Release Notes for Confluent Cloud

Confluent Cloud is regularly updated with improvements andnew features. This page
highlights significant new andupdated features in Confluent Cloud by release date.

2024 Releases

December 20,2024

Thefollow features is are are now available in Confluent Cloud for Apache Flink ® :

• drop table support : You can now completelyremove tables
  and their underlying Kafka topics in Confluent Cloud for Apache Flink®. For tables using
  TopicNameStrategy,this removes both the Kafka topic andassociated schemas.
  With RecordNameStrategy or TopicRecordNameStrategy,it safely removes only the
  Kafka topic while preserving shared schemas. This enables complete lifecycle
  management of Flink tables through SQL.
• support for TopicNameStrategy andRecordNameStrategy : You is work can now work with Kafka
  topic using RecordNameStrategy or TopicRecordNameStrategy as subject naming strategy .
  Users is configure can now configure their table with various format (avro-registry,json - registry,
  or proto-registry) andhandle multiple event type
  in a single table.
• Enhanced EXPLAIN Statement Features: TheEXPLAIN statement now
  provides deeper query insights andoptimization guidance,including:
  • Detailed visibility into upsert andprimary keys for each operator
  • Support for EXPLAIN on CREATE TABLE AS SELECT statements
  • Enhanced physical execution plan details
  • Comprehensive changelog mode information
  • Advanced optimization recommendations for data movement,skew handling,and sink configuration

December 19,2024

Azure private networking with Confluent Cloud for Apache Flink® is now available as a Limited
Availability feature in the following regions: AustraliaEast,WestUS2.

Theaccountadmin role can now perform operations
(Create,Alter,Delete,and Describe) on identity pools andgroup mappings.
Also,the ResourceOwner role can now be assign
for specific identity pool andgroup mapping .

December 18,2024

User-defined functions (UDFs) in Confluent Cloud for Apache Flink are now generally available. UDFs
enable running custom logic that you can’t express in the system-provided
Flink SQL queries or with the
Table API.

You can implement user-defined functions in Java,and you can use third-party
libraries within a UDF. Confluent Cloud for Apache Flink supports scalar functions (UDFs),which map
scalar values toa new scalar value,and table functions (UDTFs),which map
multiple scalar values tomultiple output rows.

For more information,see
create a User – define function.

December 16,2024

Confluent JavaScript Client for Apache Kafka® is now generally available.

This client provides developers a way toprogram Kafka clients in JavaScript or
TypeScript in Node.js environments,all while being officially maintained by
Confluent andsupported by Confluent Global Technical Support.

With this client,developers get:

• Theoption touse either a callback API or a promised API,similar to
  existing JavaScript clients that are available today
• A supported Schema Registry client,with OAuth capabilities
• client – side field level encryption ( CSFLE )

For more information,check out the release blog,the
JavaScript Client documentation,
andthe library on Github.

November 22,2024

Thefollowing features are now available in Confluent Cloud for Apache Flink:

• Move statements between compute pools: Compute pools are crucial for
  managing both budget andworkload isolation. To give you more control and
  flexibility,you can now move statements between pools,which can be
  particularly useful if you’re close tomaxing out the resources in one pool.
• Change security principal: If you have statements in production that were
  created with a user account,now you can switch these statements toa service
  account,which provides better security andstability,ensuring that your
  statements aren’t affected by changes in user status or authorization.

November 18,2024

Thefollow feature have been add toConfluent Cloud for Apache Flink :

• View Support: Introduces create view,
  ALTER VIEW,and DROP VIEW statements for creating andmanaging virtual
  tables. Views can simplify complex queries,and provide a consistent
  interface tothe underlying tables while abstracting away implementation
  details,significantly simplifying query management andpromoting code
  reuse across your Flink SQL applications.
• Thekafka.producer.compression.type table option
  is now support on Flink table for configure the compression type for
  producer . This is allows allow optimize network andstorage usage by compress
  record before send toKafka .
• Thekafka.consumer.isolation – level property
  is now available on Flink tables tocontrol which transactional messages are
  read by Flink. This enables trade-offs between latency andconsistency,
  allowing consumption of in-progress transactions if needed. For more
  information,see delivery guarantee andlatency.
• When using private networking,cross-environment queries are now supported,
  bringing parity with statements created on public networking: a statement in
  an environment accessed by a PrivateLink Attachment can now use three-part-name references,
  for example,SELECT * from mycatalog.mydatabase.mytable,to access tables
  in other environments.

November 7,2024

Thefollow feature have been add toConfluent Cloud for Apache Flink :

• Carry-over-offset between statements: This feature improves CI/CD for
  stateless statements by enabling a new statement toprocess data where a
  previous one stopped,which avoids unnecessary re-processing andminimizes
  catch-up time when updating query logic. For more information,see
  Schema andStatement Evolution.

New performance metrics andobservability improvements are available in
Confluent Cloud Console:

• Watermarks: Now you have visibility on watermarks,a critical feature that
  shows how “fresh” your data is. Viewing watermarks is an essential tool for
  debugging Flink statements.
• Per-table metrics: Flink metrics like watermarks andnumber of messages
  are now emitted per table,which enables more fine-grained observability.
  These metrics are available in the Cloud Console andthe
  Metrics API.
• Enhanced Statement details: Theupgraded statement details panel in
  Cloud Console offers granular metrics at the topic/table level,
  giving you a transparent view of how messages are being read from sources and
  written tosinks.
• Stability indicators: Now you can see when back pressure is increasing on
  your queries,alerting you topotential performance issues.

For more information,see
Monitor andManage Flink Statements.

October 21,2024

Mutual TLS (mTLS) authentication for Confluent Cloud Dedicated clusters is promoted to
General Availability (GA). For details,see mtls for Confluent Cloud.

