Files
certctl/internal/connector/target/validate_only_smoke_test.go
T
shankar0123 7444df01e2 feat(nginx): atomic deploy + post-deploy TLS verify + rollback + ValidateOnly + ownership preservation
Phase 4 of the deploy-hardening I master bundle. The canonical NGINX
implementation that Phases 5-9 model on. Replaces the historical
os.WriteFile flow at internal/connector/target/nginx/nginx.go:99
with deploy.Apply() and adds three production-grade competitor-gap
features: atomic deploy with rollback, post-deploy TLS verify, file
ownership preservation.

NGINX connector — internal/connector/target/nginx/nginx.go:

- DeployCertificate now wires deploy.Apply with PreCommit running
  the operator's ValidateCommand (e.g. `nginx -t`), PostCommit
  running ReloadCommand (e.g. `nginx -s reload`), and an explicit
  post-deploy TLS verify step that dials the configured endpoint,
  pulls the leaf cert SHA-256, and compares against what was just
  deployed. SHA-256 mismatch (wrong vhost / cached cert / NGINX
  still serving stale) triggers automatic rollback: backup files
  are restored + reload fired again. Failed-second-reload returns
  ErrRollbackFailed (operator-actionable; loud audit + alert).

- ValidateOnly replaces the Phase 3 stub: runs the operator's
  ValidateCommand without touching the live cert. V2 contract is
  syntax-only validation (full pre-deploy temp-config validation
  is V3-Pro). Returns ErrValidateOnlyNotSupported when no
  ValidateCommand is configured.

- New per-target Config fields: PostDeployVerify (frozen-decision-
  0.3 default ON), PostDeployVerifyAttempts (default 3 — defends
  against load-balanced targets where the verify might hit a
  different pod that hasn't picked up the new cert yet),
  PostDeployVerifyBackoff (default 2s exponential), per-file
  Mode/Owner/Group overrides (KeyFileMode, CertFileMode,
  KeyFileOwner, etc.), and BackupRetention (default 3, -1 to
  disable backups entirely — documented foot-gun).

- buildPlan honors per-distro nginx user (Debian: www-data,
  Alpine: nginx, Red Hat: nginx) by checking the local user
  database; falls back to no-chown when neither exists. Means
  the connector is portable across distros without operator
  config.

Deploy package — internal/deploy/ownership.go:

- applyOwnership now silently swallows chown failures when the
  agent isn't running as root. Production agents always run as
  root and chown failures are real bugs; dev / CI runs as a
  regular user where chown to a different uid will always fail
  with EPERM (or EINVAL on some tmpfs configs) and would
  otherwise force every test to run with sudo. Production-grade
  contract preserved (uid 0 still hard-fails on chown errors).

Test suite — internal/connector/target/nginx/nginx_atomic_test.go
ships 42 new named tests (NGINX total: 17 pre-existing + 42 new = 59,
above the prompt's >=40 bar; matches the IIS depth bar of 41):

- Atomic-deploy invariants (cert+chain+key all-or-nothing,
  validate-fails-no-files-changed, reload-fails-rollback,
  rollback-also-fails-escalation)
- SHA-256 idempotency (full match skips, partial match deploys all)
- Post-deploy TLS verify (fingerprint-match-success,
  SHA256-mismatch-rollback, dial-timeout-rollback, retries-until-
  match, retries-exhausted-rollback, no-endpoint-skips,
  disabled-skips-entirely, default-10s-timeout, endpoint-forwarded)
- Ownership / mode preservation (existing-mode-preserved, override-
  wins, KeyFileMode override applied)
- Backup retention (keeps-last-N, disabled-creates-no-backups,
  fresh-deploy-creates-backup)
- Concurrency (same-paths-serialize via deploy package's file mutex,
  different-paths-parallelize)
- ValidateOnly (happy-path-nil, command-fails-wrapped-error,
  no-config-returns-sentinel, ctx-cancelled, stderr-in-message)
- Edge cases (no-chain, no-key, no-chain-path, empty-cert-PEM,
  ctx-cancelled, all-four-one-apply)
- Result.Metadata + DeploymentID shape contracts

Coverage: NGINX 91.0% (above the >=85% prompt bar). Race detector
clean. golangci-lint v2.11.4 clean. Existing 17 tests still all pass
(no behavior change in the legacy paths exercised there).

