8.8 Release notes

These release notes identify the main new features included in the 8.8 minor release, including alpha feature releases.

Minor release date	Scheduled end of maintenance	Changelog(s)	Release blog	Upgrade guides
14 October 2025	13 April 2027	Patch Releases and Changelogs	-	Upgrade guides

8.8 resources

• See release announcements to learn more about supported environment changes and breaking changes or deprecations.
• See What's new in Camunda 8.8 for important changes to consider when planning your upgrade from Camunda 8.7.
• Refer to the quality board for an overview of known bugs by component and severity.

Agentic orchestration

Agentic orchestrationAI agentsSaaSSelf-Managed

Camunda agentic orchestration allows you to build and orchestrate AI agents within your BPMN-based workflows, enabling human tasks, deterministic rule sets, and AI-driven decisions to collaborate in a robust, end-to-end process.

Camunda Agentic orchestration

Use the following new features to build and integrate AI agents into your processes:

Feature	Description
AI agent connector	Enables AI agents to integrate with an LLM to provide interaction/reasoning capabilities. This connector is designed for use with an ad-hoc sub-process in a feedback loop, providing automated user interaction and tool selection.
MCP Client connector	Connect an AI agent connector to tools exposed by Model Context Protocol (MCP) servers.
Ad-hoc tools schema resolver connector	Can be used independently with other AI connectors for direct LLM interaction. Use this connector if you don’t want to use the AI agent connector but still want to resolve tools for an ad-hoc sub-process or debug tool definitions.
Vector database connector	Allows embedding, storing, and retrieving LLM embeddings. Use this connector to build AI-based solutions such as context document search, long-term memory for LLMs, and agentic AI interaction.
fromAi() FEEl function	Use in combination with the AI Agent connector. See AI Agent tool definitions.

APIs & tools

APIs & toolsSaaSSelf-Managed

APIs

Use the Orchestration Cluster REST API to interact programmatically with the Orchestration Cluster.

• This replaces component APIs (Operate API, Tasklist API, Zeebe API, and much of Identity API) with a single set of endpoints.
• This unified API supports both organizational (SaaS) and Self-Managed deployments.
• This is now the default and recommended integration point for developers, operators, and automation solutions.
• You can also use a Swagger UI to interact with this API.

Orchestration Cluster API

Camunda Java client

The new Camunda Java Client replaces the Zeebe Java Client as the official Java library for building process applications that integrate with Camunda 8. This client provides everything you need to programmatically interact with the Orchestration Cluster, whether you are orchestrating microservices, managing human tasks, or visualizing process data.

Get started with the Camunda Java client

Camunda Process Test H2 data layer support

Camunda Process Test supports using the H2 Database Engine as the default embedded data layer.

• H2 is now automatically provisioned when integrating the Camunda Process Test libraries, eliminating manual database configuration and reducing memory footprint.
• H2 support streamlines the developer experience for your Spring Boot and plain Java projects. Process testing is now faster to set up, simpler to maintain, and easier to integrate with your continuous integration workflows.

Camunda Process Test

Camunda Spring Boot Starter

The Camunda Spring Boot Starter replaces the Spring Zeebe SDK to simplify interaction between a Spring Boot application and Camunda 8, allowing you to:

• Easily integrate process entity management and queries within your workflows.
• Seamlessly configure endpoints and authentication via Spring Boot auto-configuration, minimizing boilerplate code.
• Rely on an official, standardized approach to guarantee consistency and reduce maintenance costs when upgrading.
• Based on Spring Boot 3.5 (version compatibility matrix).

Get started with the Camunda Spring Boot Starter

Process instance tags

Introduce optional, immutable process instance tags set at creation for lightweight routing, correlation, and future prioritization without inspecting large variable payloads. See process instance creation and job workers.

Process instance creation

Public API

Find out what’s included in Camunda 8's public API, the policies around versioning, and what to expect when upgrading.

• The public API is the official contract between Camunda and its users under SemVer.
• No breaking changes will be made to the public API in minor or patch releases.
• You can safely build on these interfaces with the expectation of stability and backward compatibility.

Public API

Run process segment

Manually execute and test individual tasks or segments (connectors, RPA bots, IDP extractions) without running full processes, improving debugging and development efficiency.

Run process segment

Connectors

ConnectorsSaaSSelf-Managed

Ad-hoc tools schema resolver connector

This connector implements the tool resolution part of the AI Agent connector, but can also be used independently with other AI connectors for direct LLM interaction. Use this connector if you don’t want to use the AI agent connector but still want to resolve tools for an ad-hoc sub-process or debug tool definitions.

Ad-hoc tools schema resolver connector

AI Agent connector

Enable AI agents to integrate with an LLM to provide interaction/reasoning capabilities. Designed for use with an ad-hoc sub-process in a feedback loop, providing automated user interaction and tool selection.

AI Agent connector

Azure Blob Storage connector

Store and retrieve documents within Camunda workflows using Azure Blob Storage. Enables seamless document management directly within processes, improves efficiency, and reduces the need for custom integrations.

Azure Blob Storage connector

CSV connector (SaaS)

Read, filter, transform, and write CSV data within processes. Reduces technical debt, accelerates development, and broadens integration capabilities with native support for this universal data format.

CSV connector

Google Cloud storage connector

Store and retrieve documents directly within Camunda workflows, streamlining document management without custom development.

Google Cloud storage connector

Hubspot Connector

Connect your BPMN service with HubSpot and manage your HubsSpot contacts, companies, and deals.

HubSpot connector

MCP Client connector

Enable Camunda processes and AI agents to auto-discover and invoke external tools, eliminating hardwired connectors and enabling dynamic, metadata-driven tool integration.

MCP client connector

Vector database connector

Enable embedding, storing, and retrieving Large Language Model (LLM) embeddings. Use to build AI-based solutions for your organizations, such as context document search, long-term LLM memory, and agentic AI interaction in combination with the AI Agent connector (RAG).

Vector database connector

Email connector Message-ID

The Email connector now exposes the Message-ID provided by the client in the connector response payload. This allows for improved traceability, easier correlation between sent messages and logs, and better integration with downstream systems that rely on Message-ID.

note

This change is backwards-compatible and does not require any action. You can now optionally use the messageId field for enhanced tracking when parsing connector responses.

Intrinsic functions

A new getJson intrinsic function accepts a document and an optional FEEL expression. It extracts and returns content from a JSON document as an object.

• The optional FEEL expression parameter specifies the part that will be extracted from the JSON document content.
• If not provided, the whole document is returned as a JSON object.

Intrinsic functions

Job header binding

Use the @Header annotation to bind a header value (for example, a FEEL expression) to a model class.

