This page captures the most important chart values and runtime settings used in the current customer deployment model. It is not yet an exhaustive schema dump.
Core run settings
| Key | Type or example | Purpose |
|---|
config.projectId | test | Resolves the input directory name under the input root |
config.runId | run_20260307T010203Z | Forces outputs into a stable run-scoped directory |
config.runMode | standard or deep | Selects the execution profile and billing metric |
config.inputPath | /data/input | Root path used to discover <project_id>/01_sources/ |
config.outputPath | /data/output | Root path where run-scoped outputs are written |
Entitlement settings
| Key | Type or example | Purpose |
|---|
config.entitlementUrl | https://YOUR_CLOUD_RUN_SERVICE | Base URL for the entitlement and seal API |
config.entitlementAuthMode | google | Enables identity-based auth for the customer job |
config.entitlementAudience | same as entitlement URL | Audience used when minting the identity token |
config.entitlementPrincipal | local-dev | Local or test principal for non-Google development flows |
For the customer deployment path, prefer config.entitlementAuthMode=google.
Storage settings
| Key | Type or example | Purpose |
|---|
storage.type | pvc or gcs | Selects the backing storage mode |
storage.pvc.storageClassName | standard | Storage class for PVC mode |
storage.pvc.size | 50Gi | Requested PVC capacity |
storage.gcs.bucket | YOUR_BUCKET | Bucket used for GCS-backed storage mode |
Workload Identity settings
| Key | Type or example | Purpose |
|---|
workloadIdentity.enabled | true | Enables Workload Identity plumbing in the chart |
workloadIdentity.gcpServiceAccount | your-sa@project.iam.gserviceaccount.com | GSA used by the job when calling Google-protected services |
Marketplace and usage reporting settings
| Key | Type or example | Purpose |
|---|
marketplace.reportingSecret | Kubernetes Secret reference | Injects the Marketplace reporting credential |
ubbagent.metricName | standard_audit_run | Legacy hard override for usage metric selection |
ubbagent.metricNameStandard | standard_audit_run | Explicit Standard metric name |
ubbagent.metricNameDeep | deep_audit_run | Explicit Deep metric name |
If ubbagent.metricName is set, it acts as a hard override. Leave it empty unless you have a controlled release reason to override the mode-aware defaults.
Operational notes
config.projectId must match the staged input folder name exactly
config.runId must be unique per run to avoid 409 reuse conflicts
- Standard and Deep should be treated as separate operational profiles, not just cosmetic labels
- Do not change storage size downward on an existing PVC-backed release
See also Kubernetes and Helm.