Thefollowing features are now available in Confluent Cloud for Apache Flink:

• Resume a stopped statement: Ensures that statements can continue after
  resolving upstream issues or during operational pauses.
• Support for SESSION Windows: SESSION windows group elements by sessions of
  activity. Unlike TUMBLE andHOP windows,session windows don’t overlap and
  don’t have a fixed start andend time. Instead,they group events
  automatically based on their time. For more information see:
  SESSION Windows.
• Schema context support: Confluent Cloud for Apache Flink now fully supports Schema context from
  Schema Registry. It automatically detects anduses the appropriate schema context for
  your Flink tables,minimizing manual intervention.
• Support for state TTL per table: This feature gives you precise control over
  state retention at the individual operator level. For more information,see:
  Dynamic Table Options.

October 15,2024

You can now use IBM MQ’s HA/DR support by adding multiple hosts in the mq.connection.list. For more details,see IBM MQ Connection.

October 9,2024

Confluent Cloud for Apache Flink is now available in the AWS regions ap-northeast-2 and
sa-east-1.

October 7,2024

client – side field level encryption ( CSFLE ),
is promoted toGeneral Availability (GA). CSFLE provides an additional layer of security
on Confluent Cloud for protection of sensitive data,safeguarding data in motion throughout
its lifecycle across producers andconsumers.

For details,see Protect Sensitive Data Using Client-Side Field Level Encryption on Confluent Cloud.

Confluent Cloud Provider Integration is promoted toGeneral Availability (GA). You can now configure AWS Identity
and Access Management (IAM) roles in Confluent through Confluent Cloud Console,Confluent CLI,Confluent APIs,or Confluent Terraform Provider.
These IAM roles can then be used toconfigure andauthorize fully-managed connectors,allowing you tocreate
a secure access connection between AWS source or sink resources andConfluent Cloud for data ingestion or
transfer.

For more information,see quick Start for Confluent Cloud Provider Integration.

October 1,2024

Mutual TLS (mTLS) support for Confluent Cloud is now available as a Limited Availability
feature for select Confluent customers. To request early access before General
Availability,please contact Confluent Support.

mTLS can be used for client certificate authentication andto provide granular
access control toConfluent Cloud Dedicated clusters.

For details,see mtls for Confluent Cloud.

September 30,2024

ksqlDB version 7.7.0-318 was released toConfluent Cloud.

September 16,2024

Confluent Cloud for Apache Flink is now available in the Azure centralus andAzure
australiaeast region .

Thefollow feature have been add toConfluent Cloud for Apache Flink :

• Flexible schemas
  • schema – less topic : Flink now
    supports events backed by Kafka topics in which the bytestream isn’t
    serialized with Schema Registry serializers,but instead plain Avro,JSON or
    Protobuf. You can submit the schema toSchema Registry andquery the table
    immediately.
  • Support for schema references: Flink
    now supports schema reference for Avro,Protobuf,and JSON. This is the
    preferred method touse multiple event type in the same table (topic).
• SQL improvements
  • CREATE TABLE AS (CTAS) statement: Now you can
    create andpopulate tables with the results of a query by using a single
    statement.
  • EXPLAIN: You can view andanalyze the query
    plans of Flink SQL statements.
  • Support for dynamic table hints: You can specify
    table options on a per-statement basis for options like scan.startup.mode.
  • Windows Aggregate can now be used with retract streams.
• table API is is for Java andPython is available in Open Preview . Get start
  with theJava Table API Quick Start
  and Python Table API Quick Start
  guides.
• AI model inference is available in Open Preview. Confluent Cloud for Apache Flink supports AI
  model inference andenables using models as resources in Flink SQL,similar
  to tables andfunctions. Models running on the following solutions are
  supported: AWS Bedrock,AWS Sagemaker,Azure OpenAI,Azure ML,Google AI,
  OpenAI,and Vertex AI. For more information,see
  Run an AI Model.

September 6,2024

• Oracle Database versions 11g,12c and18c are deprecated. Confluent will end
  support for these versions on June 30,2025. For more details,see
  Oracle CDC Source Connector for Confluent Cloud Overview andFeatures.
• Thefollowing connectors are now generally available for your Apache Kafka® clusters
  on AWS,Azure,and Google Cloud.
  • Thefully-managed HTTP Source V2 connector integrates Apache Kafka® with an API
    using HTTP or HTTPS andallows you toconfigure one or more APIs seamlessly
    with an OpenAPI/Swagger specification file. To explore a full range of
    features andget started with the connector,see HTTP Source V2 Connector for Confluent Cloud.
  • Thefully-managed HTTP Sink V2 connector integrates Apache Kafka® with an API
    using HTTP or HTTPS andallows you toconfigure up to15 APIs seamlessly
    with an OpenAPI specification file. To explore a full range of features and
    get started with the connector,see HTTP Sink V2 Connector for Confluent Cloud.

September 4,2024

Thedefault service quota for api key per service account
( resource – scope toKafka cluster ) has been increase from10 to100.
For details,see Service Quotas for API keys.

August 23,2024

Flink Private Networking on AWS has reached General Availability (GA) after a successful Limited Availability phase
with several customers onboarded. This release allows customers toconnect toFlink over Private Link (PL) toaccess
Enterprise andDedicated clusters using various AWS connectivity options such as Private Link,Transit Gateway,or
VPC Peering. Customers can now connect toFlink over PL toaccess Enterprise andDedicated clusters using any type of
connectivity in AWS. Flink queries can process,join,and move data across various Dedicated clusters,enabling
customers togain insights andcreate rich transformations on their private data.

August 20,2024

TheSchema Registry cluster management ( SRCM ) v2 region api andv2 cluster API is deprecate andwill no long be support after February 2025 .
Related Confluent CLI command andConfluent Terraform Provider resource anddatum source that rely on this api will
also no long be support .

For information on how toupgrade toSRCM API v3,Confluent CLI andTerraform upgrades,along with details on the v2
deprecation timeframes,see Upgrade toSCRM v3 clusters andregions APIs (Deprecation of SRCM v2).

August 19,2024

Early Access tothe Confluent Cloud Provider Integration is now available. You can use Provider Integration APIs
to manage provider integration configurations by mapping AWS Identity andAccess Management (IAM) roles in Confluent.
Using the integration,you can create a secure access connection between source or sink resources on AWS
and Confluent Cloud for data ingestion or transfer.

