.. _schemaregistry_intro:

Schema Management
=================

|sr-long| provides a serving layer for your metadata. It provides a RESTful interface for storing and retrieving |avro-tm| :ref:`schemas <schema_registry_tutorial_definition>`. It stores a versioned history of all schemas based on a specified :ref:`subject name strategy <sr-avro-subject-name-strategy>`, provides multiple :ref:`compatibility settings <schema_evolution_and_compatibility>` and allows evolution of schemas according to the configured compatibility settings and expanded Avro support. It provides serializers that plug into |ak-tm| clients that handle schema storage and retrieval for |ak| messages that are sent in the Avro format.

|sr| lives outside of and separately from your |ak| brokers. Your producers and consumers still talk to |ak| to publish and read data (messages) to topics. Concurrently, they can also talk to |sr| to send and retrieve schemas that describe the data models for the messages.

.. figure:: ../images/schema-registry-and-kafka.png
   :align: center

   |sr-long| for storing and retrieving Avro schemas

|sr| is a distributed storage layer for Avro Schemas which uses |ak| as its underlying storage mechanism. Some key design decisions:

* Assigns globally unique ID to each registered schema. Allocated IDs are guaranteed to be monotonically increasing but not necessarily consecutive.
* |ak| provides the durable backend, and functions as a write-ahead changelog for the state of |sr| and the schemas it contains.
* |sr| is designed to be distributed, with single-primary architecture, and |zk|/|ak| coordinates primary election (based on the configuration).

.. tip:: To see a working example of |sr|, check out :ref:`Confluent Platform demo <cp-demo>`. The demo shows you how to deploy a
         |ak| streaming ETL, including |sr|, using KSQL for stream processing.

Schemas, Subjects, and Topics
-----------------------------

First, a quick review of terms and how they fit in the context of |sr|: what is a |ak| `topic` versus a `schema` versus a `subject`.

.. include:: includes/terms-schemas-topics.rst

The :ref:`schema_registry_tutorial` shows an example of a :ref:`schema definition <schema_registry_tutorial_definition>`.

Starting with |cp| 5.2.0, you can use |crep-full| to :ref:`migrate schemas <schemas_migrate_overview>` from one |sr| to another, and automatically rename subjects on the target registry.

Avro Background
---------------

When sending data over the network or storing it in a file, we need a
way to encode the data into bytes. The area of data serialization has
a long history, but has evolved quite a bit over the last few
years. People started with programming language specific serialization
such as Java serialization, which makes consuming the data in other
languages inconvenient. People then moved to language agnostic formats
such as JSON.

However, formats like JSON lack a strictly defined format, which has two significant drawbacks:

1. **Data consumers may not understand data producers:** The lack of structure makes consuming data in these formats
   more challenging because fields can be arbitrarily added or removed, and data can even be corrupted.  This drawback
   becomes more severe the more applications or teams across an organization begin consuming a data feed: if an
   upstream team can make arbitrary changes to the data format at their discretion, then it becomes very difficult to
   ensure that all downstream consumers will (continue to) be able to interpret the data.  What's missing is a
   "contract" (cf. schema below) for data between the producers and the consumers, similar to the contract of an API.
2. **Overhead and verbosity:** They are verbose because field names and type information have to be explicitly
   represented in the serialized format, despite the fact that are identical across all messages.

A few cross-language serialization libraries have emerged that require the data structure to be formally defined by
some sort of schemas. These libraries include `Avro <http://avro.apache.org>`_,
`Thrift <http://thrift.apache.org>`_, and `Protocol Buffers <https://github.com/google/protobuf>`_.  The advantage of
having a schema is that it clearly specifies the structure, the type and the meaning (through documentation) of the
data.  With a schema, data can also be encoded more efficiently.
In particular, we recommend Avro which is supported in |cp|.

An Avro schema defines the data structure in a JSON format.

The following is an example Avro schema that specifies a user record with two fields: ``name`` and ``favorite_number``
of type ``string`` and ``int``, respectively.

.. sourcecode:: json

    {"namespace": "example.avro",
     "type": "record",
     "name": "user",
     "fields": [
         {"name": "name", "type": "string"},
         {"name": "favorite_number",  "type": "int"}
     ]
    }

You can then use this Avro schema, for example, to serialize a Java object (POJO) into bytes, and deserialize these
bytes back into the Java object.

One of the interesting things about Avro is that it not only requires
a schema during data serialization, but also during data
deserialization. Because the schema is provided at decoding time,
metadata such as the field names don't have to be explicitly encoded
in the data. This makes the binary encoding of Avro data very compact.


Schema ID Allocation
--------------------

Schema ID allocation always happens in the primary node and Schema IDs are always
monotonically increasing.

If you are using |ak| primary election, the Schema ID is always based off the last ID that was
written to |ak| store. During a primary re-election, batch allocation happens only after the new
primary has caught up with all the records in the store ``<kafkastore.topic>``.

