fix(sat): fan check uses hottest GPU load + IPMI-hang-proof polling

- GPU load switches from bee-gpu-burn (compute burn, ~88% TDP) to
  dcgmproftester -t 1004 / targeted_power via
  resolveBenchmarkPowerLoadCommand — the same engine Power/Thermal Fit
  uses and the hottest sustained NVIDIA load we have, so fans are
  actually pushed toward their ceiling.
- Sample loop is now IPMI-hang-proof: every ipmitool read is time-boxed
  in an abandonable goroutine, and the poll interval backs off
  geometrically (1s→30s) when reads are slow, tightening again on
  recovery. A plateau is only trusted while telemetry is healthy;
  degraded runs ride out to MaxLoadSec. Summary gains fan_samples /
  telemetry_degraded. Drops the per-second nvidia-smi+power+cpu-temp
  sampling from the hot loop.
- Dead code removed: FanStressRow, GPUStressMetric, sampleFanStressRow,
  sampleGPUStressMetrics, WriteFanStressCSV/WriteFanSensorsCSV,
  analyzeMaxTemp, sampleSystemPowerResolved.

Topology fan tiles:
- size encodes the fan's ceiling RPM (its class), not current speed;
  coloured fill rising from the bottom encodes live duty cycle
  (current / ceiling), shown only when the ceiling was measured.
- glyph spin rate now maps absolute RPM into a human-perceptible band
  (fanSpinPeriodSec: 2.2s/turn at <=1000 RPM, 0.35s at >=13000).
- the "N fans · N OK · tile size ∝ …" caption line is gone.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019VHG21rgTUiR1G3qFHTVmN
This commit is contained in:
Mikhail Chusavitin
2026-09-04 11:02:31 +03:00
co-authored by Claude Sonnet 5
parent bb2a501a28
commit 8cb250f3f4
6 changed files with 328 additions and 267 deletions
+20 -2
View File
@@ -202,8 +202,9 @@ func TestTopoPageRendersCoolingFansAsFlexRow(t *testing.T) {
if !strings.Contains(body, "FAN1 · 4200 RPM") || !strings.Contains(body, "FAN2 · 15000 RPM") {
t.Fatalf("topo page missing per-fan RPM tooltips: %s", body)
}
if !strings.Contains(body, "2 fans") {
t.Fatalf("topo page missing fan count caption: %s", body)
// No observed ceiling in this test → tiles show no duty fill and say so.
if !strings.Contains(body, "ceiling not measured") {
t.Fatalf("topo fan tooltip should note the ceiling is unmeasured: %s", body)
}
// Component-detail fallback endpoint must resolve the "fan" type.
@@ -217,6 +218,23 @@ func TestTopoPageRendersCoolingFansAsFlexRow(t *testing.T) {
}
}
func TestFanSpinPeriodSec(t *testing.T) {
// Clamped at both ends; monotonically faster (smaller period) with RPM.
if got := fanSpinPeriodSec(200); got != 2.2 {
t.Fatalf("low RPM: got %v want 2.2 (slowest visible)", got)
}
if got := fanSpinPeriodSec(25000); got != 0.35 {
t.Fatalf("high RPM: got %v want 0.35 (fastest visible)", got)
}
mid := fanSpinPeriodSec(7000)
if mid <= 0.35 || mid >= 2.2 {
t.Fatalf("mid RPM period %v out of band", mid)
}
if fanSpinPeriodSec(10000) >= fanSpinPeriodSec(3000) {
t.Fatalf("higher RPM must spin faster (shorter period)")
}
}
func TestTopoPageRendersStorageDisksGroupedByType(t *testing.T) {
dir := t.TempDir()
path := filepath.Join(dir, "audit.json")