Compare commits

...

2 Commits

Author SHA1 Message Date
Mikhail Chusavitin
72cf482ad3 Embed Reanimator Chart web viewer 2026-03-15 22:07:42 +03:00
Mikhail Chusavitin
a6023372b1 Use microcode as CPU firmware 2026-03-15 21:16:17 +03:00
20 changed files with 304 additions and 26 deletions

3
.gitmodules vendored
View File

@@ -1,3 +1,6 @@
[submodule "bible"]
path = bible
url = https://git.mchus.pro/mchus/bible.git
[submodule "internal/chart"]
path = internal/chart
url = https://git.mchus.pro/reanimator/chart.git

View File

@@ -12,6 +12,7 @@ import (
"bee/audit/internal/platform"
"bee/audit/internal/runtimeenv"
"bee/audit/internal/tui"
"bee/audit/internal/webui"
)
var Version = "dev"
@@ -43,6 +44,8 @@ func run(args []string, stdout, stderr io.Writer) int {
return runTUI(args[1:], stdout, stderr)
case "export":
return runExport(args[1:], stdout, stderr)
case "web":
return runWeb(args[1:], stdout, stderr)
case "sat":
return runSAT(args[1:], stdout, stderr)
case "version", "--version", "-version":
@@ -60,6 +63,7 @@ func printRootUsage(w io.Writer) {
bee audit --runtime auto|local|livecd --output stdout|file:<path>
bee tui --runtime auto|local|livecd
bee export --target <device>
bee web --listen :80 --audit-path /var/log/bee-audit.json
bee sat nvidia|memory|storage
bee version
bee help [command]`)
@@ -73,6 +77,8 @@ func runHelp(args []string, stdout, stderr io.Writer) int {
return runTUI([]string{"--help"}, stdout, stdout)
case "export":
return runExport([]string{"--help"}, stdout, stdout)
case "web":
return runWeb([]string{"--help"}, stdout, stdout)
case "sat":
return runSAT([]string{"--help"}, stdout, stderr)
case "version":
@@ -213,6 +219,38 @@ func runExport(args []string, stdout, stderr io.Writer) int {
return 1
}
func runWeb(args []string, stdout, stderr io.Writer) int {
fs := flag.NewFlagSet("web", flag.ContinueOnError)
fs.SetOutput(stderr)
listenAddr := fs.String("listen", ":8080", "listen address, e.g. :80")
auditPath := fs.String("audit-path", app.DefaultAuditJSONPath, "path to the latest audit JSON snapshot")
title := fs.String("title", "Bee Hardware Audit", "page title")
fs.Usage = func() {
fmt.Fprintln(stderr, "usage: bee web [--listen :80] [--audit-path /var/log/bee-audit.json] [--title \"Bee Hardware Audit\"]")
fs.PrintDefaults()
}
if err := fs.Parse(args); err != nil {
if err == flag.ErrHelp {
return 0
}
return 2
}
if fs.NArg() != 0 {
fs.Usage()
return 2
}
slog.Info("starting bee web", "listen", *listenAddr, "audit_path", *auditPath)
if err := webui.ListenAndServe(*listenAddr, webui.HandlerOptions{
Title: *title,
AuditPath: *auditPath,
}); err != nil {
slog.Error("run web", "err", err)
return 1
}
return 0
}
func runSAT(args []string, stdout, stderr io.Writer) int {
if len(args) == 0 {
fmt.Fprintln(stderr, "usage: bee sat nvidia|memory|storage")

View File

@@ -1,8 +1,11 @@
module bee/audit
go 1.23
go 1.24.0
replace reanimator/chart => ../internal/chart
require github.com/charmbracelet/bubbletea v1.3.4
require reanimator/chart v0.0.0
require (
github.com/aymanbagabas/go-osc52/v2 v2.0.1 // indirect