If you are using |zk| primary election, ``/<schema.registry.zk.namespace>/schema_id_counter``
path stores the upper bound on the current ID batch, and new batch allocation is triggered by both primary election and exhaustion of the current batch. This batch allocation helps guard against potential zombie-primary scenarios, (for example, if the previous primary had a GC pause that lasted longer than the |zk| timeout, triggering primary reelection).

.. _schemaregistry_design:

|ak| Backend
-------------

.. include:: includes/backend.rst


.. _schemaregistry_single_master:

Single Primary Architecture
---------------------------
|sr| is designed to work as a distributed service using single primary architecture. In this configuration, at most one |sr| instance is the primary at any given moment (ignoring pathological 'zombie primaries'). Only the primary is capable of publishing writes to the underlying |ak| log, but all nodes are capable of directly serving read requests. Secondary nodes serve registration requests indirectly by simply forwarding them to the current primary, and returning the response supplied by the primary.
Prior to |sr| version 4.0, primary election was always coordinated through |zk|.
Primary election can now optionally happen via |ak| group protocol as well.

.. note::

         Please make sure not to mix up the election modes amongst the nodes in same cluster.
         This will lead to multiple primaries and issues with your operations.

----------------------------------
|ak| Coordinator Primary Election
----------------------------------

.. figure:: ../images/schema-registry-design-kafka.png
   :align: center

   |ak| based Schema Registry

|ak| based primary election is chosen when ``<kafkastore.connection.url>`` is not configured and
has the |ak| bootstrap brokers ``<kafkastore.bootstrap.servers>`` specified. The kafka group
protocol, chooses one amongst the primary eligible nodes ``master.eligibility=true`` as the primary. |ak|-based primary
election can be used in cases where |zk| is not available, for example for hosted or cloud
|ak| environments, or if access to |zk| has been locked down.

---------------------
|zk| Primary Election
---------------------

.. figure:: ../images/schema-registry-design.png
   :align: center

   ZooKeeper based Schema Registry

|zk| primary election is chosen when |zk| URL is specified in |sr| config
``<kafkastore.connection.url>``.
The current primary is maintained as data in the ephemeral node on the ``/<schema.registry.zk.namespace>/schema_registry_master`` path in |zk|. |sr| nodes listen to data change and deletion events on this path, and shutdown or failure of the primary process triggers each node with ``master.eligibility=true`` to participate in a new round of election. Primary election is a simple 'first writer wins' policy: the first node to successfully write its own data to ``/<schema.registry.zk.namespace>/schema_registry_master`` is the new primary.


|sr| is also designed for multi-colo configuration. See :ref:`schemaregistry_mirroring` for more details.

.. _sr-high-availability-single-primary:

------------------------------------------
High Availability for Single Primary Setup
------------------------------------------
         
Many services in |cp| are effectively stateless (they store state in |ak| and
load it on demand at start-up) and can redirect requests automatically. You can
treat these services as you would deploying any other stateless application and
get high availability features effectively for free by deploying multiple
instances. Each instance loads all of the |sr| state so any node can serve a
READ, and all nodes know how to forward requests to the primary for WRITEs.

A common pattern is to put the instances behind a single virtual IP or round
robin DNS such that you can use a single URL in the ``schema.registry.url``
configuration but use the entire cluster of |sr| instances. This also makes it
easy to handle changes to the set of servers without having to reconfigure and
restart all of your applications. The same strategy applies to REST proxy or
|kconnect-long|.

A simple setup with just a few nodes means |sr| can fail over easily
with a simple multi-node deployment and single primary election protocol.

.. _schemas_migrate_overview:

Migrate Schemas
---------------

Starting with |cp| 5.2.0, you can use :ref:`connect_replicator` to migrate
schemas to another |sr|, which is either self-managed or in :ref:`Confluent Cloud <schemaregistry_migrate>`.

.. include:: ../includes/replicator-topic-rename.rst

The property ``topic.rename.format`` is described in :ref:`rep-destination-topics` under :ref:`connect_replicator_config_options` for |crep|.

  
Suggested Reading
-----------------

- Tutorial, including quick install and examples workflow: :ref:`schema_registry_tutorial`
- Blog post: `Schemas, Contracts, and Compatibility <https://www.confluent.io/blog/schemas-contracts-compatibility>`_
- Blog post: `17 Ways to Mess Up Self-Managed Schema Registry <https://www.confluent.io/blog/17-ways-to-mess-up-self-managed-schema-registry>`_

.. toctree::
   :maxdepth: 1
   :hidden:

   installation/index
   schema_registry_tutorial
   avro
   using
   monitoring
   singledc
   multidc
   Connect Schema Registry to Confluent Cloud <https://docs.confluent.io/cloud/current/cp-component/schema-reg-cloud-config.html>
   installation/migrate
   security
   ../../confluent-security-plugins/schema-registry/introduction
   develop/index
   serializer-formatter
   schema-deletion-guidelines
   connect
   changelog