Feature Requests

Right now the rollback steps can only run on certain types of non-successes (namely Failures and I believe Expiries), which leaves some potential issues unkept/unattended when pipelines are aborted or otherwise. Ideally, we want to be able to optionally perform some kinds of rollback steps on any type of status of the pipeline other than success.

We are currently encountering difficulties utilizing Harness Delegate Connector across all projects. With over hunderds of GCP projects, managing a separate delegate connector for each project presents a significant challenge.

Background / Use Case: Currently, there is no expression available to retrieve the sequence ID of a previous pipeline run. This creates challenges in scenarios where artifacts are tied to the pipeline’s sequence ID. Example Use Case: The pipeline builds an artifact and names it using the expression build-<+pipeline.sequenceId>. The artifact is published to S3 with the key <+pipeline.sequenceId>. Subsequent steps depend on the existence of this S3 object using the same key. Problem: If one of the downstream steps fails and the pipeline is rerun, the new run gets an incremented sequenceId. As a result, the steps look for an artifact at <+pipeline.sequenceId + 1>, which does not exist, causing the rerun to fail. Request: Add support for detecting reruns and accessing the previous pipeline’s sequence ID (e.g., <+pipeline.previousSequenceId>). This would allow reruns to reuse artifacts or outputs from the original run instead of failing due to sequence ID mismatches.

Currently the only way to jump to an individual pipeline view form a pipeline chain is to scroll to the very right of the unfolded individual pipeline (which can take a long time!) and find a link in the top right of its rectangle. It's very slow. For some reason there is no link in the console view at all. I'd like to have a faster way to jump to a child pipeline view. Would be great to have links in the console view.

When a pipeline has a variable with a default value, any child triggers created from it will copy that value. If the default value in the parent pipeline changes later, the child triggers will still keep the old value. We’d like an option for child triggers to automatically stay in sync with the parent pipeline’s default variable values.

Currently the TAS services can be deployed using CF CLI 7.0. This feature request is to enable the support for TAS deployments using CF cli 8.0

We have a security requirement that needs us to utilize Bearer tokens instead of other authentication. Harness JFrog Connectors do not seem to support bearer authentication.

The new GCE Instance Flexibility feature will greatly help with Waze's reliability, especially for important services that are currently under capacity constraints (e.g. services requiring C2/C4 cores). From reading the documentation ( https://cloud.google.com/compute/docs/instance-groups/configure-instance-flexibility#terraform_1 ), a new instance_flexibility_policy field is introduced to the instance template, which Spinnaker doesn't know how to parse and configure. We will like to have Spinnaker supporting this feature.

Currently helm lint feature is not available out of the box and users have to rely on the hooks on the service to lint the helm chart that requires custom scripting. This request is to have helm lint support for helm services (Both native helm and K8s) in harness.

Currently the steps for helm deployments are limited to deploy/rollback/delete etc and the helm commands are included part of these steps. However, Having a separate step that has configurable helm commands would give more control and help leverage the helm features.