fix(helm): make health path take root path into account.

[pantierra]

Fixes Bug 2 reported in developmentseed/eoapi-k8s#568

Makes the probes to automatically use ROOT_PATH (if set) before the HEALTH_PREFIX.

[claude]

Claude finished @pantierra's task in 2m 57s —— View job

---

Review of PR #189

• Gather context (diff, changed files, related app source)
• Review for correctness/security/best-practice issues
• Apply ponytail-review for over-engineering
• Post final review

---

✅ No blocking issues — safe to merge.

Traced the fix end-to-end against RemoveRootPathMiddleware (src/stac_auth_proxy/middleware/RemoveRootPathMiddleware.py:32): when ROOT_PATH is set, incoming requests must already include that prefix (or the middleware 404s), so probes do need {ROOT_PATH}{HEALTHZ_PREFIX} as this PR implements. The stac-auth-proxy.healthzPath/stac-auth-proxy.probe helpers (helm/templates/_helpers.tpl:97-113) correctly: trim the trailing slash from ROOT_PATH, no-op when httpGet.path is already set explicitly, and no-op when a probe has no httpGet (e.g. exec/tcpSocket probes) — the deepCopy before set avoids mutating the shared .Values map. New helm/tests/deployment_test.yaml cases cover default, ROOT_PATH-only, ROOT_PATH + custom HEALTHZ_PREFIX, and explicit-override — matches the implementation.

One non-blocking gap, same root cause the PR is fixing:

• helm/templates/NOTES.txt:37,50,52,57 — the post-install "Health check" output still prints {{ .Values.env.HEALTHZ_PREFIX | default "/healthz" }} without ROOT_PATH. When ROOT_PATH is set, the curl command Helm prints after install will 404, since it needs the same {ROOT_PATH}{HEALTHZ_PREFIX} this PR derives for the probes. Consider reusing include "stac-auth-proxy.healthzPath" . there too, e.g.:

  curl http://{{ .Values.ingress.host }}{{ include "stac-auth-proxy.healthzPath" . }}
  

No security issues found (chart doesn't touch auth/secrets), and no correctness issues in deployment.yaml/values.schema.json/values.yaml changes.

Simplify (ponytail): nothing to flag — the two small helpers are each used 3x (startup/liveness/readiness probes), so the factoring pays for itself; no speculative abstraction or unneeded dependency introduced.

---

💰 Estimated review cost: $1.01 · 2m57s · 37 turns

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 89.49%. Comparing base (fec33b6) to head (f8782a6).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #189   +/-   ##
=======================================
  Coverage   89.49%   89.49%           
=======================================
  Files          29       29           
  Lines        1276     1276           
  Branches      171      171           
=======================================
  Hits         1142     1142           
  Misses         92       92           
  Partials       42       42           

Flag	Coverage Δ
unittests	89.49% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[alukach]

As usual, I barely understand this stuff. I'll leave it to your discretion to consider Claude's input from the above comment.