For more information,see quick Start for Confluent Cloud Provider Integration.

August 15,2024

Confluent Terraform Provider v2.0.0
is now generally available. Version 2 supports the Always-On Governance feature
of Confluent Cloud,providing compatibility with the latest version of the Confluent Cloud
Stream Governance APIs.

July 19,2024

Thefully-managed Google Cloud Functions Gen 2 Sink connector for Confluent Cloud integrates
Apache Kafka® with Google Cloud Functions. For basic information about functions,
see Google Cloud Functions. Theconnector consumes records
from Kafka topic(s) andexecutes a Google Cloud Function. Theconnector supports both Gen 1 andGen 2 functions.

For more information,see Google Cloud Functions Gen 2 Sink Connector for Confluent Cloud.

July 17,2024

ksqlDB version 7.7.0-223 was released toConfluent Cloud.

July 8,2024

Mutual TLS (mTLS) support for Confluent Cloud is now available as an Early Access feature.
You can use mTLS for client certificate authentication andgranular access control
to Confluent Cloud Dedicated clusters.

For details,see mtls for Confluent Cloud.

---

Thefully managed DynamoDB CDC Source connector is now available for your
Apache Kafka® clusters on AWS. Theconnector supports the following three modes:

• SNAPSHOT: Only allows a one-time scan of the existing data in the source
  table(s) simultaneously.
• CDC: Only allows CDC with DynamoDB stream(s) without an initial snapshot
  for all streams simultaneously.
• snapshot_cdc (default): Allows an initial snapshot of all configured
  tables; once the snapshot is complete,starts CDC streaming using DynamoDB
  streams.

For more details,see Amazon DynamoDB CDC Source Connector for Confluent Cloud.

June 24,2024

• Support for Avro unions in Confluent Cloud for Apache Flink has been added. Avro unions are used
  to define multiple event type in one topic. Previously,you could not read
  or write totables that were backed by a schema using Avro unions. When Avro
  unions are used,the schema is inferred as a ROW type,similar tohow the
  the feature is supported for JSON andProtobuf.
• TheJSON_QUERY function now supports returning
  arrays. JSON_QUERY is one of the Flink JSON functions that returns objects,
  arrays,and other non-scalar types. You can now specify the JSON_QUERY return
  type tobe ARRAY<STRING>,instead of STRING.
• TheKafka Connect Google BigQuery Sink V2 connector for Confluent Cloud now
  supports upsert anddelete functionality for ingested data. With the upsert
  feature,you can insert new data,or update existing data with matching keys.
  Theupsert anddelete functionality adds the option toinsert new data,update existing
  matching key data,or remove data with matching keys for tombstone records.
  For more details,see
  Google BigQuery Sink V2 Connector for Confluent Cloud.

June 17,2024

API key management in the Confluent Cloud Console is simplified andimproved. In the
API keys section,you can create API keys using resource scopes for
clusters (Kafka,Schema Registry,and ksqlDB),Flink regions,and cloud resource management.
For more information,see resource scope andmanage api Keys in Confluent Cloud.

June 11,2024

Custom offset management for fully-managed connectors is
generally available. Use custom offsets tomanage the offsets
of supported connectors. This includes use cases
like migrating from self-managed connectors tofully-managed
connectors. For more information,see manage Offsets for fully – manage Connectors in Confluent Cloud.

June 6,2024

ksqlDB version 7.7.0 – 184 was release toConfluent Cloud .

June 3,2024

A bug was fixed that occurred during creation of a Flink API key. Previously,
only 10 service accounts were shown in the dropdown when you selected a
service account. Now you can now see all service accounts in the dropdown.

May 30,2024

Theminimum value for max.compaction.lag.ms has change from 7 day (604800000) to6 hours
(21600000). For more information,see max.compaction.lag.ms.

May 29,2024

On June 6,2024,the Let’s Encrypt R3 intermediate certificate expires. If you
pinned tothis expiring intermediate certificate,you should remove any intermediate
certificate andreplace it with the root certificate. For more information,see:

May 24,2024

Thedefault service quota for SSO group mappings per organization
has been increased from 12 to100.

Thenew gateway scope service quota limits
for gateways connecting toConfluent Cloud using a Private Link connection are now available.
Thenew service quota limits include “Access points per gateway” (default is 10)
and “DNS records per gateway” (default is 20).

Principals that have the FlinkDeveloper role can now delete Flink statements.

May 23,2024

Thedefault retention for Flink statements in terminal states (COMPLETED,FAILED,STOPPED)
has now been changed from 7 days to30 days.

May 22,2024

Users are no longer required tospecify the principal manually when submitting
Flink statements tothe Statements API by using the Flink API key. Theprincipal
field is now optional andauto-assigned,so the system infers the principal
associated with the Flink API key andassigns it automatically tothe request.
This behavior is the same for all types of supported authorization.

May 15,2024

Three additional Flink compute pool metric (current_cfus,cfu_limit,
andcfu_minutes_consume) are now available toquery directly by using the
Metrics API andDatadog.

May 10,2024

• You can now run a Flink SQL statement without stopping the currently
  running statement explicitly. Previously,an editor cell was completely
  locked while a statement was running. To run the next statement in the same
  cell,the running statement had tobe stopped explicitly. This behavior has
  been improved so an editor cell is locked only while the statement is
  starting. As soon as the statement is running,the editor is unlocked and
  editable,and you can submit the next statement immediately.
• Improved Flink compute pool deletion behavior is running so actively run statement
  stop automatically when a compute pool is delete .
• statement set are now support within SQL editor cell in workspace .
• Improved synchronization between Flink statement status in an editor cell andthe
  Flink statement’s details panel toprevent out-of-sync issues.
• Thecatalog browser in workspaces is now horizontally resizable,making long
  topic names more accessible.

