Add Support for Customizing Deployment maxSurge and maxUnavailable in Helm Chart

[yaguirre]

Description

Currently, the Harness Delegate Helm chart does not allow users to customize the Deployment strategy.rollingUpdate.maxSurge and maxUnavailable parameters. In the deployment.yaml template, these values are hardcoded, and there are no chart values to override them.

This limitation can cause rollout issues in real-world scenarios, especially when using Horizontal Pod Autoscaler (HPA) with a small minimum replica count. For example, the default values can result in failed rollouts or reduced availability during upgrades, as the rolling update strategy cannot be tuned to fit specific production needs.

Request

Please add support for configuring both maxSurge and maxUnavailable via the chart's values.yaml file, so users can adjust these settings as needed. For example:

strategy:
  type: RollingUpdate
  rollingUpdate:
    maxSurge: 1
    maxUnavailable: 1

Why This Is Important

• Allows teams to avoid rollout failures in deployments with HPAs or low replica counts.
• Enables safe zero-downtime upgrades by customizing the rolling update behavior to match real-world workloads.
• Brings the chart in line with Kubernetes best practices and other major Helm charts, which typically expose these parameters.

Example Use Case:

In our case, we needed to set both maxSurge and maxUnavailable to 1 to prevent rollout failures when running with only 2 replicas and HPA enabled. Not being able to configure these values from the chart forced us to edit the deployment manifest manually, which is not ideal for automated or repeatable deployments.