Settings

../_images/overview-features.png

If you are looking for installation instructions, please refer to Install Kinto.

Kinto is built to be highly configurable. As a result, the related configuration can be verbose, but don’t worry, all configuration flags are listed below.

Environment variables

In order to ease deployment or testing strategies, Kinto reads settings from environment variables, in addition to .ini files.

The environment variables are exactly the same as the settings, but they are capitalised and . are replaced by _.

For example, kinto.storage_backend is read from environment variable KINTO_STORAGE_BACKEND (if defined of course).

All settings are read first from the environment variables, then from application .ini, and finally from internal defaults.

Feature settings

Setting name Default What does it do?
kinto.readonly False If set to true, the whole service will reject every write operation. Note that with this option, the default bucket cannot be used and request will be rejected with a 405 Method Not Allowed error response.
kinto.batch_max_requests 25 The maximum number of requests that can be sent to the batch endpoint.
kinto.paginate_by None The maximum number of items to include on a response before enabling pagination. If set to None, no pagination will be used. It is recommended to set-up pagination if the server is under high load. If not defined, a collection response cannot contain more elements than defined by the kinto.storage_max_fetch_size setting.
kinto.<object-type>_id_generator UUID4 The Python dotted location of the generator class that should be used to generate identifiers on a POST endpoint. Object type is one of bucket, collection, group, record. See How to define and use a custom ID generator?.
kinto.experimental_collection_schema_validation False Experimental: Allow definition of JSON schema at the collection level, in order to validate submitted records. It is marked as experimental because the API might be subject to changes.
kinto.experimental_permissions_endpoint False Experimental: Add a new /permissions endpoint to let the user grab the list of objects (buckets, collections, groups, records) on which they have read or write permission. It is marked as experimental because the API might be subject to changes.
kinto.trailing_slash_redirect_enabled True Try to redirect resources removing slash or adding it for the root URL endpoint: /v1 redirects to /v1/ and /buckets/default/ to /buckets/default. No redirections are made when turned off.
kinto.heartbeat_timeout_seconds 10 The maximum duration of each heartbeat entry, in seconds.

Note

kinto.readonly will disable every endpoint that is not accessed with one of GET, OPTIONS, or HEAD HTTP methods. Requests will receive a 405 Method Not Allowed error response.

The cache backend will still needs read-write privileges, in order to cache OAuth authentication states and tokens for example.

Backends

Kinto relies on three types of backends: storage, cache and permission. The settings names have a different prefix for each.

For each of them, the supported services are currently PostgreSQL, Redis, and Memory. Memcached is also available as a cache backend.

Storage

Setting name Default What does it do?
kinto.storage_backend kinto.core.storage.memory The Python dotted location of the storage backend to use.
kinto.storage_url '' The URL to use to authenticate to the storage backend. e.g. redis://localhost:6378/1 or postgres://user:pass@database/db
kinto.storage_max_fetch_size 10000 The maximum number of items that can be returned by one request to the storage backend. If no pagination is enabled, this is the maximum number of items that can be stored in a collection (otherwise some of them won’t be returned). With pagination enabled, this limitation doesn’t apply.
kinto.storage_pool_size 25 The size of the pool of connections to use for the storage backend.
kinto.storage_max_overflow 5 Number of connections that can be opened beyond pool size.
kinto.storage_pool_recycle -1 Recycle connections after the given number of seconds has passed.
kinto.storage_pool_timeout 30 Number of seconds to wait before giving up on getting a connection from the pool.
kinto.storage_max_backlog -1 Number of threads that can be in the queue waiting for a connection.
kinto.storage_backend = kinto.core.storage.postgresql
kinto.storage_url = postgres://postgres:postgres@localhost/postgres

# Safety limit while fetching from storage
# kinto.storage_max_fetch_size = 10000

# Control number of pooled connections
# kinto.storage_pool_size = 50

Cache

Setting name Default What does it do?
kinto.cache_backend kinto.core.cache.memory The Python dotted location of the cache backend to use.
kinto.cache_url '' The URL to use to authenticate to the cache backend. e.g. redis://localhost:6378/1 or postgres://user:pass@database/db
kinto.cache_prefix '' A prefix added to each key. Useful when having multiple Kinto using the same cache database.
kinto.cache_max_size_bytes 524288 The maximum size the memory cache backend will allow per process. (in bytes)
kinto.cache_pool_size 25 The size of the pool of connections to use for the cache backend.
kinto.cache_max_overflow 5 Number of connections that can be opened beyond pool size.
kinto.cache_pool_recycle -1 Recycle connections after the given number of seconds has passed.
kinto.cache_pool_timeout 30 Number of seconds to wait before giving up on getting a connection from the pool.
kinto.cache_max_backlog -1 Number of threads that can be in the queue waiting for a connection.
kinto.cache_hosts '' The space separated list of Memcached hosts.

