v0.5.0: slice 5 Phase B — the host-root surface (mounts + SMART + grow + destructive gate)

The privileged write surface, isolated behind a narrow, arg-validated, adversarially-
tested seam (HostOps), the same discipline as the slice-4 gate. Completes slice 5.

- internal/storage: HostOps seam + SudoHostOps (systemd .mount units by fs-UUID, detach,
  SMART, lvs) via sudoers allowlist + fixed arg vectors, no shell; NoopHostOps fallback.
- validate.go: strict UUID/mount-path/device/LVM validators + in-process systemd-escape.
  Headline test: adversarial matrix (metacharacters/traversal/malformed) refused with
  zero exec.
- smart.go: smartctl SATA + NVMe parse, UNKNOWN-degrade; lvs thin-pool metadata fill.
- observer enrichment (Observe only): fills smart + thin-pool metadata.
- watchdog: benign re-mount response off the poll path (DevicePresent probe, rate-limited).
- reconcile: ActionResize (benign, grow-only) + proxmox.ResizeLXC; destructive storage ops
  (ClassStorageWipe/Decommission) through the slice-4 gate, target-scoped; built+tested,
  inert live.
- --selftest=storage [-watch] live harness; configs/felhom-agent.sudoers; privileged.* knobs.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-06-09 10:53:38 +02:00
parent 27b68f043b
commit 9d6e49236c
25 changed files with 2074 additions and 182 deletions
+3
View File
@@ -69,6 +69,7 @@ type ActualGuest struct {
SpecKnown bool
Cores int
MemoryMiB int64 // proxmox LXC `memory` is MiB
DiskBytes int64 // rootfs size in bytes (from the LXC list MaxDisk; for grow planning)
Description string // raw (may carry PVE's trailing newline; compared via normalizers)
}
@@ -111,6 +112,8 @@ type GuestAPI interface {
Start(ctx context.Context, vmid int) (string, error)
Stop(ctx context.Context, vmid int) (string, error)
SetConfig(ctx context.Context, vmid int, params map[string]string) (string, error)
// ResizeLXC grows a volume (grow-only; the planner never emits a shrink). Async → UPID.
ResizeLXC(ctx context.Context, vmid int, disk, size string) (string, error)
WaitTask(ctx context.Context, upid string, opts proxmox.WaitOptions) (proxmox.TaskStatus, error)
// TaskStatusOnce is a single non-blocking task-status read — used by crash
// recovery to learn the outcome of an op that was in flight when the agent died.