SQL connector Oracle database connection

The SQL connector now supports Oracle Database connections.

SQL connector

Console

ConsoleSaaSSelf-Managed

Connector manage and run supports multiple runtimes

Connector manage and run in Console now supports management of multiple connector runtime instances.

Manage your connectors

Docker distribution

Console is now available as a Self-Managed Docker distribution. Deploy Camunda Console using a Docker image, run Console outside Kubernetes using Docker Compose or other container services, and use Console with Camunda 8 clusters deployed via Docker.

Docker Compose developer quickstart

Encryption (BYOK)

The new Bring Your Own Key (BYOK) approach with Amazon Key Management Service (KMS) allows you to configure Camunda 8 SaaS to use a customer-managed key stored in your own Amazon account. This ensures you retain ownership, control, and visibility throughout the key lifecycle, including creation, rotation, and audit logging.

• Camunda 8 SaaS encrypts all cluster data at rest to protect customer workloads. By default, encryption is managed using cloud provider–supplied keys. While this satisfies many organizations, some (especially in regulated industries) require full control over their encryption keys.
• This feature is currently available only for customers provisioning their Camunda 8 SaaS clusters in AWS regions. Support for additional cloud providers may be added in the future.

Regions

New regions are available for SaaS clusters on Amazon Web Services:

Type	Region
AWS region	Singapore (ap-southeast-1)
Secondary backups region	Ireland, Europe (eu-west-1)
Secondary backups region	Jakarta, Indonesia (ap-southeast-3)
Secondary backups region	Oregon, North America (us-west-2)

Use these regions to:

• Improve overall processing speed and reduce latency if you operate in countries within that region.
• Keep cluster data within that country/region to support your local data residency and compliance needs.

Regions

Usage metrics for licence model and tenant

Usage metrics now support per-tenant reporting and align with Camunda’s updated licensing model based on the number of tenants.

Desktop Modeler

Desktop ModelerSaaSSelf-Managed

Settings

The new Settings window in Desktop Modeler allows you to configure the application and customize your modeling experience. You can select your default execution platform version, as well as other options that were previously only available as flags.

Desktop Modeler settings

Helm chart deployment

Helm chartsSelf-Managed

Alternative container images

Camunda provides alternative container images to the previous Bitnami images, offering better security and faster patch delivery. From 8.8, these are the default supported images.

• These images are hosted on registry.camunda.cloud.
• To use them, update your Helm deployment to reference the values-images-ee.yml file.

Install Camunda with Helm

Configurable volumes

The Helm chart now supports configurable volumes. You can define PersistentVolumeClaims or continue using EmptyDir through values.yaml.

ExtraVolumeClaimTemplates

You can now add custom extraVolumeClaimTemplates to the Zeebe/Core StatefulSet by supplying an array of templates in your Helm values file. This allows you to attach additional persistent volumes to each Zeebe/Core pod for use cases such as custom storage or log directories.

important

Kubernetes does not allow you to change the volumeClaimTemplates of an existing StatefulSet. If you add, remove, or modify extraVolumeClaimTemplates after initial deployment, you must delete and recreate the StatefulSet (which will also delete the pods) for the changes to take effect. This may require additional planning and data migration steps to avoid data loss.

Common labels for Camunda resources

A new commonLabels value is now available and integrates with camundaPlatform.labels. This allows you to define mutable labels that are automatically applied to all Camunda resources. By setting commonLabels, you can ensure consistent labeling across deployments, making it easier to manage, organize, and identify resources within your Camunda environment.

Configure Web Modeler replicas

The number of replicas for the Web Modeler REST API and web app deployments can be set with new configuration properties: webModeler.restapi.replicas and webModeler.webapp.replicas, respectively.

Integrations

Microsoft Teams

The new Camunda for Microsoft Teams app allows you to view, claim, and complete Camunda tasks directly in Microsoft Teams.

The app integration allows you to:

• Start processes from a channel, chat, or the app Home tab.
• Fill out start forms in Microsoft Teams and submit them to kick off workflows, with optional links to Operate for monitoring.

Camunda for Microsoft Teams app

note

The Microsoft Teams integration is released as an early access alpha feature to allow you to test and participate in development by sharing feedback before general availability, and is subject to alpha feature limitations.

SAP Advanced Event Mesh (AEM) connectivity

Connect Camunda directly to SAP Advanced Event Mesh (AEM) for event-driven process automation.

• New AEM connectors allow you to receive CloudEvents from SAP systems as BPMN messages and publish CloudEvents from Camunda to AEM.
• This enables seamless integration with SAP’s event-driven architecture, allowing your Camunda processes to react in real time to business events across SAP BTP, S/4HANA, or ECC.

SAP Eventing with SAP Advanced Event Mesh

ServiceNow

Extend the power of your process automation by integrating Camunda with ServiceNow. This integration enables seamless communication between your BPMN workflows and ServiceNow IT Service Management (ITSM), helping you automate routine tasks and accelerate service delivery.

The ServiceNow integration allows you to:

• Manage ServiceNow data: Create, read, update, and delete records in any ServiceNow table directly from Camunda workflows.
• Trigger ServiceNow flows: Start automations built in ServiceNow's Flow Designer as part of an end-to-end process.
• Orchestrate ITSM processes: Integrate Camunda tasks with ServiceNow approvals, incidents, and service requests to create unified workflows.

ServiceNow integration

Intelligent document processing (IDP)

IDPSaaS

Bring your own model (BYOM)

When using an OpenAI Compatible provider, you can use your own model for document extraction. Any model provider that implements the /chat/completions API endpoint is supported.

Supported providers are:

• AWS: Amazon Web Services with Bedrock and Textract (supports both structured and unstructured extraction).
• Azure: Microsoft Azure with AI Document Intelligence and AI Foundry (unstructured extraction only).
• GCP: Google Cloud Platform with Vertex AI and Document AI (supports both structured and unstructured extraction).
• OpenAI compatible: Supports OpenAI and any provider implementing the OpenAI /chat/completions API (unstructured extraction only).

Document extraction

Structured data form extraction

You can use form-based structured document extraction to capture data from structured documents.

• For example, you can use this extraction method for documents with a consistent layout, such as invoices, tax forms (for example, W-2s, VAT declarations), and loan or insurance applications.
• Projects can be shared organization-wide, enhancing accessibility to extraction capabilities.

Extract structured data

Migration from Camunda 7 to Camunda 8

Camunda 7 migrationSaaSSelf-Managed

Data migration tool

Use the Data Migrator to copy running process instances from Camunda 7 to Camunda 8.