View File

@@ -8,6 +8,14 @@ import (
"strings"
)
var execDmidecode = func(typeNum string) (string, error) {
out, err := exec.Command("dmidecode", "-t", typeNum).Output()
if err != nil {
return "", err
}
return string(out), nil
}
// collectBoard runs dmidecode for types 0, 1, 2 and returns the board record
// plus the BIOS firmware entry. Any failure is logged and returns zero values.
func collectBoard() (schema.HardwareBoard, []schema.HardwareFirmwareRecord) {
@@ -141,9 +149,5 @@ func cleanDMIValue(v string) string {
// runDmidecode executes dmidecode -t <typeNum> and returns its stdout.
func runDmidecode(typeNum string) (string, error) {
out, err := exec.Command("dmidecode", "-t", typeNum).Output()
if err != nil {
return "", err
}
return string(out), nil
return execDmidecode(typeNum)
}

View File

@@ -24,9 +24,7 @@ func Run(_ runtimeenv.Mode) schema.HardwareIngestRequest {
snap.Board = board
snap.Firmware = append(snap.Firmware, biosFW...)
cpus, cpuFW := collectCPUs(snap.Board.SerialNumber)
snap.CPUs = cpus
snap.Firmware = append(snap.Firmware, cpuFW...)
snap.CPUs = collectCPUs(snap.Board.SerialNumber)
snap.Memory = collectMemory()
sensorDoc, err := readSensorsJSONDoc()

View File

@@ -6,30 +6,28 @@ import (
"fmt"
"log/slog"
"os"
"path/filepath"
"strconv"
"strings"
)
// collectCPUs runs dmidecode -t 4 and reads microcode version from sysfs.
func collectCPUs(boardSerial string) ([]schema.HardwareCPU, []schema.HardwareFirmwareRecord) {
// collectCPUs runs dmidecode -t 4 and enriches CPUs with microcode from sysfs.
func collectCPUs(boardSerial string) []schema.HardwareCPU {
out, err := runDmidecode("4")
if err != nil {
slog.Warn("cpu: dmidecode type 4 failed", "err", err)
return nil, nil
return nil
}
cpus := parseCPUs(out, boardSerial)
var firmware []schema.HardwareFirmwareRecord
if mc := readMicrocode(); mc != "" {
firmware = append(firmware, schema.HardwareFirmwareRecord{
DeviceName: "CPU Microcode",
Version: mc,
})
for i := range cpus {
cpus[i].Firmware = &mc
}
}
slog.Info("cpu: collected", "count", len(cpus))
return cpus, firmware
return cpus
}
// parseCPUs splits dmidecode output into per-processor sections and parses each.
@@ -180,7 +178,7 @@ func parseInt(v string) int {
// readMicrocode reads the CPU microcode revision from sysfs.
// Returns empty string if unavailable.
func readMicrocode() string {
data, err := os.ReadFile("/sys/devices/system/cpu/cpu0/microcode/version")
data, err := os.ReadFile(filepath.Join(cpuSysBaseDir, "cpu0", "microcode", "version"))
if err != nil {
return ""
}

View File

@@ -1,6 +1,8 @@
package collector
import (
"os"
"path/filepath"
"testing"
)
@@ -63,6 +65,39 @@ func TestParseCPUs_unpopulated_skipped(t *testing.T) {
}
}
func TestCollectCPUsSetsFirmwareFromMicrocode(t *testing.T) {
tmp := t.TempDir()
origBase := cpuSysBaseDir
cpuSysBaseDir = tmp
t.Cleanup(func() { cpuSysBaseDir = origBase })
if err := os.MkdirAll(filepath.Join(tmp, "cpu0", "microcode"), 0755); err != nil {
t.Fatalf("mkdir microcode dir: %v", err)
}
if err := os.WriteFile(filepath.Join(tmp, "cpu0", "microcode", "version"), []byte("0x2b000643\n"), 0644); err != nil {
t.Fatalf("write microcode version: %v", err)
}
origRun := execDmidecode
execDmidecode = func(typeNum string) (string, error) {
if typeNum != "4" {
t.Fatalf("unexpected dmidecode type: %s", typeNum)
}
return mustReadFile(t, "testdata/dmidecode_type4.txt"), nil
}
t.Cleanup(func() { execDmidecode = origRun })
cpus := collectCPUs("CAR315KA0803B90")
if len(cpus) != 2 {
t.Fatalf("expected 2 CPUs, got %d", len(cpus))
}
for i, cpu := range cpus {
if cpu.Firmware == nil || *cpu.Firmware != "0x2b000643" {
t.Fatalf("cpu[%d] firmware=%v want microcode", i, cpu.Firmware)
}
}
}
func TestParseCPUStatus(t *testing.T) {
tests := []struct {
input string