For PostgreSQL

kinto.cache_backend = kinto.core.cache.postgresql
kinto.cache_url = postgres://postgres:postgres@localhost/postgres

# Control number of pooled connections
# kinto.cache_pool_size = 50

For Memcached

You would need to install the memcached dependencies: pip install kinto[memcached]

kinto.cache_backend = kinto.core.cache.memcached
kinto.cache_hosts = 127.0.0.1:11211 127.0.0.2:11211

Permissions

Setting name Default What does it do?
kinto.permission_backend kinto.core.permission.memory The Python dotted location of the permission backend to use.
kinto.permission_url '' The URL to use to authenticate to the permission backend. e.g. redis://localhost:6379/1
kinto.permission_pool_size 25 The size of the pool of connections to use for the permission backend.
kinto.permission_max_overflow 5 Number of connections that can be opened beyond pool size.
kinto.permission_pool_recycle -1 Recycle connections after the given number of seconds has passed.
kinto.permission_pool_timeout 30 Number of seconds to wait before giving up on getting a connection from the pool.
kinto.permission_max_backlog -1 Number of threads that can be in the queue waiting for a connection.
kinto.permission_backend = kinto.core.permission.postgresql
kinto.permission_url = postgres://postgres:postgres@localhost/postgres

# Control number of pooled connections
# kinto.permission_pool_size = 50

Bypass permissions with configuration

Permissions are usually retrieved from the permission backend. However, it is also possible to configure them from settings, and it will bypass the permission backend.

For example, for a resource named “bucket”, the following setting will enable authenticated people to create bucket records:

kinto.bucket_create_principals = system.Authenticated

The format of these permission settings is <resource_name>_<permission>_principals = comma,separated,principals.

Scheme, host, and port

By default, Kinto relies on WSGI for underlying details like host, port, or request scheme. Tuning these settings may be necessary when the application runs behind proxies or load balancers, but most implementations (such as uWSGI) provide adequate values automatically.

That said, if ever these items need to be controlled at the application layer, the following settings are available:

Check the behaviour of the server with the url value returned in the hello view.

Setting name Default What does it do?
kinto.http_host None The HTTP Host used by Kinto to refer to itself. If set to None, the HTTP host is read from HTTP headers or WSGI environment.
kinto.http_scheme None The HTTP scheme used by Kinto to refer to itself. If set to None, the HTTP scheme is read from the HTTP headers or WSGI environment.
# kinto.http_scheme = https
# kinto.http_host = production.server.com:7777

Logging and Monitoring

Setting name Default What does it do?
kinto.statsd_backend kinto.core.statsd The Python dotted location of the StatsD module that should be used for monitoring. Useful to plug custom implementations like Datadog™.
kinto.statsd_prefix kinto The prefix to use when sending data to statsd.
kinto.statsd_url None The fully qualified URL to use to connect to the statsd host. e.g. udp://localhost:8125

Standard Logging

With the following configuration, all logs are redirected to standard output (See 12factor app):

[loggers]
keys = root

[handlers]
keys = console

[formatters]
keys = generic

[logger_root]
level = DEBUG
handlers = console

[handler_console]
class = StreamHandler
args = (sys.stdout,)
level = NOTSET
formatter = generic

[formatter_generic]
format = %(asctime)s,%(msecs)03d %(levelname)-5.5s [%(name)s] %(message)s
datefmt = %H:%M:%S

Example output:

16:18:57,179 INFO  [root] Running kinto 6.1.0.dev0.
16:19:00,729 INFO  [request.summary]
16:19:22,232 WARNI [kinto.core.authorization] Permission not granted.
16:19:22,238 INFO  [request.summary]

Colored Logging

[formatters]
keys = color

[formatter_color]
class = logging_color_formatter.ColorFormatter

Example output:

../_images/color-formatter.png

JSON Logging

Using a JSON logging formatter, like this one, it is possible to output logs as JSON:

[formatters]
keys = json

[formatter_json]
class = kinto.core.JsonLogFormatter

Example output:

{"Pid": 19240, "Type": "root", "Timestamp": 1489067815875679744, "Severity": 6, "Hostname": "pluo", "Logger": "%", "EnvVersion": "2.0", "Fields": {"message": "Running kinto 6.1.0.dev0."}}
{"Pid": 19240, "Type": "root", "Timestamp": 1489067817834153984, "Severity": 4, "Hostname": "pluo", "Logger": "%", "EnvVersion": "2.0", "Fields": {"perm": "read", "userid": "basicauth:cbd3731f18c97ebe1d31d9846b5f1b95cf8eeeae586e201277263434041e99d1", "message": "Permission not granted.", "uri": "/buckets/123"}}

Handling exceptions with Sentry

Requires the raven package.