• Copy running process instances (state-preserving).
• Copy process variables and their values.
• Handle problematic instances gracefully with retry options.
• Write custom code to intercept variable migration.

Data Migrator

Migration Analyzer & Diagram Converter

The Migration Analyzer & Diagram Converter helps you get a first understanding of migration tasks when moving from Camunda 7 to Camunda 8. It analyzes Camunda 7 model files (BPMN or DMN) and generates a list of tasks required for the migration.

It can also automatically convert these files from Camunda 7 format to Camunda 8 format (updating namespaces, XML structures/properties, and simple expression transforms), with a web UI and CLI that outputs XLSX/CSV reports, for prioritization and batch conversion.

Migration Analyzer & Diagram Converter

Code conversion

Code conversion utilities provide code mapping tables, conversion patterns, and automatable refactoring recipes to systematically translate Camunda 7 implementation patterns to Camunda 8 equivalents.

Code conversion

Optimize

OptimizeSelf-Managed

Amazon OpenSearch support

Camunda 8 Self-Managed now fully supports the use of Amazon OpenSearch with Optimize.

Orchestration Cluster

Orchestration ClusterZeebeOperateTasklistIdentitySaaSSelf-Managed

Architecture

The primary architectural change in 8.8 is the consolidation of the core Zeebe, Operate, Tasklist, and Identity components into the Orchestration Cluster application as a single deployable artifact, distributed as a JAR file or Docker container.

This impacts how Camunda 8 is deployed, managed, and scaled.

• The Orchestration Cluster (previously automation cluster) is now the core component of Camunda 8.
• Use the Orchestration Cluster REST API to interact programmatically with the Orchestration Cluster.
• The new unified exporter architecture improves cluster management and data migration, bringing importer and archiving logic of web components (Tasklist and Operate) closer to the distributed platform (Zeebe). The index schema is also harmonized.
• The unified configuration for Orchestration Cluster components allows you to define cluster and component behavior.

What's new in Camunda 8.8

Dynamic activation of ad-hoc sub-processes using job workers

Ad-hoc sub-process elements can be activated by job workers:

• Define a task in the process model.
• When the engine reaches the ad-hoc sub-process, a job is created. Completing this job with a job result lets you define which elements in the ad-hoc sub-process to activate.
• Once any flows in the ad-hoc sub-process complete, a new job is created, giving job workers control over what to do next.

Dynamic partition scaling

You can now add new Zeebe partitions to a running cluster to increase capacity without downtime:

• New process instances will be started on new partitions immediately.
• Existing process instances do not migrate between partitions, so it can take time for the cluster to reach equilibrium.
• Existing messages remain on original partitions, potentially causing slight imbalances for message-heavy workloads (future updates will address this)

Cluster scaling

Identity and Management Identity

In Camunda 8.8, Orchestration Cluster Identity and Management Identity are two separate components used for Identity management, each with distinct areas of responsibility.

• Identity: Access and permission management for all Orchestration Cluster components: Zeebe, Operate, Tasklist, and the Orchestration Cluster REST and gRPC API.
• Management Identity: Continues to manage access for platform components such as Web Modeler, Console, and Optimize.

What's new in Camunda 8.8

Migrate taken sequence flows flowing to the joining gateway

Enhanced migration now supports taken sequence flows leading to joining gateways:

• Define migration plans mapping active elements and taken sequence flows
• Configure plans via Operate UI or API

Process instance migration

Enhanced process instance migration allows you to solve problems with process definitions and use the latest process improvements. You can now migrate compensation boundary event subscriptions, escalation boundary events, and escalation event subprocesses.

Process instance migration

React to expired messages with a custom exporter

You can now use a custom exporter to react to expired messages.

• The original message is retrieved from primary storage upon expiration, and the expiry event enhanced with the original message.
• Existing functionality remains unchanged, so there is no disruption for current customers or custom exporters.

With these updates, developers can subscribe to the expired messages through a custom exporter, examine the event content, and optionally re-publish or handle the message differently.

Run Orchestration Cluster without secondary storage

You can now run Orchestration Cluster in "Zeebe-only" mode without secondary storage:

• Ideal for setups not using Query APIs, Operate, or Tasklist
• Enable by setting camunda.database.type=none
• Starts only the required Zeebe components

User task listeners

Task lifecycle management is enhanced with user task listeners, allowing users to react to specific user task lifecycle events.

• Process designers can now model task listeners for different events, such as assigning and completing.
• Developers can use the same job infrastructure to activate and complete task listener jobs.
• Operations engineers can easily check details of active and completed task listeners within instances, and efficiently resolve task listener incidents.

This enhancement streamlines operations and ensures smoother incident handling, improving time to unblock process execution.

User task listeners

Zeebe-managed resilient batch operations

All batch operations, such as canceling or resolving incidents in bulk, are now handled by Zeebe instead of Operate.

• This change ensures region failovers in the multi-region setup no longer risk losing critical batch commands.
• Users will initiate and manage batch operations through the Orchestration Cluster REST API and the Operate UI, but the underlying processing occurs within Zeebe.
• By moving batch operations to the core engine, multi-region deployments gain reliability and resilience.

Robotic Process Automation (RPA)

RPASelf-Managed

Multi-file script support

RPA supports multi-file script support to allow you to organize scripts modularly, reuse common automation components, and integrate existing Robot Framework scripts. The execution engine fully supports multi-file scripts and linked resources, improving scalability, maintainability, and flexibility for enterprise automation projects.

RPA worker offline installer

An offline installer package for the RPA worker allows installation without internet connectivity. The offline installer removes reliance on external repositories or downloads, ensuring consistent, secure, and hassle-free deployment into air-gapped or restricted environments.

Web Modeler

Web ModelerSaaSSelf-Managed

Bring your own model (BYOM)

BPMN Copilot and FEEL Copilot in the Web Modeler are now available for Camunda Self-Managed customers in a safe, secure way.

You can connect these features to your own AI provider and model, even fine-tuning the model yourself to improve its performance and relevance to your organization.

CI/CD building blocks guide

Element templates documentation is improved to support scalable reuse of building blocks, focusing on CI/CD practices for managing templates across teams and environments.

Key updates include:

• Guidance on integrating element templates and dependencies into CI/CD pipelines
• Restructuring the documentation to improve discoverability and usability
• Tool-agnostic content covering Web Modeler, dependency management, and local development setups
• Moving relevant information from Connectors to the element templates section

Element templates in Modeler

Cluster basic authentication

In addition to bearer token authentication, you can now configure Web Modeler in Self-Managed to use basic authentication for cluster access.

• Set the CAMUNDA_MODELER_CLUSTERS_0_AUTHENTICATION environment variable value to BASIC.
• Web Modeler sends a username and password with every request to the cluster.