View File

@@ -0,0 +1,77 @@
package webui
import (
"errors"
"fmt"
"net/http"
"os"
"strings"
"reanimator/chart/viewer"
chartweb "reanimator/chart/web"
)
const defaultTitle = "Bee Hardware Audit"
type HandlerOptions struct {
Title string
AuditPath string
}
func NewHandler(opts HandlerOptions) http.Handler {
title := strings.TrimSpace(opts.Title)
if title == "" {
title = defaultTitle
}
auditPath := strings.TrimSpace(opts.AuditPath)
mux := http.NewServeMux()
mux.Handle("GET /static/", http.StripPrefix("/static/", chartweb.Static()))
mux.HandleFunc("GET /healthz", func(w http.ResponseWriter, r *http.Request) {
w.Header().Set("Cache-Control", "no-store")
w.WriteHeader(http.StatusOK)
_, _ = w.Write([]byte("ok"))
})
mux.HandleFunc("GET /audit.json", func(w http.ResponseWriter, r *http.Request) {
data, err := loadSnapshot(auditPath)
if err != nil {
if errors.Is(err, os.ErrNotExist) {
http.Error(w, "audit snapshot not found", http.StatusNotFound)
return
}
http.Error(w, fmt.Sprintf("read audit snapshot: %v", err), http.StatusInternalServerError)
return
}
w.Header().Set("Cache-Control", "no-store")
w.Header().Set("Content-Type", "application/json; charset=utf-8")
_, _ = w.Write(data)
})
mux.HandleFunc("GET /", func(w http.ResponseWriter, r *http.Request) {
snapshot, err := loadSnapshot(auditPath)
if err != nil && !errors.Is(err, os.ErrNotExist) {
http.Error(w, fmt.Sprintf("read audit snapshot: %v", err), http.StatusInternalServerError)
return
}
html, err := viewer.RenderHTML(snapshot, title)
if err != nil {
http.Error(w, fmt.Sprintf("render snapshot: %v", err), http.StatusInternalServerError)
return
}
w.Header().Set("Cache-Control", "no-store")
w.Header().Set("Content-Type", "text/html; charset=utf-8")
_, _ = w.Write(html)
})
return mux
}
func ListenAndServe(addr string, opts HandlerOptions) error {
return http.ListenAndServe(addr, NewHandler(opts))
}
func loadSnapshot(path string) ([]byte, error) {
if strings.TrimSpace(path) == "" {
return nil, os.ErrNotExist
}
return os.ReadFile(path)
}