Sentry logging can be enabled as explained in official documentation.

Note

The application sends an INFO message on startup (mainly for setup check).

Monitoring with StatsD

Requires the statsd package.

StatsD metrics can be enabled (disabled by default):

kinto.statsd_url = udp://localhost:8125
# kinto.statsd_prefix = kinto-prod

Monitoring with New Relic

Requires the newrelic package.

Setting name Default What does it do?
kinto.newrelic_config None Location of the newrelic configuration file.
kinto.newrelic_env dev The environment the server runs into

New Relic can be enabled (disabled by default):

kinto.newrelic_config = /location/of/newrelic.ini
kinto.newrelic_env = prod

Authentication

By default, Kinto relies on Basic Auth to authenticate users.

User registration is not necessary. A unique user idenfier will be created for each username:password pair.

Kinto is compatible with any third-party authentication service.

Setting name Default What does it do?
kinto.userid_hmac_secret '' The secret used to create the user ID from a username:password pair. This value should be unique to each instance and kept secret.
multiauth.policies basicauth The list of authentication policies aliases that are enabled. Each alias is configuration using dedicated settings as explained below.
multiauth.authorization_policy kinto.authorization.AuthorizationPolicy Python dotted path the authorisation policy to use for the permission mecanism.

Authentication setup

Kinto relies on pyramid multiauth to initialise authentication.

Therefore, any authentication policy can be specified through configuration.

In the following example, Basic Auth, Persona, and IP Auth are all enabled:

multiauth.policies = basicauth pyramid_persona ipauth

multiauth.policy.ipauth.use = pyramid_ipauth.IPAuthentictionPolicy
multiauth.policy.ipauth.ipaddrs = 192.168.0.*
multiauth.policy.ipauth.userid = LAN-user
multiauth.policy.ipauth.principals = trusted

Permission handling and authorisation mechanisms are specified directly via configuration. This allows for customised solutions ranging from very simple to highly complex.

Basic Auth

basicauth is enabled via multiauth.policies by default.

multiauth.policies = basicauth

By default an internal Basic Auth policy is used, where any login:password pair will be accepted, meaning that no account creation is required.

In order to replace it by another one:

multiauth.policies = basicauth
multiauth.policy.basicauth.use = myproject.authn.BasicAuthPolicy

Custom Authentication

Using the various Pyramid authentication packages, it is possible to plug in any kind of authentication.

Firefox Accounts

Enabling Firefox Accounts consists of including kinto_fxa in configuration, mentioning fxa among policies, and providing appropriate values for OAuth2 client settings.

See mozilla-services/kinto-fxa.

Plugins

It is possible to extend the default Kinto behaviors by using “plugins”.

The list of plugins to load at startup can be specified in the settings, as a list of Python modules:

kinto.includes = kinto.plugins.default_bucket
                 kinto.plugins.history
                 kinto.plugins.admin
                 kinto-attachment
                 custom-myplugin
Built-in plugins What does it do?
kinto.plugins.default_bucket It enables a personal bucket default, where collections are created implicitly (more details).
kinto.plugins.history It tracks every action performed on objects within a bucket (more details).
kinto.plugins.admin It is a Web admin UI to manage data from a Kinto server. (more details).
kinto.plugins.flush Adds an endpoint to completely remove all data from the database backend for testing/staging purposes. (more details).

There are many available packages in Pyramid ecosystem, and it is straightforward to build one, since the specified module must just define an includeme(config) function.

See our list of community plugins.

See also: How to write a Kinto plugin? for more in-depth informations on how to create your own plugin.

Pluggable components

Pluggable components can be substituted from configuration files, as long as the replacement follows the original component API.

kinto.logging_renderer = your_log_renderer.CustomRenderer

This is the simplest way to extend Kinto, but will be limited to its existing components (cache, storage, log renderer, …).

In order to add extra features, including external packages is the way to go!

Notifications

Kinto has a notification system, and the event listeners are configured using the event_handlers setting, which takes a list of aliases.

In the example below, the Redis listener is activated and will send events data in the queue Redis list.

kinto.event_listeners = redis

kinto.event_listeners.redis.use = kinto_redis.listeners
kinto.event_listeners.redis.url = redis://localhost:6379/0
kinto.event_listeners.redis.pool_size = 5
kinto.event_listeners.redis.listname = queue

Filtering

It is possible to filter events by action and/or types of object. By default actions create, update and delete are notified for every kinds of objects.

kinto.event_listeners.redis.actions = create
kinto.event_listeners.redis.resources = bucket collection

Third-party

Enabling push notifications to clients consists in enabling an event listener that will be in charge of forwarding events data to remote clients.

A Kinto plugin was made using the Pusher (commercial) service. See How to setup push notifications using WebSockets?.

