Skip to content

TRAILHEAD

TRAILHEAD is the AzureLocal live-cluster validation framework. Unlike MAPROOM (which is fixture-based and runs in CI), TRAILHEAD runs structured cycles against real target clusters and captures timestamped evidence.

Start here

Read When
Overview What TRAILHEAD is, how it differs from MAPROOM, when to use it
Cycle planning Plan a cycle — scenario, cluster, phase, operator
Evidence capture How logs, timings, and outputs are recorded
Safety checklist Pre-flight before running against a real cluster
Post-run review After-action: comparing runs, filing issues, regressions

Key facts

  • Harness: testing/trailhead/scripts/ — platform-owned helpers
  • Scripts: Start-TrailheadRun.ps1, TrailheadLog-Helpers.ps1
  • Evidence location: consumer repo's tests/trailhead/logs/<date>-<scenario>/
  • Authority: contract — platform owns the harness; consumers supply scenario scripts
  • Typical cadence: pre-release smoke, pre-release full, regression after upgrade, post-upgrade HCI

Relationship to MAPROOM

MAPROOM TRAILHEAD
Input JSON fixture Live cluster
Runs in CI Operator workstation / bastion
Output Pester pass/fail Evidence artifacts + pass/fail
Duration Seconds Minutes to hours
Frequency Every PR Scheduled cycles
Classification (ADR-0004) Scope: infra-fabric, Authority: canonical Scope: workload, Authority: contract

Current consumers

All PS module and IaC solution repos that deploy or validate against a cluster:

  • azurelocal-s2d-cartographer
  • azurelocal-ranger
  • Additional consumers onboard as their deploy paths mature

Next steps