2 Commits

Include renames

Author	SHA1	Message	Date
shankar0123	b8b7e1e3dd	tlsprobe: add VerifyWithExponentialBackoff + rewire all connectors' runPostDeployVerify ... Closes Top-10 fix #8 of the 2026-05-02 deployment-target audit re-run (see cowork/deployment-target-audit-2026-05-02-rerun/ RESULTS.md). Pre-fix, every connector's runPostDeployVerify used linear backoff (default 3 attempts × 2s linear waits). Linear backoff misbehaves under load-balanced rollouts: the verify probe hits a random LB-backed pod, and 3 × 2s often falls into the worst case where match-fingerprint pods stop responding by attempt 3 due to LB session-stickiness cycles. This commit: 1. New shared helper internal/tlsprobe/retry.go:: VerifyWithExponentialBackoff. Default 3 attempts; 1s initial, 16s cap. Doubling pattern: 1s → 2s → 4s → 8s → 16s. probe func(ctx) error signature so connectors compose handshake + fingerprint-compare into one lambda. 2. Each connector's runPostDeployVerify (nginx, apache, haproxy, traefik, envoy, postfix, dovecot) rewired to call the shared helper. Per-connector signature unchanged. 3. New PostDeployVerifyMaxBackoff time.Duration field added to each connector's Config. Operators preserving V2 linear behavior set PostDeployVerifyMaxBackoff equal to PostDeployVerifyBackoff. 4. Tests: - tlsprobe/retry_test.go: TestVerifyWithExponentialBackoff_ GrowthAndCap + TestVerifyWithExponentialBackoff_ StopsOnFirstSuccess + TestVerifyWithExponentialBackoff_ CtxCancellation. - One Test<Connector>_VerifyExponentialBackoff_ GrowsBetweenAttempts per connector (6 total across postfix, nginx, apache, haproxy; traefik and envoy connectors use unique test signatures so test wiring deferred to future unification). 5. docs/deployment-atomicity.md Section 4 updated: 'linear backoff' → 'exponential backoff (1s → 16s cap)'; YAML example shows the new field. Backward-compat note: PostDeployVerifyBackoff was interpreted as the linear interval pre-fix; post-fix it's interpreted as the initial backoff (which doubles each attempt). Operators using the default value (2s) see waits of 2s → 4s → 8s instead of 2s → 2s → 2s. For LB-rollout cases this is the intended behavior; for single-target deploys the wall-clock is slightly longer (12s vs 6s for 3 attempts). Operators preserving V2 linear semantics: set PostDeployVerifyMaxBackoff equal to PostDeployVerifyBackoff. Verified locally: - gofmt clean. - go test -short -count=1 ./internal/tlsprobe/... ./internal/connector/target/{postfix,nginx,apache,haproxy}/... green. Audit reference: cowork/deployment-target-audit-2026-05-02-rerun/ RESULTS.md Top-10 fix #8.	2026-05-02 22:56:07 +00:00
shankar0123	919a92bf1b	feat(haproxy): atomic deploy + post-deploy TLS verify + rollback + ValidateOnly + test-depth uplift to 36 tests ... Phase 6 of the deploy-hardening I master bundle. HAProxy connector follows the canonical Phase 4 NGINX template with the HAProxy- specific quirk: combined PEM file (cert + chain + key in one file, in that order). Test count lifts 3 → 36. HAProxy specifics: - buildCombinedPEM concatenates cert, chain, key in HAProxy's required order. The combined file goes through deploy.Apply as a single File entry (vs NGINX/Apache's 2-3 separate File entries). - Default mode 0600 unconditionally (combined file contains the private key); operators rely on this back-compat behavior. PEMFileMode override is the supported escape hatch. - Validate command is `haproxy -c -f <config>`. Reload via `systemctl reload haproxy` (NOT `restart` — reload uses socket activation to drain in-flight connections). - Default user/group: haproxy (cross-distro consistent). DeployCertificate refactor: - Replaces the duplicated os.WriteFile flow with deploy.Apply. - PreCommit runs `haproxy -c -f` validation (gated on ValidateCommand being non-empty — HAProxy historically allowed empty validate). - PostCommit runs the operator's ReloadCommand. - Post-deploy TLS verify (frozen-decision-0.3 default ON when Endpoint is configured): probes the configured target, fingerprint-matches against the deployed cert (the leaf cert block from the combined PEM), retries with backoff for load- balanced targets. - Rollback wires identical to NGINX/Apache: backup restore + reload retry on PostCommit failure; verify-fail also triggers rollback. ValidateOnly real impl: returns sentinel when no ValidateCommand; otherwise runs the operator's command without touching the live combined PEM. Tests (36 total: 33 in haproxy_atomic_test.go + 3 pre-existing in haproxy_test.go): - Atomic invariants (happy, validate-fail, reload-fail-rollback, rollback-also-fail-escalation) - Combined PEM order (cert + chain + key — verified via PEM block headers, not base64 bodies) - Mode handling (default 0600 even when existing is 0640 — back-compat; PEMFileMode override; existing-mode unchanged when override matches) - Idempotency (full skip) - Verify (match, mismatch, dial-timeout, retries, disabled, no-endpoint, rollback-runs-reload) - ValidateOnly (happy, fails, no-command-sentinel, stderr-in-error) - Concurrency (same-paths-serialize) - Edge cases (no-chain, no-key, ctx-cancelled, no-validate-command, config-validation rejects missing pem_path / reload / shell-injection) Coverage: HAProxy 88.0% (above >=85% prompt bar). Race detector clean. golangci-lint v2.11.4 clean. Smoke test connectorsAtPhase3 list shrinks 11→10 (haproxy removed alongside nginx + apache). Phase 7 next: Traefik + Caddy + Envoy + Postfix — the remaining file-based connectors get the same treatment.	2026-04-30 15:01:23 +00:00