Phase 5 next: mirror this implementation for Apache + lift its
test count from 3 to >=30. Same template applies through Phases
6-9 for the remaining 11 connectors.
2026-04-30 14:50:56 +00:00

105 lines
4.9 KiB
Go

package target_test
// Phase 3 of the deploy-hardening I master bundle: per-connector
// regression smoke pinning the default ValidateOnly stub returns
// the sentinel for every one of the 13 connectors. This test lives
// in target_test (external test package) so it can import each
// connector concretely + assert the interface contract.
//
// As Phases 4-9 replace each connector's stub with a real
// validate-with-the-target implementation, the corresponding
// per-connector entry in TestEveryConnectorDefaultsToSentinel
// MUST be deleted (or the test will fail because the real
// implementation no longer returns the sentinel). That deletion
// IS the bookkeeping that the operator-visible bit + behavior
// change are wired together.
import (
"context"
"encoding/json"
"errors"
"testing"
"github.com/shankar0123/certctl/internal/connector/target"
"github.com/shankar0123/certctl/internal/connector/target/apache"
"github.com/shankar0123/certctl/internal/connector/target/caddy"
"github.com/shankar0123/certctl/internal/connector/target/envoy"
"github.com/shankar0123/certctl/internal/connector/target/f5"
"github.com/shankar0123/certctl/internal/connector/target/haproxy"
"github.com/shankar0123/certctl/internal/connector/target/iis"
"github.com/shankar0123/certctl/internal/connector/target/javakeystore"
"github.com/shankar0123/certctl/internal/connector/target/k8ssecret"
// nginx removed Phase 4 — real ValidateOnly implementation now in nginx.go.
"github.com/shankar0123/certctl/internal/connector/target/postfix"
"github.com/shankar0123/certctl/internal/connector/target/ssh"
"github.com/shankar0123/certctl/internal/connector/target/traefik"
"github.com/shankar0123/certctl/internal/connector/target/wincertstore"
)
// connectorsAtPhase3 is the canonical list of connectors that, as
// of Phase 3, return ErrValidateOnlyNotSupported from
// ValidateOnly. Each entry is a (name, factory) tuple; the factory
// returns a target.Connector via the connector's bare-NewConnector
// constructor pattern. As Phases 4-9 land, the corresponding
// connector is REMOVED from this list — its real ValidateOnly
// implementation is then exercised in the per-connector test
// suite, NOT here.
//
// CI guard rationale: a future PR that adds a 14th connector
// without wiring ValidateOnly fails this test (the sentinel
// contract is not satisfied). A future PR that implements a real
// ValidateOnly for, say, NGINX, but forgets to remove its entry
// from this list, fails this test (real impl no longer returns
// the sentinel). Both are the load-bearing bookkeeping protections.
var connectorsAtPhase3 = []struct {
name string
// new returns a fresh Connector instance. The default
// ValidateOnly stub doesn't dereference any field on the
// receiver, so a zero-value &pkg.Connector{} is sufficient
// to satisfy the interface and exercise the sentinel return.
// Phases 4-9 introduce real validate-with-the-target impls
// that DO read fields; those connectors will need a populated
// constructor here OR (more likely) be removed from this list
// entirely and exercised in their own per-connector test
// suite.
new func() target.Connector
}{
{"apache", func() target.Connector { return &apache.Connector{} }},
{"caddy", func() target.Connector { return &caddy.Connector{} }},
{"envoy", func() target.Connector { return &envoy.Connector{} }},
{"f5", func() target.Connector { return &f5.Connector{} }},
{"haproxy", func() target.Connector { return &haproxy.Connector{} }},
{"iis", func() target.Connector { return &iis.Connector{} }},
{"javakeystore", func() target.Connector { return &javakeystore.Connector{} }},
{"k8ssecret", func() target.Connector { return &k8ssecret.Connector{} }},
// nginx removed Phase 4 — its ValidateOnly is now the real
// implementation; tested directly in
// internal/connector/target/nginx/nginx_test.go.
{"postfix", func() target.Connector { return &postfix.Connector{} }},
{"ssh", func() target.Connector { return &ssh.Connector{} }},
{"traefik", func() target.Connector { return &traefik.Connector{} }},
{"wincertstore", func() target.Connector { return &wincertstore.Connector{} }},
}
func TestEveryConnectorDefaultsToSentinel(t *testing.T) {
// Expected list size shrinks as Phases 4-9 land their real
// ValidateOnly implementations. Phase 4 removed nginx.
const expectedAtCurrentPhase = 12
if len(connectorsAtPhase3) != expectedAtCurrentPhase {
t.Fatalf("connectors-at-phase list = %d entries, want %d (drift in the 13-connector inventory)", len(connectorsAtPhase3), expectedAtCurrentPhase)
}
for _, c := range connectorsAtPhase3 {
t.Run(c.name, func(t *testing.T) {
conn := c.new()
err := conn.ValidateOnly(context.Background(), target.DeploymentRequest{
CertPEM: "ignored-by-stub",
ChainPEM: "ignored",
TargetConfig: json.RawMessage(`{}`),
})
if !errors.Is(err, target.ErrValidateOnlyNotSupported) {
t.Errorf("got %v, want ErrValidateOnlyNotSupported", err)
}
})
}
}