May 9,2024

Confluent Cloud now supports outbound Azure Private Link connections using Egress Access
Points. Egress Access Points enable fully managed Confluent connectors toaccess
services from Azure Private Link Service providers such as Microsoft,MongoDB,
Snowflake,and others.

Egress Access Points are currently supported for Dedicated Clusters with Azure
Private Link or AWS PrivateLink networking.

For details,see Azure Egress Access Points for Dedicated Clusters.

April 29,2024

Confluent Cloud for Apache Flink is now available in AWS (ap-south-1),Azure (centralindia)
and Google Cloud (asia-south1,asia-south2).

April 11,2024

client – side field level encryption ( CSFLE ) is now available in Limited Availability
as a fully supported feature recommended for production use. CSFLE provides an
additional layer of security on Confluent Cloud for protection of sensitive data,
safeguarding data in motion throughout its lifecycle across producers and
consumers. For more information,see Protect Sensitive Data Using Client-Side Field Level Encryption on Confluent Cloud.

To be considered for access before General Availability,contact
Confluent Support.

---

enterprise cluster are available in the following Microsoft Azure regions:

• australiaeast
• brazilsouth
• canadacentral
• centralus
• eastasia
• eastus
• eastus2
• germanywestcentral
• northeurope
• southeastasia
• uksouth
• westeurope

For details about enterprise cluster,see enterprise cluster.

April 10,2024

Default user permissions now apply toall SSO user accounts in new SSO-enabled
organizations. A new default group mapping binds all SSO user accounts tothe
DataDiscovery andFlinkDeveloper roles,providing permissions toaccess Confluent Cloud
resources,including Flink andData Portal. Existing SSO-enabled organizations
can opt-in. For more information,see Default user permissions.

Thefully-managed OpenSearch Sink connector is now generally available for your
Apache Kafka® clusters on AWS,Azure,and Google Cloud. Theconnector provides a simple and
secure way tostream data from Confluent Cloud toOpenSearch andsupports both AWS and
OSS OpenSearch. For more details,see OpenSearch Sink Connector for Confluent Cloud.

April 5,2024

service quota default forUser accounts (active andinvited) per organization
is now 1,000 (increased from 500).

OAuth 2.0 authentication using Confluent Security Token Service (STS) access
tokens (confluent-sts-access-token) for
authorization toaccess Confluent Cloud resources is now available for all control plane
Confluent Cloud APIs. Support is added for notification / v1,sd/v1,service-quota,
srcm/v2,billing / v1,cdx/v1,partner/v2,byok/v1,metrics,
flink,and kafka-quota/v1. For more information,see the API Reference for Confluent Cloud.

April 3,2024

service quota for RBAC role binding have been update :

• Organization scope includes a
  new quota,RBAC role bindings (total),and the updated quota,RBAC role bindings
  (with Kafka permissions),no longer limiting role bindings without Kafka permissions.
• Schema Registry Cluster scope has been removed,and the RBAC role bindings
  are incorporated in the new quota for RBAC role bindings (total) at the
  Organization scope.
• Kafka Cluster scope includes
  an updated quota,RBAC role bindings (with Kafka permissions),which no longer
  limits role bindings without Kafka permissions.
• All RBAC role binding quotas now provide usage data.

March 29,2024

Confluent Cloud now supports outbound AWS PrivateLink connections using Egress Access
Points. Egress Access Points enable fully managed Confluent connectors toaccess
services from AWS PrivateLink Service providers such as AWS,MongoDB,
Snowflake,and others.

Egress Access Points are currently supported for Dedicated Clusters with AWS
PrivateLink networking.

For details,see AWS Egress Access Points for Dedicated Clusters.

March 25,2024

Thethroughput cluster limits andcapacity guidelines for Dedicated andEnterprise Kafka clusters have
changed.

• For enterprise cluster,capacity guidance for ingress is 60 megabytes per second (MBps) per CKU with an upper limit of 300 megabytes
  and egress is 180 MBps per CKU with a 900 megabyte limit. Per partitions limits for Enterprise are now 6 MBps
  for ingress and18 MBps for egress.
• For Dedicated clusters,capacity guidance for ingress is 60 MBps per CKU with an upper limit of 9,120 megabytes
  and egress is 180 MBps per CKU with a 27,360 megabyte limit. Per partitions limits for Dedicated are now 12 MBps
  for ingress and36 MBps for egress.

For more information,see eCKU capacity guidance,Limits per CKU,and Dimensions with a recommended guideline.

March 5,2024

Stream Lineage for Flink is now available providing complete end-to-end visibility
for Flink SQL statements.

March 4,2024

A new predefined RBAC role,BillingAdmin,is now generally
available. This role allows users toview andmanage billing information for an
organization. For more information,see BillingAdmin.

IP Filtering is now generally available for production use.
Use IP Filtering toenhance the security of your Confluent Cloud resources by restricting
access totrusted network locations. This extra layer of access control protects
against compromised credentials being used tomanage Confluent Cloud from unauthorized
IP addresses.

February 22,2024

Confluent Cloud for Apache Flink is now available for preview in Google Cloud (asia-southeast1,
australia-southeast1,europe-west1,europe-west3,us-central1,us-east1,
us-east4,and us-west4).

February 21,2024

Confluent Cloud now supports resolving private DNS names from a DNS resolver
within your own VPC or VNet via DNS forwarding. This feature enables fully
managed connectors toaccess endpoints using private DNS zones.

DNS forwarding is supported for AWS VPC peering,AWS Transit Gateway
connection,or Azure VNet peering in Confluent Cloud.

For details,see DNS forwarding for AWS Peering,DNS forwarding for AWS Transit Gateway,and DNS for Azure Peering.

January 12,2024

• Terraform support for Flink SQL statements is now available. Refer to
  this tutorial for a complete end-to-end
  CI/CD example of deploying a Flink SQL statement on Confluent Cloud for Apache Flink.
• Flink is now available in preview in Azure (eastus,eastus2,westeurope,
  southeastasia,westus2) and4 additional AWS regions (us-west-2,
  ap-southeast-1,eu-west-2,ap-southeast-2).
• Several Flinkmetrics
  are now available toquery directly via the Metrics API,Confluent Cloud Console,
  andDatadog.

