It came, it went, it conquered! We’re talking about {unscripted}, of course - what an amazing 2 days packed with knowledge, great speakers, insightful panels, and industry-shaking announcements! In case you missed it, we have some helpful recaps for Day 1 and Day 2 that are sure to get some excitement bubbling up! We also have a handy dandy YouTube playlist with all sessions available to watch. Read on below to learn more about what we've been up to in our June 2021 product update!
New Features
Harness CI Enterprise
Harness has entered the CI space with the addition of the CI module to its platform. Harness took the best concepts from its open-source counterpart, Drone, and added in some new features to enhance the developer experience by making it easier for developers to build and test code. New features include containerized steps, visual pipeline builder, YAML code editor, GitOps functionality, fine-grained RBAC, and for the first time ever in a CI solution: test intelligence. Test Intelligence uses ML to select tests that are required to confirm the quality of code changes that triggered the build, and rank tests in an optimal order to increase the rate of fault detection. Learn more about Test Intelligence and CIE’s many benefits in our blog post.
Harness Feature Flags
Harness now has a product module in the feature flags space with the introduction of Harness Feature Flags. Feature flags empower engineering teams to deliver features faster, with less risk. Built for developers, it features a streamlined developer experience that includes a visual pipeline builder and automated feature verification. Additional features include: boolean and multivariate flags; flexible targeting across users, groups, regions, and infrastructure; streaming and polling architecture support; RBAC; and audit trails. Harness Feature Flags is uniquely designed in the feature flags space to be easy to get started with and get your first flag out the door quickly. Find out more in our Feature Flags post.
Harness CCM - Cloud AutoStopping
Harness now supports active cloud cost management with its latest addition of Cloud AutoStopping, which reduces cloud compute waste associated with non-production environments by up to 75%. Harness can proactively manage your cloud compute resources so that they’re automatically turned off when idle, and turned back on when traffic is detected to the resource. It also features full orchestration on spot instances - it’s “set it and forget it” at its best. Find out more in our Cloud AutoStopping post.
Harness CD - Supporting Secrets with Github Webhook triggers
This allows users to authenticate incoming GitHub webhooks using secrets to ensure the webhooks are coming from a trusted source. Even if an unintended user gets access to the webhook URL for the harness trigger, the user will not be able to set it up in Github since it will require the secret value as well. This improves the security aspect of triggers.
Harness CD - Granular Access Control for Workflow Deployments
Harness has added a new permission to rollback workflows. This allows users to roll back a service to the last successfully deployed version or initiate a rollback for a running workflow, even if they don't have the permission to execute the workflows.
This is totally not a virtual event, but you should come watch Black Widow with us on July 8th! We’re hosting a private, pre-release event in the US - all you need to do is register and we’ll send you two tickets - maybe more if you’re really nice about it! We have 11 locations in the US. Register now!
It’s hard not to want to keep looking back at June because we released so much new stuff already… but you know what they say about bees: they’re always busy. In the next few weeks/months, here’s what’s coming at ya:
Publish Events for Deployments - Harness will be able to publish the events for pipeline deployments on a webhook URL. Users will be able to set up dashboards on the events to run analysis on their deployments.
Approve/Reject Harness Approvals via API - Harness users will be able to approve or reject the approvals via API. This will enable them to achieve the deployments containing Harness Approvals via scripts and there will be no need to navigate through Harness UI.
Deployment Freeze Window Enhancements - Harness will now allow users to set up recurring freeze windows on daily, weekly, monthly etc. basis. Also, users will be able to set up freeze windows on Services. With this, there will be a new permission to allow deployments during freeze.
Capabilities Support With AWS CloudFormation - Users will now be able to define and use the “--capabilities” functionality that AWS CloudFormation offers when provisioning a new stack. Capabilities allow for the user to define what this stack can do, and it's a form of acknowledgement that the stack will be creating and manipulating sensitive resources within a users environment.
Pre-Processing and Fetching Helm Charts From Any Repository Source - Users will be able to define a method on how Harness is to fetch the user’s Helm Charts and Values yaml for deployment. A common use case with this feature is that there are secrets that need to be generated or fetched from a specific source, or organizations have custom yaml templating mechanisms that they wish to leverage with Harness. These Helm charts can be fetched from any source as long as the delegate can communicate with the source.
Exporting the Deployment.yaml as a Variable - Users will be able to take the manifest-dry-run.yaml that Harness generates during deployment and pass that file to a third-party system. This allows users to run policies or checks against their yaml to make sure the manifest configuration meets organization standards.
Helm Chart as an Artifact Source (Experience Revamp) - To our users who have tried the new Helm Chart functionality, we would like to thank you! We loved hearing from you and collecting your feedback. We are working on making the experience even better by providing a more intuitive user interface for users to configure their Helm Charts for Kubernetes and Helm services. The revamped experience includes API support for Helm and Kubernetes Deployments with the Helm Chart as the deployable unit. We will poll for new Helm Charts published to the users Helm repo and deploy it!
HashiCorp Vault Agent-Based Authentication for HashiCorp Vault - Users will be able to use the HashiCorp Vault Agent (in addition to App Role and Token) to authenticate with HashiCorp Vault. Vault Agent-based authentication eliminates the need for users to store their HashiCorp Vault credentials within Harness. Further, Vault token renewal can also be managed using the Vault Agent instead of being managed within Harness.
Delegate Token Rotation - Users will be able to further secure the communication between their Delegates and Harness Manager by replacing their default Delegate tokens. Delegate tokens are used by Harness to encrypt communication between Harness Delegates and the Harness Manager. Users can now create custom delegate tokens and use these custom tokens with one or more of their delegates instead of using the default Delegate token. Further, users can revoke Delegate tokens per their governance policies and replace revoked tokens with new custom tokens.
Custom Remote Manifest Support for Native Helm - Multiple customers have requested the ability to pull the Helm Chart linked to a Harness Service prior to running a deployment that allows them to run some form of preprocessing before the deployment. This will be available shortly.
Support the Option to Deploy Without a Values File for Templatizing - Harness users will be able to use their existing Helm manifest files which contain their own custom go-templating and be able to do Kubernetes deployments.
Handle CloudFormation Statuses Gracefully - Harness will now be able to handle CloudFormation statuses gracefully. This will help with deployments and with rollbacks.