Available authentication methods

Element template editor onboarding

Web Modeler now provides a low-coder friendly UX for creating building blocks. Whether you create an element template from scratch or from a task, you can quickly set and narrow down the properties so process developers using these templates can quickly and confidently wire building blocks together into an E2E process.

Generate an element template

Element template support for all tasks

You can now save any configured task as a reusable element template directly from the Web Modeler properties panel:

• Save configured tasks (service, user, send, receive, business rule, script tasks, or call activities) as templates.
• Edit templates to adjust input/output bindings, validation rules, categories, and more.
• Publish templates to your project or organization for reuse.

Save tasks as element templates

Unlock element template fields

Element template management is now more flexible for developers and DevOps teams.

• You can assign custom semantic IDs and use an intuitive versioning scheme, ensuring templates are portable and retain stable references across different environments.
• Template names and file names can be managed independently, and you can quickly import templates using copy and paste, git sync, or CI/CD pipeline.
• Safeguards now notify you of ID or version conflicts to prevent accidental overwrites when publishing templates.

Manage element templates

FEEL Copilot

Chat with the AI FEEL Copilot for help generating FEEL (Friendly Enough Expression Language) expressions in Web Modeler.

FEEL Copilot

FEEL Playground

Use the FEEL Playground to validate and troubleshoot your FEEL expressions when modeling process diagrams in Web Modeler.

FEEL Playground

Git sync Azure DevOps and Bitbucket integration

In addition to GitHub and GitLab, Web Modeler now supports integration with Azure DevOps and Atlassian Bitbucket.

Git sync

Low-code process testing

You can now save and rerun versioned test scenarios in Web Modeler:

• Supports user tasks, connectors, and basic branching logic
• Ideal for process developers and CoEs
• Enables behavior-driven development with Camunda Process Test

Scenarios

RDBMS support for Oracle and MS SQL in Self-Managed

Web Modeler Self-Managed now supports Oracle Database and Microsoft SQL Server for simpler setup and maintenance.

Set up Oracle or MS SQL

Username claim configuration

Configure the ID token claim used for usernames via the CAMUNDA_IDENTITY_USERNAMECLAIM environment variable to account for differences between identity providers and ensure user-friendly names.

Version description

Use the version Description field to track changes alongside the version tag (for example, as a change log or Git commit message). This helps make versioning more intuitive and collaborative, keeps teams aligned, and reduces ambiguity.

Web Modeler versioning

Technical Changelogs for all 8.8.x releases

Overview of all patch releases and their Changelogs in GitHub

8.8.0-alpha8

Release date	Changelog(s)	Blog
9 September 2025	Camunda 8 core Connectors	Release blog

Ad-hoc sub-process dynamic activation using job workers SaaSSelf-ManagedAgentic orchestration

Ad-hoc sub-process elements can now be activated by job workers:

• Define a task in the process model.
• When the engine reaches the ad-hoc sub-process, a job is created. Completing this job with a job result lets you define which elements in the ad-hoc sub-process to activate.
• Once any flows in the ad-hoc sub-process complete, a new job is created, giving job workers control over what to do next.

note

The job result also supports fulfilling the completion condition for the ad-hoc sub-process. When this condition is met, the sub-process waits for active children to complete before finishing, unless you explicitly cancel remaining instances in the job result.

CI/CD building blocks guide SaaSSelf-ManagedModeler

Element templates documentation is improved to support scalable reuse of building blocks, focusing on CI/CD practices for managing templates across teams and environments.

Key updates include:

• Guidance on integrating element templates and dependencies into CI/CD pipelines
• Restructuring the documentation to improve discoverability and usability
• Tool-agnostic content covering Web Modeler, dependency management, and local development setups
• Moving relevant information from Connectors to the element templates section

See element templates in Modeler for details.

Connectors SaaSSelf-ManagedConnectors

Job header binding

Use the @Header annotation to bind a header value (for example, a FEEL expression) to a model class.

SQL connector Oracle database connection

The SQL connector now supports Oracle Database connections.

See SQL connector for more information.

note

You must manually download the Oracle JDBC driver from Oracle due to licensing restrictions. Run the connector in hybrid mode to enable Oracle Database connections.

Console SaaSSelf-ManagedConsole

Docker distribution

Console is now available as a Self-Managed Docker distribution:

• Deploy Camunda Console using a Docker image
• Run Console outside Kubernetes using Docker Compose or other container services
• Use Console with Camunda 8 clusters deployed via Docker

See Docker Compose developer quickstart for setup instructions.

Usage metrics for licence model and tenant

Usage metrics now support per-tenant reporting and align with Camunda’s updated licensing model based on the number of tenants.

Element template support for all tasks SaaSSelf-ManagedModeler

You can now save any configured task as a reusable element template directly from the Web Modeler properties panel:

• Save configured tasks (service, user, send, receive, business rule, script tasks, or call activities) as templates
• Edit templates to adjust input/output bindings, validation rules, categories, and more
• Publish templates to your project or organization for reuse

See save tasks as element templates for details.

Additional support includes:

• zeebe:assignmentDefinition for assigning users or groups
• zeebe:priorityDefinition for setting task priorities (integer or FEEL expression)
• zeebe:taskSchedule for templating task scheduling configurations

Migrate taken sequence flows flowing to the joining gateway

Enhanced migration now supports taken sequence flows leading to joining gateways:

• Define migration plans mapping active elements and taken sequence flows
• Configure plans via Operate UI or API

Operate and Tasklist API deprecation SaaSSelf-ManagedAPI

The Operate and Tasklist APIs are now deprecated in favor of the Orchestration Cluster API for task and process management.

note

The deprecated APIs remain functional in this release but will no longer receive feature updates and will be removed in version 8.10.

Orchestration Cluster SaaSSelf-ManagedOrchestration Cluster

Orchestration Cluster Identity

Self-Managed Identity Management:

• Create and manage users, groups, roles, and memberships directly in Identity’s database
• Integrate external providers like Keycloak or Microsoft Entra via OIDC
• Assign resource-level permissions using RBAC
• Map users, groups, and roles to resources using token claims and application/client mappings
• Simplify migration with built-in tools

SaaS enhancements:

• Integrate organizational identity providers for centralized user management
• Assign roles, groups, and authorizations per cluster

Orchestration Cluster scaling

Add Zeebe partitions to a running cluster in SaaS to increase capacity without downtime:

• New partitions start processing tasks immediately
• Existing messages remain on original partitions, potentially causing slight imbalances for message-heavy workloads (future updates will address this)

Run Orchestration Cluster without secondary storage

You can now run Orchestration Cluster in "Zeebe-only" mode without secondary storage:

• Ideal for setups not using Query APIs, Operate, or Tasklist
• Enable by setting camunda.database.type=none
• Starts only the required Zeebe components

Web Modeler SaaSSelf-ManagedWeb Modeler

Cluster Identity compatibility

Web Modeler is fully compatible with Camunda 8.8 Orchestration Clusters for a smooth migration.

