F11 matrix-correction: re-arm on guest-blind (active mounts not inherited by rebooted guest — live finding)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017CDMFpFx84pfviCTVuGGhf
2026-07-12 07:58:39 +02:00
parent 47eb0bf967
commit b6300250aa
5 changed files with 90 additions and 10 deletions
@@ -156,6 +156,17 @@ func (h *SudoHostOps) resetNetworkAutomountIfFailed(ctx context.Context, where s
 	return reset
 }

+// RearmNetworkAutomount is the caller-driven remediation (CAMPAIGN-3 F11, matrix-corrected): when a
+// running guest does NOT see a share after the reassert pass, re-arm its trigger regardless of the
+// host's mount state. LIVE FINDING (matrix row 1/2, 2026-07-12): a freshly-rebooted guest does NOT
+// inherit an ACTIVE host mount either — only a fresh trigger-mount event propagates into the new
+// namespace — so skip-active is safe to skip PROACTIVELY (don't churn shares guests already see) but
+// must be remediated REACTIVELY when a specific guest is blind. reset-failed first (F10), then re-arm.
+func (h *SudoHostOps) RearmNetworkAutomount(ctx context.Context, where string) error {
+	h.resetNetworkAutomountIfFailed(ctx, where)
+	return h.rearmNetworkAutomount(ctx, where)
+}
+
 // rearmNetworkAutomount stops then re-enables+starts the .automount for a mountpoint. The stop is
 // tolerated failing (unit not loaded); the enable --now is the action that must succeed. Both verbs
 // are the existing FELHOM_NETMOUNT sudoers grants.