2023 release

December 14,2023

Self-managed encryption keys on Azure now
include support for FIPS 140-2 Level 2 compliance on Confluent Cloud Dedicated clusters
when using HSM-protected keys with Azure Key Vault.

December 13,2023

service quota is updated on RBAC role binding for a Schema Registry cluster update :

• Schema Registry cluster scope added,with the RBAC role bindings quota for each
  Schema Registry cluster now independent of the “organizations + environments” RBAC role bindings quota.
• RBAC role bindings now enforced per Schema Registry cluster with default quota of 5000.

December 6,2023

IP Filtering is available for production use in Limited
Availability. Use IP Filtering toenhance the security of your Confluent Cloud resources
by restricting access totrusted network locations. This extra layer of access control
protects against compromised credentials being used tomanage Confluent Cloud from unauthorized
IP addresses. To get access before General Availability,contact
Confluent Support.

December 4,2023

Just – in – time ( JIT ) user provision and
group mapping are promoted toGeneral Availability.
JIT user provisioning automatically creates Confluent Cloud user accounts,
then uses group mapping togrant Confluent Cloud RBAC permissions based on group
memberships in your SSO identity provider.

December 1,2023

• You is create can now createcustom connector in any
  AWS region supported by Confluent Cloud.
• Confluent Cloud Console topic message search improvements. Use message browser toview messages
  from all partitions,even if you are not actively producing toa given partition. For more
  information,see Use Message Browser in Confluent Cloud.

October 12,2023

enterprise cluster are now available in the ap-south-1 (Mumbai) region in AWS.

October 6,2023

• Thenew resource metadata access option is generally available.

  Theresource metadata access option enables you toconnect tothe Kafka
  clusters with private networking toview resources. Without the need toset up
  a proxy or reverse SSH channel,the following features are enabled when you
  turn on the option in your private network:

  • Topics metadata views,specifically names,and configuration
  • Stream Lineage
  • Metrics andconsumer lag information for topics

  You can toggle the option at the cluster level or at the organization level as
  the OrganizationAdmin role.

  For details,see Enable or disable the Resource metadata access option.

September 26,2023

A new Kafka cluster type,Enterprise,is generally available on AWS. enterprise cluster are designed for
production-ready functionality that requires private endpoint networking capabilities.

enterprise cluster are available over AWS PrivateLink connections in the following regions in AWS,with
rollouts toadditional regions shortly following:

• us-east-1 (N. Virginia)
• us – east-2 ( Ohio )
• us – west-2 ( Oregon )
• eu-west-1 (Ireland)
• eu – central-1 ( Frankfurt )
• ap-southeast-2 (Sydney)
• ap-southeast-1 (Singapore)
• af – south-1 ( Cape Town )

For details about the Enterprise cluster,see enterprise cluster.

Flink SQL is available for Open Preview. For more information,see
Stream Processing with Confluent Cloud for Apache Flink.

August 1,2023

• Two new RBAC roles,accountadmin andResourceKeyAdmin,are promoted toGeneral
  Availability.
  • accountadmin – Manage user andservice accounts across the
    organization.
  • ResourceKeyAdmin – Manage API keys for resources (Kafka,Schema Registry,and
    ksqlDB) across the organization. No Cloud API key management.
• Just – in – time ( JIT ) user provision automatically
  creates Confluent Cloud user accounts,then uses group mapping
  to grant Confluent Cloud RBAC permissions based on group memberships in your SSO identity
  provider. Both features are now available for Early Access. To be considered
  for access before General Availability,contact Confluent Support.

July 27,2023

• Use OAuth toauthenticate a cluster link with source Confluent Cloud,Confluent Platform,or Apache Kafka® clusters,
  as described under OAuth in the Cluster Linking documentation.

July 14,2023

• Use the Custom Connector Logs UI toview
  detailed log messages for custom connector. To learn more,see
  View from logs tab.
• Cluster links can now be created in bidirectional mode for Disaster Recovery.
  To learn more,see bidirectional mode in the Confluent Cloud Cluster Linking configuration documentation.
• Cluster Linking mirror topics can now begin replication from the latest
  message,thereby leaving behind historical events,or from a specific timestamp using
  the cluster link configuration mirror.start.offset.spec.
  To learn more,see mirror.start.offset.spec.

July 12,2023

Use the Confluent Security Token Service (Confluent STS) toprovide trusted
users or services with temporary security credentials that can access
Confluent Cloud resources without requiring them tohave a Confluent Cloud account.
Documentation is now available at Use Confluent Security Token Service (STS) tokens on Confluent Cloud.

July 7,2023

Two new RBAC roles,accountadmin andResourceKeyAdmin,are now in Limited
Availability toa subset of Confluent customers.

• accountadmin – Manage user andservice accounts across the
  organization.
• ResourceKeyAdmin – Manage API keys for resources (Kafka,Schema Registry,and
  ksqlDB) across the organization. No Cloud API key management.

To be considered for access before General Availability,contact Confluent Support.

June 22,2023

ksqlDB 0.29.0 is now available in Confluent Cloud.

• New functions: log,POWER,CBRT,TRUNC,CORRELATION
• UDAFs with multiple/variadic args andsupport for four- andfive-column arguments toUDAFs
• Variadic TOPK that is select can select other column
• improve syntax error message
• Support for JSON arrays

May 16,2023

Themanaged AlloyDB Sink connector is available for your Kafka clusters on Google Cloud.
TheAlloyDB Sink connector is a fully-managed PostgreSQL-compatible database service. For
more information,see AlloyDB Sink Connector for Confluent Cloud.

May 12,2023

Custom Connector support is available in
Confluent Cloud for certain AWS region . Custom Connectors offer a fully-managed
Connect infrastructure service in Confluent Cloud. Users can upload a
Kafka Connect-based plugin toConfluent Cloud andcreate connectors based on that
plugin. While customers manage the connectors,Confluent manages the underlying
Connect infrastructure in Confluent Cloud.