Low-code Process Testing

You can now save and rerun versioned test scenarios in Web Modeler:

• Supports user tasks, connectors, and basic branching logic
• Ideal for process developers and CoEs
• Enables behavior-driven development with Camunda Process Test

RDBMS support for Oracle and MS SQL in Self-Managed

Web Modeler Self-Managed now supports Oracle Database and Microsoft SQL Server for simpler setup and maintenance.

Test scenario files

You can now save, export, and share test scenarios:

• Portable, Git-syncable files enable quick creation of Camunda Process Test scenarios

Username claim configuration

Configure the ID token claim used for usernames via the CAMUNDA_IDENTITY_USERNAMECLAIM environment variable to account for differences between identity providers and ensure user-friendly names.

8.8.0-alpha7

Release date	Changelog(s)	Blog
12 August 2025	Camunda 8 core Connectors	Release blog

Camunda 8 API SaaSSelf-ManagedAPI

Development continues on the single unified Camunda 8 REST API that consolidates multiple fragmented APIs into a single, coherent interface, simplifying development and improving clarity across Camunda components.

Process instance tags SaaSSelf-ManagedAPI

Introduce optional, immutable process instance tags set at creation for lightweight routing, correlation, and future prioritization without inspecting large variable payloads.

For a full feature overview see process instance creation and job workers.

Camunda 8 Run supports 8.8 architecture Self-ManagedDeveloper

Camunda 8 Run now includes Identity, allowing all core applications to run locally in configurations similar to production. This simplifies local development by enabling multiple user authentications and credentials.

Connectors SaaSConnectors

AI Agent connector

• Hybrid mode/customization: The AI Agent connector can now be customized in hybrid mode. This includes an API to define custom memory storage backends apart from the provided ones.
• Added model provider support: Added support for Azure OpenAI and Google Vertex AI models.

To learn more, see AI Agent connector.

Azure Blob Storage connector

Use the new Azure Blob Storage connector to store and retrieve documents within Camunda workflows using Azure Blob Storage. This enables seamless document management directly within processes, improves efficiency, and reduces the need for custom integrations.

To learn more, see Azure Blob Storage connector.

CSV connector

Use the new CSV connector for SaaS to read, filter, transform, and write CSV data within processes. This reduces technical debt, accelerates development, and broadens integration capabilities with native support for this universal data format.

To learn more, see CSV connector.

Google Cloud storage connector

Use the new Google Cloud storage connector for easy document storage and retrieval directly within Camunda workflows, streamlining document management without custom development.

To learn more, see Google Cloud storage connector.

MCP Client connector early access

Use the new MCP Client connector to allow Camunda processes and AI agents to auto-discover and invoke external tools, eliminating hardwired connectors and enabling dynamic, metadata-driven tool integration.

To learn more, see MCP client.

note

The MCP Client connector is released as an early access alpha feature to allow you to test and participate in development by sharing feedback before general availability, and is subject to alpha feature limitations.

Vector database connector

Improvements are made to the Vector database connector as follows:

• Updated the OpenSearch vector store to support non-AWS managed instances.
• Added support for OpenAI embedding models.

To learn more, see vector database connector.

Fetch latest process definitions

The connectors runtime is optimized by fetching only the latest process definition versions from the Orchestration Cluster. This reduces CPU consumption and improves deployment performance, particularly in environments with frequent CI/CD-generated process versions.

Unlock element template fields

Element template management is now more flexible for developers and DevOps teams.

• You can assign custom semantic IDs and use an intuitive versioning scheme, ensuring templates are portable and retain stable references across different environments.
• Template names and file names can be managed independently, and you can quickly import templates using copy and paste, git sync, or CI/CD pipeline.
• Safeguards now notify you of ID or version conflicts to prevent accidental overwrites when publishing templates.

Run process segment SaaSSelf-ManagedAPI

This feature allows developers to manually execute and test individual tasks or segments (connectors, RPA bots, IDP extractions) without running full processes, improving debugging and development efficiency.

To learn more, see run process segment

Intelligent document processing (IDP) form extraction SaaSIDP

You can use form-based structured document extraction to capture data from structured documents.

• For example, you can use this extraction method for documents with a consistent layout, such as invoices, tax forms (for example, W-2s, VAT declarations), and loan or insurance applications.
• Projects can be shared organization-wide, enhancing accessibility to extraction capabilities.

To learn more, see extract structured data.

Migration to Orchestration Cluster Identity support Self-ManagedSecurity

Allows smooth migration from Camunda 8.7 to 8.8 by transferring tenants, roles, and authorizations to the new Orchestration Cluster Identity, minimizing manual administration effort during upgrade.

RPA multi-file script support RPA

Robotic process automation (RPA) now supports multi-file script support, allowing you to organize scripts modularly, reuse common automation components, and integrate existing Robot Framework scripts. The execution engine fully supports multi-file scripts and linked resources, improving scalability, maintainability, and flexibility for enterprise automation projects.

User task listener metadata and filtering SaaSSelf-ManagedTasklist

With this release, user task listener jobs are improved as follows:

• Task metadata is now directly embedded in the task listener jobs' properties instead of being exposed as custom headers. This includes attributes such as assignee, dueDate or userTaskKey.
• User tasks can now be filtered using partial user task states to understand the current lifecycle state of the user task fully.

These improvements simplify job worker development, reduce errors, and enable better observability of the user task lifecycle.

Unified configuration for the Orchestration Cluster Self-Managed

Simplifies configuration by consolidating Operate, Tasklist, and Identity profiles into a unified Camunda 8 Orchestration Cluster application, reducing duplication and complexity for easier deployment and management.

Zeebe-managed resilient batch operations Self-ManagedZeebe

All batch operations, such as canceling or resolving incidents in bulk, are now handled by Zeebe instead of Operate.

• This change ensures region failovers in the multi-region setup no longer risk losing critical batch commands.
• Users will initiate and manage batch operations through the Orchestration Cluster REST API and the Operate UI, but the underlying processing occurs within Zeebe.
• By moving batch operations to the core engine, multi-region deployments gain reliability and resilience.

