE to use our options-backed feature flagging system:
manager.add("organizations:onboarding", OrganizationFeature, FeatureHandlerStrategy.FLAGPOLE)
# pass FeatureHandlerStrategy.INTERNAL if you don't plan to use options automator:
manager.add("organizations:onboarding", OrganizationFeature, FeatureHandlerStrategy.INTERNAL)
# [DEPRECATED] pass FeatureHandlerStrategy.OPTIONS to use options automator:
manager.add("organizations:onboarding", OrganizationFeature, FeatureHandlerStrategy.OPTIONS)
The feature can be enabled with the following in your sentry.conf.py
, usually located at ~/.sentry/
:
SENTRY_FEATURES["organizations:onboarding"] = True
You can modify the state of feature flags in tests using a context manager.
Generally you want your feature names to be unique to help in their
removal. For example a feature flag like trends
may prove difficult to find because trends
may appear throughout the codebase.
But a name like performance-trends-view
is more likely to be unique and easier
to remove later
Creating a new Feature Flag
Determine what scope the feature should have
Features can be scoped by organization, and projects. If you're not confident
you want a project feature, create an organization level one. In this example
we'll build a feature called test-feature
scoped at the organization level.
Determine the permanency of your feature
Typically we use feature flags for development. They are usually intended to be graduated. The only exceptions to this are permanent flags which control subscription plan-specific features in GetSentry.
Most Sentry feature flags are placed in temporary.py
, while permanent Sentry flags live in
permanent.py
.
GetSentry only flags are typically placed in features.py
.
Determine whether your feature needs to be exposed in the API
If you plan on using your feature in the frontend, you need to set
api_expose=True
when adding your feature. Features that have api_expose
will
be included in the results of the organization details response.
Add your feature to the FeatureManager
If you want to back your feature flag via options, you can do so using the Flagpole library
by adding the feature to the FeatureManager
with the FLAGPOLE
enum set as the feature strategy:
default_manager.add('organizations:test-feature', OrganizationFeature, FeatureHandlerStrategy.FLAGPOLE, api_expose=True)
When defining a feature you can also set the default state of the feature flag.
If no default is provided, False
will be used.
# Example of a feature set with a default value of True
default_manager.add('organizations:test-feature', OrganizationFeature, FeatureHandlerStrategy.FLAGPOLE, default=True, api_expose=True)
If you don't plan to use Flagpole, use FeatureHandlerStrategy.INTERNAL
with a custom feature handler instead, for example:
default_manager.add('organizations:test-feature', OrganizationFeature, FeatureHandlerStrategy.INTERNAL)
Add it to the Organization Model Serializer
The Organization model serializer
(src/sentry/api/serializers/models/organization.py
) builds a list
called feature_list
that is given to the front-end to use. By default, all
organization features are checked and those that are present are added into the list. If
your feature requires additional custom logic, you will have to update the
organization serializer to check and include it manually.
Using Model Flags (Less common)
Sometimes a flag on the model is used to indicate a feature flag as shown
below. This is not recommended unless there is a specific reason to make
changes to the model. For example, the require_2fa
flag affects behavior on
the backend to enforce two-factor authentication.
feature_list = []
if getattr(obj.flags, 'allow_joinleave'):
feature_list.append('open-membership')
if not getattr(obj.flags, 'disable_shared_issues'):
feature_list.append('shared-issues')
if getattr(obj.flags, 'require_2fa'):
feature_list.append('require-2fa')
Checking your feature
In Python code
The FeatureManager's has
method checks see if the feature exists. The has
method takes in the feature's name, the objects that correspond to the scope of
the feature (i.e. an organization for an organization level feature or
a project for a project level feature), and the actor (aka user). Here's an example
Organization feature check:
if features.has('organizations:test-feature', obj, actor=user):
feature_list.append('test-feature')
The example code only adds the feature to the feature_list
if that feature is enabled for
the organization and the type of user given. Note that when we give the feature
to the frontend, we remove the scope prefix, and
our 'organizations:test-feature'
becomes 'test-feature'
.
In JavaScript
In order to check a feature flag in JavaScript, your feature flag will need
api_expose=True
.
Declarative features with the Feature component
React uses a declarative programming paradigm. As such, we have a utility component that we use to hide components based on the feature flags available to a organization/project
import Feature from 'sentry/components/acl/feature';
const toRender = (
<Feature features={['test-feature']}>
<MyComponentToFlag />
</Feature>
);
Imperative feature flag checks
There are some exceptions when React components are generated imperatively (e.g.
headers/columns for Tables). In difficult times like these, the Organization
/ Project
object has a array of the feature flags, which you can use in this
way:
const {organization} = this.props;
// Method 2
organization.features.includes('test-feature'); // evals to True/False
Enabling features in development
In Sentry you can run sentry devserver
to view your changes in development
mode. If you would like to view a change behind a feature flag, you will need to
open the file ~/.sentry/sentry.conf.py
on your local machine. This file
contains your local settings for the sentry application, and can be viewed and
edited. If you would like to toggle a flag on or off, add this to your
configuration file:
SENTRY_FEATURES['organizations:test-feature'] = True
Alternatively, you can test Flagpole features by setting custom options locally. See the Flagpole Local Development docs for more information on this.
Enabling your feature in production
Feature flags are declared in Sentry's codebase. For self-hosted users, those
flags are then configured via sentry.conf.py
. For Sentry's SaaS deployment,
you have the choice of using an option backed rollout via Options Automator with Flagpole,
or by writing a custom feature flag handler.
Flagpole is Sentry's internal feature flagging library, allowing a feature with multiple target segments and condition filters to be defined in YAML within Options Automator.
Options based features [DEPRECATED] allow a feature to be rolled out to a specific subset of LA orgs, a percentage of EA orgs, and/or a percentage of all orgs. These can be used in high scale situations, and are generally preferred over customer feature handlers. This strategy predates Flagpole, which is the new standard way to define an option-backed feature flag.
After launch (Graduation)
After your feature has been mainlined and is available for all customers on sentry.io, you have a few potential paths:
- If the feature cannot be disabled, or you don't need to conditionally disable the feature, remove the feature flag and all related checks from the Sentry code base. If necessary, also remove references to the feature from the self-hosted, getsentry, and options-automator repositories.
- If the feature will only be available to SaaS customers on specific plans, you
need to add your feature flag to the appropriate plans and update feature
handlers (see below). You should also move the feature flag from
temporary.py
topermanent.py
and update the default value to enable the feature in self-hosted instances.
Getsentry feature handlers
Getsentry contains a variety of feature handlers that override the
SENTRY_FEATURES
map, which are defined in the features.py
file.
Plan specific features
If your feature is available for a subset of plans (eg. only on Business plans) you need to:
- Disable the feature in
getsentry/conf/settings/defaults.py
by updatingSENTRY_FEATURES
. - Add your feature to the appropriate plan feature list.
- Update
SubscriptionPlanFeatureHandler
to handle your feature.
Custom handlers
If your feature requires additional logic to become active, you can also implement a feature handler in getsentry. For example, you can create a feature flag that is backed by options. follow these steps:
- Disable the feature in
getsentry/conf/settings/defaults.py
by updatingSENTRY_FEATURES
. - Add a new feature handler class in
getsentry/features.py
that determines availability of the feature based on the organization or actor. - Register the handler at the bottom of
getsentry/features.py
.