View File

@@ -0,0 +1,82 @@
package webui
import (
"net/http"
"net/http/httptest"
"os"
"path/filepath"
"strings"
"testing"
)
func TestRootRendersLatestSnapshot(t *testing.T) {
dir := t.TempDir()
path := filepath.Join(dir, "audit.json")
if err := os.WriteFile(path, []byte(`{"collected_at":"2026-03-15T00:00:00Z","hardware":{"board":{"serial_number":"SERIAL-OLD"}}}`), 0644); err != nil {
t.Fatal(err)
}
handler := NewHandler(HandlerOptions{
Title: "Bee Hardware Audit",
AuditPath: path,
})
first := httptest.NewRecorder()
handler.ServeHTTP(first, httptest.NewRequest(http.MethodGet, "/", nil))
if first.Code != http.StatusOK {
t.Fatalf("first status=%d", first.Code)
}
if !strings.Contains(first.Body.String(), "SERIAL-OLD") {
t.Fatalf("first body missing old serial: %s", first.Body.String())
}
if got := first.Header().Get("Cache-Control"); got != "no-store" {
t.Fatalf("first cache-control=%q", got)
}
if err := os.WriteFile(path, []byte(`{"collected_at":"2026-03-15T00:05:00Z","hardware":{"board":{"serial_number":"SERIAL-NEW"}}}`), 0644); err != nil {
t.Fatal(err)
}
second := httptest.NewRecorder()
handler.ServeHTTP(second, httptest.NewRequest(http.MethodGet, "/", nil))
if second.Code != http.StatusOK {
t.Fatalf("second status=%d", second.Code)
}
if !strings.Contains(second.Body.String(), "SERIAL-NEW") {
t.Fatalf("second body missing new serial: %s", second.Body.String())
}
if strings.Contains(second.Body.String(), "SERIAL-OLD") {
t.Fatalf("second body still contains old serial: %s", second.Body.String())
}
}
func TestAuditJSONServesLatestSnapshot(t *testing.T) {
dir := t.TempDir()
path := filepath.Join(dir, "audit.json")
body := `{"hardware":{"board":{"serial_number":"SERIAL-API"}}}`
if err := os.WriteFile(path, []byte(body), 0644); err != nil {
t.Fatal(err)
}
handler := NewHandler(HandlerOptions{AuditPath: path})
rec := httptest.NewRecorder()
handler.ServeHTTP(rec, httptest.NewRequest(http.MethodGet, "/audit.json", nil))
if rec.Code != http.StatusOK {
t.Fatalf("status=%d", rec.Code)
}
if got := strings.TrimSpace(rec.Body.String()); got != body {
t.Fatalf("body=%q want %q", got, body)
}
if got := rec.Header().Get("Content-Type"); !strings.Contains(got, "application/json") {
t.Fatalf("content-type=%q", got)
}
}
func TestMissingAuditJSONReturnsNotFound(t *testing.T) {
handler := NewHandler(HandlerOptions{AuditPath: "/missing/audit.json"})
rec := httptest.NewRecorder()
handler.ServeHTTP(rec, httptest.NewRequest(http.MethodGet, "/audit.json", nil))
if rec.Code != http.StatusNotFound {
t.Fatalf("status=%d want %d", rec.Code, http.StatusNotFound)
}
}