8.8.0-alpha6

Release date	Changelog(s)	Blog
8 July 2025	Camunda 8 core Connectors	Release blog

Bitbucket sync SaaSWeb Modeler

Camunda 8 now supports integration with Atlassian Bitbucket Cloud, in addition to GitHub, GitLab, and Azure DevOps.

• This helps customers who use Jira for their development processes.
• Organization owners and administrators can connect their Web Modeler process applications to Bitbucket Cloud, allowing users to keep their Web Modeler, Desktop Modeler, and official version control projects synced.

To learn more, see Git sync.

Camunda 8 REST API renamed to Orchestration Cluster REST API SaaSSelf-ManagedAPI

The Camunda 8 REST API is now called the Orchestration Cluster REST API.

• This name better reflects its role as a unified REST API for interacting with entities in a Camunda 8 Orchestration Cluster, such as processes, tasks, and variables.
• The functionality and structure of the API remain unchanged. The name change improves clarity and onboarding across Camunda documentation and resources.

To learn more, see Orchestration Cluster REST API.

Connectors SaaSSelf-ManagedConnectors

AI Agent connector

• Structured outputs/JSON mode: Configurable response formats allow you to choose whether the connector returns plain text or JSON for downstream processing. For some models, you can define a JSON schema for returned data.
• Conversation history storage: History can now be stored in Camunda's document storage rather than in process variables—allowing longer histories without process variable size limits.

To learn more, see AI Agent connector.

Intrinsic functions

A new getJson intrinsic function accepts a document and an optional FEEL expression. It extracts and returns content from a JSON document as an object.

• The optional FEEL expression parameter specifies the part that will be extracted from the JSON document content.
• If not provided, the whole document is returned as a JSON object.

To learn more, see intrinsic functions.

Dynamic partition scaling Self-ManagedZeebe

You can now add new Zeebe partitions to a running cluster.

• Scaling can be performed concurrently when the cluster is running, with zero downtime.
• New process instances also start on new partitions, distributing cluster load evenly across partitions.
• Process instances do not migrate between partitions, so it can take time for the cluster to reach equilibrium.
• New partitions do not take part in correlating messages/signals, except for message/signal start events.

To learn more, see cluster scaling.

caution

This feature is not yet fully compatible with backup/restore.

Helm charts Self-Managed

Alternative container images

Camunda now provides alternative container images to the previously used Bitnami images. These images are hosted on registry.camunda.cloud.

• From version 8.8, these are the default supported images, offering better security and faster patch delivery.
• To use them, update your Helm deployment to reference the values-images-ee.yml file. See the installation guide for details.

Configurable volumes

The Helm chart now supports configurable volumes. You can define PersistentVolumeClaims or continue using EmptyDir through values.yaml.

Singapore region available for SaaS on Amazon Web Services SaaS

A new Singapore (ap-southeast-1) region is now available for SaaS clusters on Amazon Web Services. Use this region to:

• Improve overall processing speed and reduce latency if you operate in Singapore and Southeast Asian (SEA) countries.
• Keep cluster data within Singapore to support your local data residency and compliance needs.

To learn more about supported SaaS regions, see regions.

Tasklist uses the Orchestration Cluster REST API SaaSSelf-ManagedTasklist

Tasklist now uses the Orchestration Cluster REST API, replacing the soon-to-be-deprecated Tasklist V1 API.

• This change improves compatibility with Camunda 8 RDBMS support and continues to work with Elasticsearch/OpenSearch.
• It ensures consistent functionality, better performance, and access to new features—without breaking existing workflows.

User task listener types SaaSSelf-ManagedTasklist

New user task listener types are available:

creating event

This event triggers before a user task is created.

Functionality	Description
Configurable creation listener	Executes logic before a task appears to users. Task is visible only after all listener jobs finish.
Controlled task initialization	The creation continues only after listeners succeed. Incidents are raised if logic fails, enabling safe retries.
Operate UI insights	A “Creating” event appears in the listener tab in Operate. Incidents are flagged for troubleshooting.
Assign user task	Assign a task programmatically during creation, useful when assignment depends on an external system.

canceling event

This event triggers when a user task is canceled (e.g., by a boundary event or process termination).

Functionality	Description
Configurable cancelation listener	Executes logic when a task is canceled. Allows inspection or modification of task data before completion.
Consistent lifecycle control	Cancelation waits for listener logic to complete. Failures can raise incidents for safe retry.
Operate UI insights	A “Canceling” event is shown in the listener tab. Incidents are highlighted for visibility.

To learn more, see user task listeners.

Documentation SaaSSelf-Managed

Get started updates

The getting started documentation now includes:

• Example BPMN files and Spring Boot/NodeJS starter projects.
• Practical code snippets, such as payment handling.
• Updated code to match recent Camunda versions.
• Annotations in BPMN files to guide usage and explain results.

Public API

The new Public API documentation outlines what’s included in Camunda 8's public API, the policies around versioning, and what to expect when upgrading.

• The public API is the official contract between Camunda and its users under SemVer.
• No breaking changes will be made to the public API in minor or patch releases.
• You can safely build on these interfaces with the expectation of stability and backward compatibility.

8.8.0-alpha5

Release date	Changelog(s)	Blog
10 June 2025	Camunda 8 core Connectors	Release blog

Agentic orchestration

The following agentic orchestration features are available in this alpha release:

AI Agent connector ConnectorsWeb ModelerDektop Modeler

The AI Agent connector enables AI agents to integrate with an LLM to provide interaction/reasoning capabilities. This connector is designed for use with an ad-hoc sub-process in a feedback loop, providing automated user interaction and tool selection.

This connector provides:

• Support for different LLM providers, such as OpenAI or Anthropic.
• Conversational/short-term memory handling to enable feedback loops and follow-up tasks.
• Tool resolution and orchestration through tools defined in an ad-hoc sub-process.

To learn more about this connector, see AI Agent connector.

Vector database connector Connectors

The vector database connector allows embedding, storing, and retrieving Large Language Model (LLM) embeddings. This enables building AI-based solutions for your organizations, such as context document search, long-term LLM memory, and agentic AI interaction in combination with the AI Agent connector (RAG).

To learn more about this connector, see vector database connector.

Azure DevOps integration for Git sync Web Modeler

Camunda now supports an integration with Azure DevOps that allows for direct syncing with Azure repositories.

FEEL Copilot Web Modeler

Chat with the AI FEEL Copilot for help generating FEEL (Friendly Enough Expression Language) expressions in Web Modeler.

To learn more about this feature, see FEEL Copilot.

note

The FEEL Copilot is released as an early access alpha feature to allow you to test and participate in development by sharing feedback before general availability, and is subject to alpha feature limitations.