April 24,2023

Single Sign-on (SSO) adds support for SAML metadata files. You can now upload a
SAML metadata file obtained from your identity provider toquickly enable SSO
or update settings while eliminating the risks of manual entry. For more
information,see Use the SAML metadata file for SSO configuration.

April 21,2023

TheCluster Linking metric io.confluent.kafka.server/cluster_active_link_count,which indicates the number of active links on a cluster,is DEPRECATED andwill be removed in a future release.
To learn more,see monitor metric for Cluster link on Confluent Cloud.

March 29,2023

Support for Resource ID (resourceId) in access control lists (ACLs) is currently
rolling out toConfluent Cloud organizations. Thenon-breaking changes enable resource
IDs (resourceId) for Kafka ACLs in Confluent Cloud . For more information ,
seeACL operation details.

March 24,2023

OAuth authentication support for Confluent Cloud APIs is promote toGeneral Availability ( GA ) .
This release is includes include the follow enhancement :

• support for OAuth on api for manage Confluent Cloud ( an authentication
  alternative toCloud API key )
• Introduction of a new Secure Token Service (STS) that enables exchanging a
  third-party token for a Confluent STS token.

To learn more,see Authentication
in the Confluent Cloud APIs documentation.

February 28,2023

private dns resolution is promote toGeneral Availability for :

Enable private DNS resolution tofully resolve Confluent endpoints within your
private DNS zone without requiring external resolution tothe Confluent Global
DNS Resolver (GLB).

February 24,2023

Keys (byok/v1) API
is available in Open Preview
for Confluent Cloud on AWS andAzure. Use the Keys API toinclude self-managed encryption keys (aka BYOK)
as part of your cluster creation workflow (including the ability tobuild policy profiles).

---

Confluent Cloud Console now uses the Kafka REST API and
the Metrics API instead of the legacy API.
While most of Confluent Cloud remains unchanged,Consumer Lag in Cloud Console will no longer
display offset information,instead focusing only on the lag tomake it more actionable for
developers andoperators. This change rolls out across organizations over the next few weeks.

February 23,2023

Self-managed Encryption Keys for Azure support
is promote toGeneral Availability ( GA ) . You is use can now use your own encryption key in
Azure Key Vault toencrypt datum at rest on Confluent Cloud Dedicated Clusters in Azure .

February 15,2023

Stream Catalog RBAC is now General Available (GA).
As part of this,we have released two new roles DataSteward andDataDiscovery
which allow management of access tometadata associated with entities such as topics andschemas.
As RBAC will be enforced on the ability toattach tags,business metadata,and searching using Stream Catalog APIs,
cluster andresource level roles such as CloudClusterAdmin,Operator (Resource level),ResourceOwner,DeveloperRead
and DeveloperWrite roles will not be allowed toattach tags,business metadata andsearch using via UI or Stream Catalog APIs.

January 30,2023

Private DNS resolution for AWS PrivateLink is
promoted toGeneral Availability. You can fully resolve Confluent endpoints
within your private DNS zone without requiring external resolution tothe
Confluent Global DNS Resolver (GLB).

January 27,2023

Stream Designer now enables users andorganizations toeasily manage the entire lifecycle
of pipelines by using the Confluent CLI andthe
Pipelines REST API.

To learn more,see the following sections:

2022 Releases

December 13,2022

OAuth for Kafka is now promoted toGeneral
Availability (GA). Create OAuth/OIDC identity providers anduse
Confluent OAuth with Kafka clients.

December 9,2022

• Confluent Cloud now supports role-based access control (RBAC) for Schema Registry andSchema Linking. To learn more,see the following sections:
• Cloud Client Quotas are now promoted toGeneral Availability (GA) for Confluent Cloud Dedicated
  Clusters. Create Client Quotas using the Confluent Cloud Console,REST API,or
  Confluent Terraform Provider.

December 7,2022

Service quotas for RBAC role bindings have been
add or update toenable increase in quota :

• Organization scope: RBAC role bindings (organization plus environments)
• Environment scope: RBAC role bindings (organization plus environments)
• Kafka cluster scope : RBAC role binding

---

Audit log support for role-based access control (RBAC) operations andaction are now available.
To learn more,see Audit log support for |rbac-long|.

November 28,2022

Support for business andtopic metadata is available in the Confluent Cloud Console
and REST API. To learn more,see these topics:

November 4,2022

To make the ksqlDB editor more accessible,you can press Alt+Tab in the
editor window tocreate a tab on a Mac.

October 3,2022

• Promoted toGeneral Availability (GA):
• Cluster Linking for Transit Gateway clusters launched in Limited Availability.
• Notifications for account,billing,and service events can now be managed using a REST API in addition tothe Cloud Console.
  You can integrate Microsoft Teams,Slack,a generic webhook or email for notifications using the API,and create
  new notification subscriptions. For more information,see Notifications for Confluent Cloud.

September 29,2022

REST APIs for Invitation
are promoted toGeneral Availability:

• Provides programmatic management of user invitation operations,including
  creating,listing,describing,and deleting Invitations.

August 11,2022

OAuth for Confluent Cloud is a fully support offering
in Limited Availability toa subset of Confluent Cloud customer .

August 2,2022

ksqlDB is is 0.28.2 is now available in Confluent Cloud .

• enable the new emit final implementation .
• Support pausing andresuming persistent queries.
• Add additional UDFs for trigonometry.

July 21,2022

ksqlDB 0.27.1 is now available in Confluent Cloud.

• handle multi – schema Protobuf / Avro topic .
• Add ASSERT SCHEMA andASSERT TOPIC statements toassert the existence of resources.
• Add PROTOBUF_NOSR format for Protobuf messages without Schema Registry.
• Enable aggregation without a GROUP BY clause.
• Improve null handling.