Cross Origin requests (CORS)

Kinto supports CORS out of the box. Use the cors_origins setting to change the list of accepted origins.

Setting name Default What does it do?
kinto.cors_origins * This List of CORS origins to support on all endpoints. By default allow all cross origin requests.

Backoff indicators

In order to tell clients to back-off (on heavy load for instance), the following flags can be used. Read more about this at Backoff header on heavy load.

Setting name Default What does it do?
kinto.backoff None The Backoff time to use. If set to None, no backoff flag is sent to the clients. If set, provides the client with a number of seconds during which it should avoid doing unnecessary requests.
kinto.backoff_percentage None If specified, then send the backoff header with probability equal to the backoff_percentage. This should be a number between 0 and 100. This setting will have no effect if the backoff is None.
kinto.retry_after_seconds 30 The number of seconds after which the client should issue requests.
# kinto.backoff = 10
kinto.retry_after_seconds = 30

Similarly, the end of service date can be specified by using these settings.

Setting name Default What does it do?
kinto.eos None The End of Service Deprecation date. If the date specified is in the future, an alert will be sent to clients. If it’s in the past, the service will be declared as decomissionned. If set to None, no End of Service information will be sent to the client.
kinto.eos_message None The End of Service message. If set to None, no End of Service message will be sent to the clients.
kinto.eos_url None The End of Service information URL.
kinto.eos = 2015-01-22
kinto.eos_message = "Client is too old"
kinto.eos_url = http://website/info-shutdown.html

Enabling or disabling endpoints

Specific resource operations can be disabled.

To do so, a setting key must be defined for the disabled resources endpoints:

'kinto.{endpoint_type}_{resource_name}_{method}_enabled'

Where:

  • endpoint_type is either collection (plural, e.g. /buckets) or record (single, e.g. /buckets/abc);
  • resource_name is the name of the resource (e.g. bucket, group, collection, record);
  • method is the http method (in lower case) (e.g. get, post, put, patch, delete).

Important

These settings are confusing because of the naming used in kinto.core. Especially collection and record are endpoint types in kinto.core and resource names in Kinto. We want to solve this in #710.

For example, to disable the POST on the list of buckets and DELETE on single records, the following setting should be declared in the .ini file:

kinto.collection_bucket_post_enabled = false
kinto.record_record_delete_enabled = false

Activating the permissions endpoint

The Permissions endpoint is used to get a list of all user accessible objects in the server as well as their permissions. It enables applications such as the kinto-admin to discover what the user is allowed to do and which data can be managed.

kinto.experimental_permissions_endpoint = true

Then, issue a GET request to the /permissions endpoint to get the list of the user permissions on the server ressources.

Client caching

In addition to per-collection caching, it is possible to add cache control headers for every Kinto object. The client (or cache server or proxy) will use them to cache the collection records for a certain amount of time, in seconds.

The setting can be set for any kind of object (bucket, group, collection, record), and concerns GET requests (GET /buckets, GET /buckets/{}/groups, GET /buckets/{}/collections, GET /buckets/{}/collections/{}/records).

# kinto.bucket_cache_expires_seconds = 3600
# kinto.group_cache_expires_seconds = 3600
# kinto.collection_cache_expires_seconds = 3600
kinto.record_cache_expires_seconds = 3600

It can also be specified per bucket or collections for records:

kinto.blog_record_cache_expires_seconds = 30
kinto.blog_articles_record_cache_expires_seconds = 3600

If set to 0 then the resource becomes uncacheable (no-cache).

Note

In production, Nginx can act as a cache-server using those client cache control headers.

Project information

Setting name Default What does it do?
kinto.version_json_path ./version.json Location of the file containing the information to be shown in the version endpoint.
kinto.error_info_link https://github.com/kinto/kinto/issues/ The HTTP link returned when uncaught errors are triggered on the server.
kinto.project_docs https://kinto.readthedocs.io The URL where the documentation of the Kinto instance can be found. Will be returned in the hello view.
kinto.project_version '' The version of the project. Will be returned in the hello view. By default, this is the major version of Kinto.
kinto.version_prefix_redirect_enabled True By default, all endpoints exposed by Kinto are prefixed by a version number. If this flag is enabled, the server will redirect all requests not matching the supported version to the supported one.

Example:

kinto.project_docs = https://project.readthedocs.io/
# kinto.project_version = 1.0

Application profiling

It is possible to profile the stack while its running. This is especially useful when trying to find bottlenecks.

Update the configuration file with the following values:

kinto.profiler_enabled = true
kinto.profiler_dir = /tmp/profiling

Run some request on the server (for example):

http GET http://localhost:8888/v1/

Render execution graphs using GraphViz:

sudo apt-get install graphviz
pip install gprof2dot
gprof2dot -f pstats POST.v1.batch.000176ms.1427458675.prof | dot -Tpng -o output.png