General information
Anjana Installer is the deployment tool for Anjana Data Platform, designed as an evolution of the traditional Ansible kit. It is an application with a web graphical interface and a command-line interface (CLI) that allows managing the complete lifecycle of Anjana installations in IaaS, PaaS, and OnPremise environments.
The installer's goal is to simplify and standardize the platform's deployment, update, and maintenance process, eliminating the need for advanced Ansible knowledge and reducing errors resulting from manual configuration.
What's new compared to the Ansible kit
New user experience
-
Web graphical interface: all configuration, deployment, and operation is done from a browser, without needing to edit YAML files or run playbooks.
-
Standalone binary: it is distributed as a single executable that includes everything needed, without requiring Ansible, Python, or additional dependencies. Styling is unified with the Anjana platform for a consistent visual experience.
Deployment improvements
-
Automatic dependency management: the installer detects which persistence layers are available and does not allow installing the core unless they are satisfied.
-
Service descriptor-level dependencies: microservices automatically wait for their dependencies to be up before starting, eliminating the manual ordered startup of the previous kit. Unified deployment of microservices and plugins by iterating over categorized lists (Core / Plugins).
-
Automatic adaptation to the deployment mode: in Single VM environments without cloud persistence layers, it is not necessary to fill in connection URLs.
-
Preflight phase before deployment: before Platform Setup and Full Install, the installer validates the configuration and the target hosts (connectivity, permissions, resources, ports, packages) and offers guided remediation of detected issues directly from the interface.
Security improvements
-
Encrypted secrets management: all credentials are stored encrypted at rest, eliminating the plain-text environment files of the previous kit.
-
Random TLS keystore password per deployment (ksPass): the installer generates a random password for the TLS keystore on every platform setup run, storing it encrypted in the SecretsService. The microservices' service descriptors never contain a predictable static password.
-
Compatibility with AWS Secrets Manager with support for a JSON blob shared with Horus (requires IRSA).
-
Encrypted local storage (AES-256-GCM) for environments without access to cloud services.
-
Improved name resolution in the hosts file, per machine instead of per microservice.
-
Persistence credential rotation with a mandatory security policy, and protection of deployed credentials during configuration import.
-
IP-based access control for the platform proxy managed from the interface (general and internal whitelist), with hot application and automatic validation.
Operations and maintenance improvements
-
Update with automatic rollback in case of failure.
-
Enriched status panel that cross-references multiple sources (service, logs, and Eureka registration) with automatic version detection.
-
Audit logs separated into an independent pipeline for integration with SIEM systems.
-
Full command-line interface for automation and integration with CI/CD pipelines.
-
Complete persistence tools: backup, restore, restore from file (.sql, .tar.gz), data export/import for migration between environments, reset with optional restoration modes, sample data loading, and a real-time log viewer.
-
Automatic re-population of Horus configuration after operations that modify the database (RESET_ALL, RESTORE_ALL, LOAD_SAMPLE_DATA).
Configuration import
The installer allows importing existing configurations in two ways:
-
Import from a previous Ansible kit (all.yaml + hosts.yaml, and optionally anjanauihosts.yaml to carry over access whitelists) with automatic topology detection and value mapping.
-
Import from a YAML previously exported by the installer itself. During import, credentials for persistence layers that are already deployed are not overwritten; a notice is shown to the user so they can manage them from Tools > Credentials.
Support
|
Supported versions |
26.1 (beta) |
|---|---|
|
Supported system |
Ubuntu 24.04, RedHat 9 |
|
Deployment modes |
Single VM, Hybrid, Distributed, Balanced, Dual VM |
Included versions
|
Component |
Version |
|---|---|
|
Java (JDK) |
17 |
|
PostgreSQL |
16 |
|
Solr |
9.9.0 |
|
Zookeeper |
3.9.4 |
|
SeaweedFS |
4.16 |
|
MongoDB |
7.0 |
|
Valkey |
7.2 |
|
RabbitMQ |
3.12 |
|
OpenSearch |
2.19 |
The versions are based on the latest release available in the Ubuntu 24.04 package repository at the time of release. On RedHat 9, versions may vary slightly depending on the availability of official packages. Solr + ZooKeeper and OpenSearch are mutually exclusive: the configuration wizard allows choosing one or the other as the indexing/search engine.
NOTE: OpenSearch is not yet available as a selectable option in the wizard (marked "Coming soon"); the active engine today is Solr + ZooKeeper.
Included assets
The installer includes the following:
|
Microservices - Core |
|
|---|---|
|
portuno kerno zeus drittesta minerva viator hermes tot horus marketplace |
Anjana microservices |
|
mcp |
AI Bridge, core service enabled by default |
|
anjana-ui |
Frontend UIs at / and /configpanel |
|
Persistence layers |
|
|
PostgreSQL |
Relational database service |
|
Solr + ZooKeeper |
Indexing and coordination engine (mutually exclusive with OpenSearch) |
|
SeaweedFS |
S3-compatible object storage service |
|
MongoDB |
Document database (required by portuno 26.1) |
|
Valkey |
In-memory cache, a fork of Redis (required by kerno 26.1) |
|
RabbitMQ |
Messaging broker (required by hermes 26.1) |
|
OpenSearch |
Search and analytics engine (alternative to Solr+ZooKeeper) |
|
Plugins (20) |
|
|
tot-plugin-jdbc tot-plugin-jdbc-oracle tot-plugin-jdbc-snowflake tot-plugin-jdbc-redshift tot-plugin-jdbc-sqlserver tot-plugin-jdbc-denodo tot-plugin-aws-s3 tot-plugin-aws-glue tot-plugin-aws-iam tot-plugin-azure-ad tot-plugin-azure-storage tot-plugin-gcp-bigquery tot-plugin-gcp-iam tot-plugin-gcp-storage tot-plugin-ldap tot-plugin-aqtiva tot-plugin-powerbi tot-plugin-ranger tot-plugin-tableau tot-plugin-databricks |
Anjana plugins |
Recommendations
Version control
To ensure effective control of the versions and changes to the installer's configurations, using Git repositories as part of the workflow is essential. It is recommended that anyone using this installer have a repository or tool that allows tracking changes over time, to ensure the integrity of the work and recovery from unwanted changes.
Migration from the Ansible kit
For existing environments deployed with the Ansible kit, the installer provides a built-in migration procedure. From the graphical interface, the user can import the all.yaml and hosts.yaml files from the previous Ansible kit (and optionally anjanauihosts.yaml to carry over the IP-based access whitelists). The installer will automatically map the configuration, detect the topology, and offer to migrate from Solr to OpenSearch if applicable. Credentials for persistence layers that are already deployed are protected during import and are not overwritten. It is recommended to check the Installer deployment manual to become familiar with the configuration to be migrated.