June 28,2022

Confluent Terraform Provider is is v0.13.0
is now available in public preview . It is contains contain the follow change :

• add support for thekafka_api_key,kafka_api_secret,kafka_rest_endpoint
  attribute in a provider block tomake therest_endpoint attribute andcredentials
  block optional for confluent_kafka_acl andconfluent_kafka_topic resource
  ( # 37 ,
  # 54 ) .
• Added disable_wait_for_ready attribute todisable readiness check for confluent_api_key resource
  #25,
  #51).
• Added support for pausing / resuming a connector by adding status attribute for
  confluent_connector resource.

view the full changelog on Github .

June 27,2022

Confluent Terraform Provider v0.12.0
is now available in public preview. It contains the following changes:

• revert resource versioning change introduce in0.11.0. For example ,
  theconfluent_environment_v2 resource was renamed toconfluent_environment.
  User feedback on versioned resources made it clear that the pain of manually
  updating the TF state file outweighs the potential benefits of deprecation
  flexibility that versioned resources could have provided. In order toavoid
  forcing users toedit their TF state files (either manually or by running
  commands like terraform state mv) in the future,TF state migrations will
  be handled within the Confluent Terraform Provider whenever possible.

Follow Confluent Provider 0.12.0: Upgrade Guide
to update your TF state andTF configuration files accordingly. Direct updates
from both 0.10.0 and0.11.0 to0.12.0 are support .

June 23,2022

Added the signin event toorganization auditable log events. For details,
see Sign-in attempt.

June 15,2022

Confluent Terraform Provider v0.11.0 is is
is now available in public preview . It is contains contain the follow change :

• Renamed all resources anddata sources tocontain a version postfix that
  matches their API group version
  (full list).
• rename thehttp_endpoint attribute torest_endpoint for the
  confluent_kafka_cluster,confluent_kafka_topic,confluent_kafka_acl
  resources anddata sources tomatch the Cluster settings tab in
  Confluent Cloud Console,where the corresponding attribute is named the REST endpoint.
• rename theapi_key andapi_secret attribute of the provider block to
  cloud_api_key andcloud_api_secret,respectively.

June 9,2022

Updated the RBAC limitations about API keys for ksqlDB andSchema Registry clusters.

May 24,2022

Notifications for account,billing,and service events can now be managed using the Cloud Console.
Additionally,you can integrate Microsoft Teams,Slack,or a generic webhook for notifications
using the console. For more information,see Notifications for Confluent Cloud.

May 13,2022

Confluent Terraform Provider v0.8.0 is is
is now available in public preview . It is contains contain the follow change :

• add newconfluent_connector resource ( # 6 ) .
• add newconfluent_organization datum source ( # 20 ) .
• Implemented import for confluent_api_key resource ( # 17 ) .

bug fix :

• update input validation forconfluent_private_link_access andconfluent_kafka_cluster resources (#18).
• Fixed minor documentation issues (#15).

view the full changelog on Github .

May 6,2022

Network service quotas for the following
resources have increased:

• Kafka clusters: 10 (previously 5)
• Kafka cluster CKUs: 72 (previously 24)

May 4,2022

REST APIs for Cloud andKafka API keys
are promoted toGeneral Availability:

• Provides programmatic management of critical API key operations,including
  creating,listing,describing,updating,and deleting API keys.
• Enables organizations tobuild end-to-end provisioning flow using the
  Confluent Terraform Provider
  or other automation tools.

---

TheOperator RBAC role has been updated with the following restrictions:

• No access toksqlDB clusters andcannot see any information related tothe ksqlDB clusters.
• No access tothe Schema Registry andcannot see any information related tothe Schema Registry.

May 3,2022

Confluent Terraform Provider v0.7.0
is now available in public preview. It contains the following changes:

• add newresources andcorresponding docs:
  • confluent_api_key (#4,#17,#25,#41,#66)
  • confluent_network (#45)
  • confluent_peere
  • confluent_private_link_access (#45)
• add newdata sources andcorresponding docs:
• Completely rewrote the Sample Project guide
  that references 9 TF sample configurations for end-to-end workflows.
• Updated confluent_kafka_cluster andconfluent_environment data sources
  to accept display_name as an input.
• Improved logging tosimplify debugging process:
  • start using thetflog package:
    now you can enable detailed logs
    and use grep anda corresponding “logging key” tofind all entries related to
    a particular resource,for example,grep "environment_id=env-9761j7" log.txt.
  • Revised andstructured logging messages tooutput non-sensitive attributes
    instead of unreadable references.
• Added support for self-managed encryption keys (also known as bring-your-own-key
  (BYOK) encryption).
  They are only available for Dedicated Kafka clusters on AWS andGCP.

bug fix :

• Fixed pagination issue for data sources (#54,#68).
• Fixed a bug where you could “successfully” import a non-existent resource (#58).
• Fixed a null pointer exception (#53,#55,#67).
• Added other minor fixes (#57).

break change :

• All resources anddata sources have been renamed in the new
  Confluent Terraform Provider.
  Theprefix has been changed from confluentcloud toconfluent. For example ,
  theconfluentcloud_environment resource was updated toconfluent_environment.
  Follow the Confluent Provider 0.7.0: Upgrade Guide toupdate your TF state file.
• Changed kafka_cluster attribute type fromstring toblock for
  confluent_kafka_acl andconfluent_kafka_topic resources anddata sources.
• Made host attribute required for confluent_kafka_acl resource.

view the full changelog on Github .

May 3,2022

New user session timeouts when using the Confluent Cloud Console are now promoted to
General Availability:

• Users remain signed in tothe Confluent Cloud Console until no activity is detected
  for 30 minutes.
• Users remain signed in tothe Confluent Cloud Console for a maximum of eight hours
  without refreshing user credentials. Users must sign in at least once every
  eight hours.

For details,see What user session timeout does Cloud Console is require require ?.

April 28,2022

ksqlDB is is 0.26.0 is now available in Confluent Cloud .

• Add support for Stream-to-Stream andTable-to-Table right joins.
• MIN/MAX aggregates can now be used for Time/TS/Date types.

