Feature Requests

Would like to create dashboards based on what infrastructure each Build stage is using. This way I can see how much each VM pool or K8s cluster is being used across the account, sliced by org, project and pipeline. Would help with usage metrics, capacity planning and other SRE decisions. For example: Project A used 200 minutes of K8S cluster Y and 50 minutes of VM pool Z for builds last week.

Right now, within Harness, it evaluates whether a deployment can proceed based on all existing freeze windows configured and the exceptions in them. We want Harness to make it easier to get a boolean response on whether a deployment can proceed. Inputs for the API would be service_name and environment,error-prone custom template, and Output as a boolean value of true/false. Currently, we've had to resort to a custom template with complex logic to determine deployment feasibility. If Harness could provide an API to simplify this process, it would significantly enhance our workflow. A relevant Support Request raised on a similar topic is https://support.harness.io/hc/en-us/requests/69139?page=1

Git supports signing commits with SSH or GPG keys. This can be used to block pushes or it can display which commits are signed for non-repupdiation.

How can parameters of gitops update app step be used as runtime inputs to a pipeline as we don't see any runtime inputs when we click "Run" on the pipeline ?

Harness RBAC should allow restricting access to the Events and Logs within the Resource View. The underlying ArgoCD allows this via their RBAC, but it'd be nice if we could restrict access via Harness roles so we could get finer grained control: Some operators only need access to see the health of a resource, not the logs.

We use the plugin kaniko-ecr to build and push images to ECR. We would like to know the imageDigest of the image that was built and pushed. The value we need is already being written to a file within the plugin execution by using the --digest-file flag on the kaniko command. Sending that back to us as an output variable would be fantastic. Without it, we have to make a network request to describe-image based on the tag to get info about the image it just pushed.

Hi Team, I understand that Different services have chosen different DBs . but in this case will casue the SMP setup more complex and addtional cost , we have to manage 4 DBs. From customer perspective , it doesn't make sense we suggest to reduce the db usage . we notice that Postgres is not use too much please merge it to mongo or move mongo to PG and remove timescale db

Hi Team, We request a feature - To not throw warning to the user if only a stage is run which uses output expression from previous stage. Rather pass with blank or empty value. We can have it as a configurable option at project setting level.

Need a way to get output (variable) value from executed pipelines/stages/steps via API. Use case: Creating a Jira ticket inside a Harness pipeline triggered via webhook API, need to use Harness API to get the Jira ticket ID outside Harness.

When deleting resources from Harness, Harness Terraform Provider Plugin does not handle the removal of the resources gracefully and provides a segmentation fault error. There should be a validation of resources by the plugin so that it can handle the missing resources more gracefully. Submitted on behalf of LL Bean