reanimator/bee
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 1 Wiki Activity

Refactor validate modes, fix benchmark report and IPMI power

Browse Source 
- Replace diag level 1-4 dropdown with Validate/Stress radio buttons
- Validate: dcgmi L2, 60s CPU, 256MB/1p memtester, SMART short
- Stress: dcgmi L3 + targeted_stress in Run All, 30min CPU, 1GB/3p memtester, SMART long/NVMe extended
- Parallel GPU mode: spawn single task for all GPUs instead of splitting per model
- Benchmark table: per-GPU columns for sequential runs, server-wide column for parallel
- Benchmark report converted to Markdown with server model, GPU model, version in header; only steady-state charts
- Fix IPMI power parsing in benchmark (was looking for 'Current Power', correct field is 'Instantaneous power reading')

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Michael Chus
2026-04-08 00:42:12 +03:00
parent dd26e03b2d
commit b2f8626fee
12 changed files with 332 additions and 164 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
audit/internal/platform/sat.go
View File

@@ -531,9 +531,13 @@ func memoryStressSizeArg() string {
return fmt.Sprintf("%dM", targetMB)
}
func (s *System) RunMemoryAcceptancePack(ctx context.Context, baseDir string, logFunc func(string)) (string, error) {
sizeMB := envInt("BEE_MEMTESTER_SIZE_MB", 128)
passes := envInt("BEE_MEMTESTER_PASSES", 1)
func (s *System) RunMemoryAcceptancePack(ctx context.Context, baseDir string, sizeMB, passes int, logFunc func(string)) (string, error) {
if sizeMB <= 0 {
sizeMB = 256
}
if passes <= 0 {
passes = 1
}
return runAcceptancePackCtx(ctx, baseDir, "memory", []satJob{
{name: "01-free-before.log", cmd: []string{"free", "-h"}},
{name: "02-memtester.log", cmd: []string{"memtester", fmt.Sprintf("%dM", sizeMB), fmt.Sprintf("%d", passes)}},
@@ -590,7 +594,7 @@ func (s *System) RunCPUAcceptancePack(ctx context.Context, baseDir string, durat
}, logFunc)
}
func (s *System) RunStorageAcceptancePack(ctx context.Context, baseDir string, logFunc func(string)) (string, error) {
func (s *System) RunStorageAcceptancePack(ctx context.Context, baseDir string, extended bool, logFunc func(string)) (string, error) {
if baseDir == "" {
baseDir = "/var/log/bee-sat"
}
@@ -622,7 +626,7 @@ func (s *System) RunStorageAcceptancePack(ctx context.Context, baseDir string, l
break
}
prefix := fmt.Sprintf("%02d-%s", index+1, filepath.Base(devPath))
commands := storageSATCommands(devPath)
commands := storageSATCommands(devPath, extended)
for cmdIndex, job := range commands {
if ctx.Err() != nil {
break
@@ -1086,17 +1090,25 @@ func listStorageDevices() ([]string, error) {
return parseStorageDevices(string(out)), nil
}
func storageSATCommands(devPath string) []satJob {
func storageSATCommands(devPath string, extended bool) []satJob {
if strings.Contains(filepath.Base(devPath), "nvme") {
selfTestLevel := "1"
if extended {
selfTestLevel = "2"
}
return []satJob{
{name: "nvme-id-ctrl", cmd: []string{"nvme", "id-ctrl", devPath, "-o", "json"}},
{name: "nvme-smart-log", cmd: []string{"nvme", "smart-log", devPath, "-o", "json"}},
{name: "nvme-device-self-test", cmd: []string{"nvme", "device-self-test", devPath, "-s", "1", "--wait"}},
{name: "nvme-device-self-test", cmd: []string{"nvme", "device-self-test", devPath, "-s", selfTestLevel, "--wait"}},
}
}
smartTestType := "short"
if extended {
smartTestType = "long"
}
return []satJob{
{name: "smartctl-health", cmd: []string{"smartctl", "-H", "-A", devPath}},
{name: "smartctl-self-test-short", cmd: []string{"smartctl", "-t", "short", devPath}},
{name: "smartctl-self-test-short", cmd: []string{"smartctl", "-t", smartTestType, devPath}},
}
}