Feature Requests

Hello, Currently, custom runner images can only be configured through YAML, as described in the Harness documentation: Harness Plugin Images Documentation: https://developer.harness.io/docs/infra-as-code-management/pipelines/plugin-images/ step: type: IACMTerraformPlugin name: apply identifier: apply timeout: 2h spec: command: apply image: plugins/private_harness_terraform_plugin # (1) connectorRef: privateConnector # (2) We would like to request the ability to add custom runner images directly through the UI, which would improve usability and reduce reliance on YAML configuration. Please let me know if you need any additional details or have questions regarding this request.

This feature request is for below: Allow users to override the folder path within the repository during workspace creation from template. Avoid unlocking the entire repository block, but still provide flexibility for users to specify their own path.

Currently, the Harness IACM stage doesn't have a concept of environments within it. As the infrastructure we are creating can be environment-specific. Please add the environment to the IACM stage as well; this will help provide more granular control over IACM creation.

Currently, when we call the GCP OIDC provider from an IACM execution, the attributes we pass to generate the OIDC tokens are - account_id, organization_id, project_id, and connector_id. Also, Context is always set to PIPELINE_EXECUTION. These attributes don't have any IACM-specific info by which we can add any attribute condition in GCP. We like more attributes to be passed here, such like we pass in cd, Here is the example of attributes harness send in CD - "account_id": "someaccountid", "organization_id": "someordf", "project_id": "sometest", "pipeline_id": "k8s_Pipeline_Rolling", "environment_id": "Env", "environment_type": "Production", "connector_id": "OIDC_Connector", "connector_name": "OIDC Connector", "step_type": "INFRASTRUCTURE_TASKSTEP_V2", "context": "PIPELINE_EXECUTION"

Our current architecture utilizes a single Terraform state file and workspace (monolith) across multiple distinct pipelines (e.g., separate pipelines for S3, VPC, and EC2) for an AWS account. Since we had 19 different aws accounts, different state file for each resources was not feasible. So currently all my resources pipeline updates the same tfvars based on target apply. While we can create resources granularly, we currently lack the ability to safe-delete them. Running a standard terraform destroy in any single pipeline creates a plan to destroy all resources in the shared state file. So we need to update the CI/CD pipeline destroy logic to accept a --target argument (e.g., -target=module.s3_bucket).

Please enable following features PIE_FLEXIBLE_TEMPLATES PIE_FLEXIBLE_TEMPLATES_PHASE2

Building a pipeline that bulk updates multiple workspaces. Everything is working generally but the UI "resources" tab only shows a single workspace.

Retro ticket for request made in Late July/ Early August meeting with iHerb/Harness on IaCM, showing iherb's evaluation. We discuessed with Uri Scheiner that we need a workspace locking mechanism that will pause/suspend workspaces with pending approvals, from running more than plans for new runs. This will avoid situations where state mismatch could roll back changes. Eg; Change 1 applies a code version and is pending an approved change window to complete. In the mean time, change 2 applies a different version of other code referenced in the same terraform/tofu workspace, and is applied completely. If change 1 is continued after change 2, the updated from change 2 will be reverted as the state does not match.

When a pipeline is chained inside another pipeline it is possible to run the pipeline without selecting a workspace-id. The workspace-id variable should be required

I want a private registry for our ORG and we should not be making connectors and secrets at the account level to access our org level material. this is not thought out well.