View File

@@ -18,8 +18,10 @@ local-fs.target
├── bee-network.service (starts `dhclient -nw` on all physical interfaces, non-blocking)
├── bee-nvidia.service (insmod nvidia*.ko from /usr/local/lib/nvidia/,
│ creates /dev/nvidia* nodes)
── bee-audit.service (runs `bee audit` → /var/log/bee-audit.json,
never blocks boot on partial collector failures)
── bee-audit.service (runs `bee audit` → /var/log/bee-audit.json,
never blocks boot on partial collector failures)
└── bee-web.service (runs `bee web` on :80,
reads the latest audit snapshot on each request)
```
**Critical invariants:**
@@ -29,6 +31,7 @@ local-fs.target
Reason: the modules are shipped in the ISO overlay under `/usr/local/lib/nvidia/`, not in the host module tree.
- `bee-audit.service` does not wait for `network-online.target`; audit is local and must run even if DHCP is broken.
- `bee-audit.service` logs audit failures but does not turn partial collector problems into a boot blocker.
- `bee-web.service` binds `0.0.0.0:80` and always renders the current `/var/log/bee-audit.json` contents.
- Audit JSON now includes a `hardware.summary` block with overall verdict and warning/failure counts.
## Console and login flow

View File

@@ -26,6 +26,7 @@ Fills gaps where Redfish/logpile is blind:
- NVIDIA proprietary driver loaded at boot for GPU enrichment via `nvidia-smi`
- SSH access (OpenSSH) always available for inspection and debugging
- Interactive Go TUI via `bee tui` for network setup, service management, and acceptance tests
- Read-only web viewer via `bee web`, rendering the latest audit snapshot through the embedded Reanimator Chart
- Local `tty1` operator UX: `bee` autologin, `menu` auto-start, privileged actions via `sudo -n`
## Network isolation — CRITICAL
@@ -95,6 +96,7 @@ Fills gaps where Redfish/logpile is blind:
| `iso/builder/` | ISO build scripts and `live-build` profile |
| `iso/overlay/` | Source overlay copied into a staged build overlay |
| `iso/vendor/` | Optional pre-built vendor binaries (storcli64, sas2ircu, sas3ircu, arcconf, ssacli, …) |
| `internal/chart/` | Git submodule with `reanimator/chart`, embedded into `bee web` |
| `iso/builder/VERSIONS` | Pinned versions: Debian, Go, NVIDIA driver, kernel ABI |
| `iso/builder/smoketest.sh` | Post-boot smoke test — run via SSH to verify live ISO |
| `iso/overlay/etc/profile.d/bee.sh` | `menu` helper + tty1 auto-start policy |

1
internal/chart Submodule

Submodule internal/chart added at 05db6994d4

View File

@@ -1,6 +1,6 @@
FROM debian:12
ARG GO_VERSION=1.23.6
ARG GO_VERSION=1.24.0
ARG DEBIAN_KERNEL_ABI=6.1.0-43
ENV DEBIAN_FRONTEND=noninteractive

View File

@@ -1,5 +1,5 @@
DEBIAN_VERSION=12
DEBIAN_KERNEL_ABI=6.1.0-43
NVIDIA_DRIVER_VERSION=590.48.01
GO_VERSION=1.23.6
GO_VERSION=1.24.0
AUDIT_VERSION=0.1.1

View File

@@ -39,6 +39,9 @@ echo "=== bee ISO build ==="
echo "Debian: ${DEBIAN_VERSION}, Kernel ABI: ${DEBIAN_KERNEL_ABI}, Go: ${GO_VERSION}"
echo ""
echo "=== syncing git submodules ==="
git -C "${REPO_ROOT}" submodule update --init --recursive
# --- compile bee binary (static, Linux amd64) ---
BEE_BIN="${DIST_DIR}/bee-linux-amd64"
GPU_STRESS_BIN="${DIST_DIR}/bee-gpu-stress-linux-amd64"

View File

@@ -9,6 +9,7 @@ echo "=== bee chroot setup ==="
systemctl enable bee-network.service
systemctl enable bee-nvidia.service
systemctl enable bee-audit.service
systemctl enable bee-web.service
systemctl enable bee-sshsetup.service
systemctl enable ssh.service
systemctl enable qemu-guest-agent.service 2>/dev/null || true

View File

@@ -12,7 +12,7 @@
set -e
. "$(dirname "$0")/VERSIONS" 2>/dev/null || true
GO_VERSION="${GO_VERSION:-1.23.6}"
GO_VERSION="${GO_VERSION:-1.24.0}"
DEBIAN_VERSION="${DEBIAN_VERSION:-12}"
DEBIAN_KERNEL_ABI="${DEBIAN_KERNEL_ABI:-6.1.0-28}"

View File

@@ -96,7 +96,7 @@ done
echo ""
echo "-- systemd services --"
for svc in bee-nvidia bee-network bee-audit; do
for svc in bee-nvidia bee-network bee-audit bee-web; do
if systemctl is-active --quiet "$svc" 2>/dev/null; then
ok "service active: $svc"
else
@@ -151,6 +151,20 @@ else
warn "audit: no log found at /var/log/bee-audit.log"
fi
echo ""
echo "-- bee web --"
if [ -f /var/log/bee-web.log ]; then
info "last web log line: $(tail -1 /var/log/bee-web.log)"
else
warn "web: no log found at /var/log/bee-web.log"
fi
if bash -c 'exec 3<>/dev/tcp/127.0.0.1/80 && printf "GET /healthz HTTP/1.0\r\nHost: localhost\r\n\r\n" >&3 && grep -q "^ok$" <&3'; then
ok "web: health endpoint reachable on 127.0.0.1:80"
else
fail "web: health endpoint not reachable on 127.0.0.1:80"
fi
echo ""
echo "-- network --"
if ip route show default 2>/dev/null | grep -q "default"; then

View File

@@ -1,6 +1,7 @@
[Unit]
Description=Bee: run hardware audit
After=bee-network.service bee-nvidia.service
Before=bee-web.service
[Service]
Type=oneshot

View File

@@ -0,0 +1,15 @@
[Unit]
Description=Bee: hardware audit web viewer
After=bee-network.service bee-audit.service
Wants=bee-audit.service
[Service]
Type=simple
ExecStart=/usr/local/bin/bee web --listen :80 --audit-path /var/log/bee-audit.json --title "Bee Hardware Audit"
Restart=always
RestartSec=2
StandardOutput=append:/var/log/bee-web.log
StandardError=append:/var/log/bee-web.log
[Install]
WantedBy=multi-user.target