FEEL Playground Web Modeler

Use the FEEL Playground to validate and troubleshoot your FEEL expressions when modeling process diagrams in Web Modeler.

To learn more about this feature, see FEEL Playground.

Identity Identity Self-ManagedSaaS

Camunda’s new Identity service enhances authentication and authorization for Self-Managed and SaaS environments:

Feature/enhancement	Description
Self-Managed Identity management	Admins can create and manage users, groups, roles, and memberships directly in the Identity database.
OIDC integration	Integrate external identity providers (IdP) such as Keycloak and Microsoft Entra.
Role-based access control (RBAC)	Assign roles and group permissions on a per-resource basis for fine-grained access control. Supported resources include Authorization, Claim Mapping Rules, Messages, Batches, Applications, Tenants, Deployments, Process Definitions, Decision Definitions, and more.
Flexible mapping	Map users, groups, and roles to resource authorizations and tenants. Leverage OIDC token claims and application/client mappings to streamline permission assignments.
Migration support	Simplified migration tools make it easy for existing customers to transition to the new service.
Organizational Identity	In SaaS environments, integrate your own IdP to manage organizational users and assign resources cluster-by-cluster.
Cluster-specific Roles and Groups	In SaaS environments, manage distinct roles, groups, and authorizations for each cluster independently.

Identity management for SaaS clusters SaaS

Orchestration Cluster Identity is now available for SaaS clusters. Starting with this alpha version, you can manage groups, roles, and authorizations at the cluster level.

The following known limitations apply for this alpha version release:

Known limitation	Description
Authorizations	Before enabling authorization checks in the cluster settings, users must assign themselves to the admin role in Identity for the Orchestration Cluster. Note: As authorizations are disabled by default, no changes are required for initial access. Authorizations cannot be assigned to users via the UI, only to groups. Authorizations are not correctly loaded in the UI.
Navigation, Notifications, and Logout	Links to the other Camunda components in the Orchestration Cluster web applications (Operate, Tasklist, Identity) do not currently work. SaaS notifications are not displayed in Orchestration Cluster components. Log out from Orchestration Cluster web applications is not fully functional.
Documentation	Documentation is incomplete.

Identity management for Helm Chart setups Self-Managed

Orchestration Cluster Identity is now available for OIDC setups in Helm chart deployments. Starting with this alpha version, you can configure the Orchestration Cluster components to use the identity provider (IdP) of your choice and enable single sign-on (SSO).

The following known limitations apply for this alpha version release:

Known limitation	Description
Authorizations	Authorizations cannot be assigned to users via the UI, only to groups. Authorizations are not correctly loaded in the UI.
Logout	Log out from Orchestration Cluster web applications is not fully functional.
Documentation	Documentation is incomplete.

React to expired messages with a custom exporter

Camunda now introduces a mechanism to react to expired messages. The original message is retrieved from the primary storage upon message expiration and the expiry event is enhanced with the original message. Existing functionality remains unchanged, so there is no disruption for current customers or custom exporters.

With these updates, developers can subscribe to the expired messages through a custom exporter, examine the event content, and optionally re-publish or handle the message differently. By providing an enhanced event and re-publish flow, this feature strengthens reliability and transparency in business processes without requiring a major upgrade or modifying existing exporters.

RPA worker offline installer

This feature introduces an offline installer package for the Camunda RPA worker, allowing installation without internet connectivity. The offline installer removes reliance on external repositories or downloads, ensuring consistent, secure, and hassle-free deployment into air-gapped or restricted environments.

Tasklist frontend application migration to use Orchestration Cluster REST API Tasklist

The Tasklist frontend application is transitioning from the soon-to-be-deprecated Tasklist V1 API to the unified Orchestration Cluster REST API.

• This ensures Tasklist remains fully compatible with Camunda 8’s new RDBMS support while continuing to work seamlessly with Elasticsearch and OpenSearch.
• You can expect consistent functionality across different data layers, improved performance, and access to new platform features - all without losing existing capabilities or disrupting task management workflows.

8.8.0-alpha4

Release date	Changelog(s)	Blog
13 May 2025	Camunda 8 core Connectors	Release blog

Camunda Process Test H2 data layer support Self-ManagedZeebe

Camunda Process Test now supports using the H2 Database Engine as the default embedded data layer.

• H2 is now automatically provisioned when integrating the Camunda Process Test libraries, eliminating manual database configuration and reducing memory footprint.
• H2 support streamlines the developer experience for your Spring Boot and plain Java projects. Process testing is now faster to set up, simpler to maintain, and easier to integrate with your continuous integration workflows.

To learn more about Camunda Process Test, see Camunda Process Test.

Connector manage and run supports multiple runtimes SaaSSelf-ManagedConsole

Connector manage and run in Console now supports management of multiple connector runtime instances.

To learn more about this feature, see manage your connectors.

Connectors SaaSSelf-ManagedConnectors

Email connector

The Email connector now exposes the Message-ID provided by the client in the connector response payload. This allows for improved traceability, easier correlation between sent messages and logs, and better integration with downstream systems that rely on Message-ID.

For example:

{
  "sent": true,
  "subject": "Email subject"
  "messageId": "<abc123@clientdomain.com>"
}

note

This change is backwards-compatible and does not require any action. You can now optionally use the messageId field for enhanced tracking when parsing connector responses.

Hubspot connector

Hubspot connector enhancements include:

• The Get contact by ID operation now supports the retrieval of properties and default contact properties.
• The new Enroll contact to workflow operation allows you to enroll contacts into a specified workflow.

To learn more about this connector, see HubSpot connector.

Desktop Modeler settings Self-ManagedDesktop Modeler

The new Settings window in Desktop Modeler allows you to configure the application and customize your modeling experience. You can select your default execution platform version, along with other options that were previously only available as flags.

To learn more about these settings, see Desktop Modeler settings.

Version descriptionSaaSSelf-ManagedWeb Modeler

Use the version Description field to track changes alongside the version tag (for example, as a change log or Git commit message). This helps make versioning more intuitive and collaborative, keeps teams aligned, and reduces ambiguity.

To learn more about versioning your diagrams, see versions.

Web Modeler cluster basic authentication Self-ManagedWeb Modeler

As well as bearer token and client credentials authentication, you can now configure Web Modeler in Self-Managed to use basic authentication for cluster access.

• To use basic authentication, set the CAMUNDA_MODELER_CLUSTERS_0_AUTHENTICATION environment variable value to BASIC.
• Web Modeler sends a username and password with every request to the cluster.

To learn more about basic authentication, see available authentication methods.