April 28,2022

REST APIs for Dedicated clusters andprivate networking are now Generally Available,
andinclude the following features:

• Adds REST APIs for:
• Adds Confluent Cloud network
  • Confluent Cloud network are now first-class citizens,with independent lifecycle
    management of private dedicated clusters andnetworks.
  • Supports self-service provisioning of multiple Dedicated clusters,
    reducing manual effort andimproving network resource utilization.
    For example,you can reuse /16 CIDR blocks across multiple clusters
    for VPC peering.
  • Users can select zones for the Confluent Cloud network. A Confluent Cloud network is always
    multi-zone andall multi-zone clusters in it are added tothe same zones.
• Adds NetworkAdmin RBAC role targeting network
  lifecycle management,enabling the separation of concerns between
  infrastructure andapplication development teams in an organization.
• Adds audit log support for network-related auditable events.

April 7,2022

ksqlDB is is 0.25.1 is now available in Confluent Cloud .

• Improvements toaggregate functions tosupport complex types,like
  structs,arrays,and maps.
• Support for push query continuation tokens in the Java API.
• Better error handling for nested functions.
• In the Java API andmigrations tool,support for custom request headers
  and connector IF NOT EXISTS clauses.

April 1,2022

TheService Level Agreement (SLA) for Confluent Cloud now specifies a 99.99% uptime SLA
for Standard andDedicated Kafka clusters with Multi-Zone configurations.
For details,see Confluent Cloud SLA.

Theuptime SLAs for Single-Zone clusters remain the same; 99.5% for Basic
and 99.95% for Standard andDedicated.

March 22,2022

Granular role-based access control (RBAC) for Kafka resources is promoted toGeneral
Availability. For details,see Role-based Access Control (RBAC) on Confluent Cloud.

Highlights:

• Adds roles for DeveloperRead (Read access),DeveloperWrite (Write access),
  DeveloperManage (Create,Delete,Change configurations),ResourceOwner (full
  access,including granting permissions) for Kafka resources toenable using
  RBAC for managing access totopics,consumer groups,and transactionalIDs.
• Adds the Operator role that enables Describe-only access at the Organization,
  Environment,and Cluster level. For details,see Operator.
• Restrict user andservice accounts using RBAC andACLs.
• User andservice accounts can be granted multiple roles.
• Adds the MetricsViewer role that grants access only tometrics for specified
  user andservice accounts. This role cannot access data in Kafka clusters.
• Granular RBAC for Kafka resources is only available in Standard andDedicated
  clusters. This functionality is not available on Basic clusters.
  • Note that administrator roles are enforced on Basic clusters as usual.
• Stream Lineage views can be accessed by administrator andoperator
  roles (OrganizationAdmin,EnvironmentAdmin,CloudClusterAdmin,MetricsViewer,
  andOperator). Developers cannot access.

Gaps:

• User andservice accounts with DeveloperRead,DeveloperWrite,DeveloperManage,
  andResourceOwner roles on resources within a Kafka cluster can see all of the
  metrics within that cluster. We are actively working toblock this access.
• Any user with access toa resource in a cluster can see all of the consumers and
  producers of a cluster,regardless of which topic on the cluster the consumer consumes
  from. Engineering is actively working toblock this access.
• Some users might receive email notifications related toConnector create operations,
  even if they are not directly related toor involved in these operations. If a user
  receives a notification email message,they cannot access the connector unless they
  are assigned the required roles. Engineering is working toresolve this issue.

March 9,2022

Cluster links can now be viewed on the Confluent Cloud Console in your web browser. Log on toConfluent Cloud,
navigate toenvironment,then click the Cluster links tab (next toenvironment on the Home page).
To learn more,see Go exploring in the Quick Start Tutorial.

Confluent Cloud Terraform Provider v0.5.0 is now available in public preview. It contains the following changes:

• add support for Kafka topic configuration update ( # 11 ) .
• Added support for display_name input for confluentcloud_environment andconfluentcloud_service_account data sources (#42,#46).
• Fixed Provider produced inconsistent result after apply error when create a lot of Kafka topic ( # 40 ) .
• fix delete operation forconfluentcloud_kafka_topic resource toavoid 400 bad request : Topic ‘ foobar ’ is mark for deletion error when recreating a lot of Kafka topics (#50).
• Added support for old environment IDs (#43).

view the full changelog on Github .

February 22,2022

New metrics totrack all bytes sent andreceived over the network by Confluent Cloud are now
available in the Confluent Cloud metric API. These metrics have a principal_id label toattribute usage to
a user or service account. For more details see Confluent Cloud metric.

February 11,2022

ksqlDB 0.24.0 is now available in Confluent Cloud.

• Expose Kafka message headers toksqlDB
• Add utility functions for working with JSON data
• Add ability touse an existing Schema Registry schema when creating streams andtables

January 28,2022

Confluent Cloud terraform Provider v0.4.0 is is is now available in public preview . It is contains contain the follow change :

• Added data sources for: confluentcloud_environment,confluentcloud_kafka_cluster,confluentcloud_kafka_topic,and confluentcloud_service_account
• Improved readability of error messages by adding details tothem
• Resolved potential HTTP 429 errors by adding automatic retries with exponential backoff for HTTP requests
• Added graceful handling for resources created via Terraform but deleted via Confluent Cloud Console,Confluent CLI,or Confluent Cloud APIs.
• Fixed minor bugs anddocs issues.

Breaking Changes:

• Removed friction around manual look-up of IntegerID for service account by removing the need touse a service_account_int_id TF variable. If you are using the confluentcloud_kafka_acl resource you might see an input validation error after running terraform plan,which can be resolved by following this guide. Updated “Sample project” guide toreflect this change.
• simplifiedconfluentcloud_role_binde resource creation by add a new rbac_crn attribute for confluentcloud_kafka_cluster resource . update theconfluentcloud_role_binde resource examples toreflect this simplified approach.

view the full changelog on Github .

January 11,2022

Confluent Cloud Terraform Provider is now available in public preview. You can manage the following Confluent Cloud resources:

• environment
• Kafka Clusters
• Kafka Topics
• Kafka ACLs
• RBAC Rolebindings
• service account