Skip to main content

Helm installation requirements and included components

Harness recommends Helm package manager for the installation and deployment of Self-Managed Enterprise Edition. Helm offers benefits including:

  • Declarative and dynamic application management
  • Built-in scalability
  • Chart reuse across environments
  • Repeatable results you can automate

The Harness Helm chart packages the core modules and components that are required to operate Harness at scale.

For instructions on installing Self-Managed Enterprise Edition using Helm, go to Install using Helm.

Requirements

The following infrastructure is required to install Self-Managed Enterprise Edition using Helm.

note

Support each node with 8 cores vCPU and a minimum of 32 GB memory.

Production environment

The production environment requires the following resources.

ModulesPodsCPUMemory (GB)Storage (GB)
Platform (including CD, GitOps, OPA)4050.4125.41090
CCM111227.31124
CI22120
STO4370
FF3360
Chaos15122450
CET1012280

Development environment

The development environment requires the following resources.

ModulesPodsCPUMemory (GB)Storage (GB)
Platform (including CD, GitOps, OPA)2223.960.6550
CCM111227.31124
CI1160
STO21.53.50
FF3360
Chaos54820
CET21.550

Included components

Harness Helm chart includes the following components.

Table 1. Platform components for Continuous Delivery

ComponentDescription
Access controlProvides pipelines with access controls including Kubernetes Role-Based Access Control (RBAC).
Data captureResponsible for the capture of data related to the operation of Harness Pipelines, including but not limited to events, tasks, metrics, and logs.
CV NextgenProvides continuous verification (CV) services to Pipeline components including deployments, services, and logs. Aggregates data from multiple providers, including performance metrics, from monitoring activities for dashboard presentation.
GatewayManages application gateway services across Harness Pipelines.
Harness ManagerResponsible for the analysis and presentation of actionable data from the end-to-end Harness Pipeline in an administrative user interface.
Harness Storage ProxySupplies proxy services for storage.
LE NextgenSupplies Harness Learning Engine (LE), a machine-learning component used to fine-tune Pipelines and identify and flag anomalies.
Log serviceProvides frontend logging services to Harness Pipelines.
MinIOA distributed object storage system providing Kubernetes-based, S3-compatible storage. You can use MinIO to store runtime logs (build logs) for Harness Pipelines.
MongoDBA NoSQL database offering high-volume storage of data represented as key-value pairs contained in documents and collections of documents.
NextGen UIProvides the user interface for Harness NextGen.
NgAuth UIUser interface component for the AngularJS ng-auth authentication service.
NgManagerProvides NextGen Harness Manager.
Pipeline serviceSupports creating a pipeline.
Platform serviceRepresents the Harness Platform service.
RedisProvides services for Redis, an in-memory data structure store.
Scm serviceProvides source code management services.
Template serviceProvides Harness templates to enable the design of reusable content, logic, and parameters in Pipelines.
Test Intelligence ServiceProvides the Test Intelligence service.
TimescaleDBProvides the TimescaleDB time-series SQL database.

Table 2. Harness Platform components

The following components are included in addition to the Harness Platform components.

ComponentModule
Ci-managerContinuous Integration
Sto-coreEnables the creation and management of Harness Security Testing Orchestration
Sto-managerProvides core services for Harness Security Testing Orchestration
Et-serviceProvides core services for Continuous Error Tracking (CET)
Et-collectorCollects Continuous Error Tracking (CET) Agent data for processing
Et-receiverScalable service responsible for processing data for Continous Error Tracking (CET)

Table 3. Optional dependencies

DependencyDescription
IstioIstio is an open-source service mesh.
Ingress ControllerSupported by default.