8.8.0-alpha3

Release date	Changelog(s)	Blog
08 April 2025	Camunda 8 core Connectors	-

Ad-hoc sub-process activation API & completion configuration SaaSSelf-Managed

Agentic process orchestration enhancements include:

• An optional completionCondition boolean expression for ad-hoc sub-processes that is evaluated every time an inner element is completed. A cancelRemainingInstances boolean attribute can also be configured to influence the ad-hoc sub-process behavior when the completion condition is met.
• An Activate activities within an ad-hoc sub-process API used to activate selected activities within an ad-hoc sub-process.

To learn more about these features, see ad-hoc sub-processes.

Advanced User Task Listeners for Updating Events SaaSSelf-ManagedTasklist

Advanced User Task Listeners for Updating Events allow you to define listeners that trigger whenever certain task properties or variables change.

• These listeners generate jobs similar to other event-based task listeners, granting direct access to task data as well as the ability to accept or roll back updates (in certain scenarios).
• Operators can also view, manage, and resolve incidents caused by these listeners in Operate, ensuring a unified and transparent approach to handling task changes.

To learn more about this feature, see advanced user task listeners for updating events.

HubSpot connector SaaSSelf-ManagedConnectors

Use the new outbound HubSpot connector to connect your BPMN service with HubSpot and manage your HubsSpot contacts, companies, and deals.

This connector supports the following operations:

• Contacts: Get all contacts, Get contact by id, Get multiple contacts by id, Search contact, Create contact, Update contact, Delete contact.
• Companies: Get all companies, Get company by id, Search company, Get all contacts of a company, Add contact to company, Remove contact from company, Create company, Delete company.
• Deals: Get all deals, Get deal by id, Search deal, Delete deal.

To learn more about this connector, see HubSpot connector.

8.8.0-alpha2

Release date	Changelog(s)	Blog
11 March 2025	Camunda 8 core Connectors	Release blog

Camunda Spring Boot Starter for the Camunda 8 REST API

A Spring Boot Starter is provided for the Orchestration Cluster REST API to unify process management, user tasks, and identity features under a single dedicated starter. This simplifies the interaction between a Spring Boot application and Camunda 8, allowing you to:

• Easily integrate process entity management and queries within your workflows.
• Seamlessly configure endpoints and authentication via Spring Boot auto-configuration, minimizing boilerplate code.
• Rely on an official, standardized approach to guarantee consistency and reduce maintenance costs when upgrading.
• Based on Spring Boot 3.5 (version compatibility matrix).

To learn more about this feature, see the Camunda Spring Boot Starter.

Camunda 8 Run authentication updates

Camunda 8 Run no longer requires authentication when working with APIs. Authentication and authorizations can be optionally enabled to allow requests using basic authentication, and to test authorizations and permissions.

To learn more about this feature, see the API documentation for Camunda 8 Run.

Identity management updates SaaSSelf-Managed

The Identity service is enhanced to deliver greater flexibility, control, and security for both Self-Managed and SaaS users. These updates are part of our broader effort to streamline the platform’s architecture.

Cluster-level identity management

Identity settings will be configured at the Orchestration Cluster level, allowing each cluster to have unique OIDC configurations. This cluster-specific setup empowers organizations to assign different identity providers (IdPs) across clusters, offering improved control over permissions and user group mappings, resulting in a more streamlined and efficient configuration experience.

For SaaS customers, identity management in Camunda 8.8 remains consistent with Camunda 8.7, allowing the attachment of a single IdP per organization. However, cluster-level identity capabilities are provided for SaaS as well as Self-Managed. This means that user groups, roles, and access permissions can now be managed at the cluster level, giving SaaS customers the same granular access control as in Self-Managed environments.

Decoupling from Keycloak Self-Managed

Built-in Keycloak integration in Self-Managed is removed, allowing customers to use any compatible IdP.

• Keycloak remains fully supported as an external option. For cluster-level identity management it must be connected as an external OIDC provider moving forward.
• OpenID Connect (OIDC) remains the standard for seamless integration with chosen IdPs.

Resource-based permissions

Resource-level permissions are introduced to control read and write permissions per specific resource.

• Admin users retain full access, but regular users must be granted specific permissions to perform operations or view resources.
• For organizations that build custom front-ends and access Camunda via API, users or Clients with API permissions can still access data through the V2 API, respecting their resource permissions.

8.8.0-alpha1

Release date	Changelog(s)	Blog
11 February 2025	Camunda 8 core Connectors	Release blog

note

Some features available in 8.8.0-alpha1 were originally released in 8.7.0 alphas. These features are longer planned for release in 8.7.0. For more information, see the Camunda 8.7 and 8.8 release update blog.

User task listeners SaaSSelf-ManagedModelerOperate

Task lifecycle management is enhanced with user task listeners, allowing users to react to specific user task lifecycle events.

• Process designers can now model task listeners for different events, such as assigning and completing.
• Developers can use the same job infrastructure to activate and complete task listener jobs.
• Operations engineers can easily check details of active and completed task listeners within instances, and efficiently resolve task listener incidents.

This enhancement streamlines operations and ensures smoother incident handling, improving time to unblock process execution.

To learn more about this feature, see user task listeners.

Orchestration Cluster REST API Query API API

You can now use a single Query API in the Orchestration Cluster REST API to find process and decision data instead of using multiple component APIs.

For example, send a request to the Search decision definitions endpoint to search for decision definitions.

New Query API endpoints are added as follows:

• Decision definitions
• Decision instances
• Decision requirements
• Flownode instances
• Incidents
• Process definitions
• Process instances
• User tasks
• Variables

To learn more about these endpoints, see the Orchestration Cluster REST API documentation.

Amazon OpenSearch Optimize support Self-ManagedOptimize

Camunda 8 Self-Managed now fully supports the use of Amazon OpenSearch with Optimize.

Process instance migration SaaSSelf-ManagedZeebe

Enhanced process instance migration allows you to solve problems with process definitions and use the latest process improvements.

You can now migrate the following:

• Compensation boundary event subscriptions
• Escalation boundary events
• Escalation event subprocesses

To learn more about migration, see process instance migration.

Camunda Exporter Self-Managed

A new Camunda Exporter brings the importer and archiving logic of web components (Tasklist and Operate) closer to the distributed platform (Zeebe). The index schema is also being harmonized.

To learn more about this feature, see the Camunda Exporter documentation.

Backup and restore improvements Self-Managed

Camunda backups have been improved and made easier to use. The web application backups are now merged together under one endpoint.

Connector Runtime SaaSSelf-ManagedConnectors

Spring SDK and Orchestration Cluster REST API Migration

The Connectors experience is enhanced with the migration from the Spring Zeebe to the Orchestration Cluster REST API, and the removal of dependency on the Operate client.