- Collect hardware event logs (last 7 days) from Systems and Managers/SEL LogServices - Parse AMI raw IPMI dump messages into readable descriptions (Sensor_Type: Event_Type) - Filter out audit/journal/non-hardware log services; only SEL from Managers - MSI ghost GPU filter: exclude processor GPU entries with temperature=0 when host is powered on - Reanimator collected_at uses InventoryData/Status.LastModifiedTime (30-day fallback) - Invalidate Redfish inventory CRC groups before host power-on - Log inventory LastModifiedTime age in collection logs - Drop SecureBoot collection (SecureBootMode, SecureBootDatabases) — not hardware inventory - Add build version to UI footer via template - Add MSI Redfish API reference doc to bible-local/docs/ ADL-032–ADL-035 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
75 lines
2.7 KiB
Go
75 lines
2.7 KiB
Go
package redfishprofile
|
|
|
|
import "strings"
|
|
|
|
func msiProfile() Profile {
|
|
return staticProfile{
|
|
name: "msi",
|
|
priority: 20,
|
|
safeForFallback: true,
|
|
matchFn: func(s MatchSignals) int {
|
|
score := 0
|
|
if containsFold(s.SystemManufacturer, "micro-star") || containsFold(s.ChassisManufacturer, "micro-star") {
|
|
score += 80
|
|
}
|
|
if containsFold(s.SystemManufacturer, "msi") || containsFold(s.ChassisManufacturer, "msi") {
|
|
score += 40
|
|
}
|
|
for _, hint := range s.ResourceHints {
|
|
if strings.HasPrefix(hint, "/redfish/v1/Chassis/GPU") {
|
|
score += 10
|
|
break
|
|
}
|
|
}
|
|
return min(score, 100)
|
|
},
|
|
extendAcquisition: func(plan *AcquisitionPlan, _ MatchSignals) {
|
|
ensureSnapshotWorkers(plan, 6)
|
|
ensurePrefetchWorkers(plan, 8)
|
|
ensureETABaseline(plan, AcquisitionETABaseline{
|
|
DiscoverySeconds: 12,
|
|
SnapshotSeconds: 120,
|
|
PrefetchSeconds: 25,
|
|
CriticalPlanBSeconds: 35,
|
|
ProfilePlanBSeconds: 25,
|
|
})
|
|
ensureRatePolicy(plan, AcquisitionRatePolicy{
|
|
TargetP95LatencyMS: 1000,
|
|
ThrottleP95LatencyMS: 2200,
|
|
MinSnapshotWorkers: 2,
|
|
MinPrefetchWorkers: 2,
|
|
DisablePrefetchOnErrors: true,
|
|
})
|
|
ensureRecoveryPolicy(plan, AcquisitionRecoveryPolicy{
|
|
EnableProfilePlanB: true,
|
|
})
|
|
addPlanNote(plan, "msi gpu chassis probes enabled")
|
|
},
|
|
refineAcquisition: func(resolved *ResolvedAcquisitionPlan, discovered DiscoveredResources, _ MatchSignals) {
|
|
for _, chassisPath := range discovered.ChassisPaths {
|
|
if !strings.HasPrefix(chassisPath, "/redfish/v1/Chassis/GPU") {
|
|
continue
|
|
}
|
|
addPlanPaths(&resolved.SeedPaths, chassisPath)
|
|
addPlanPaths(&resolved.Plan.SeedPaths, chassisPath)
|
|
addPlanPaths(&resolved.CriticalPaths, joinPath(chassisPath, "/Sensors"))
|
|
addPlanPaths(&resolved.Plan.CriticalPaths, joinPath(chassisPath, "/Sensors"))
|
|
addPlanPaths(&resolved.Plan.PlanBPaths, joinPath(chassisPath, "/Sensors"))
|
|
}
|
|
},
|
|
applyAnalysisDirectives: func(d *AnalysisDirectives, _ MatchSignals) {
|
|
d.EnableGenericGraphicsControllerDedup = true
|
|
},
|
|
refineAnalysis: func(plan *ResolvedAnalysisPlan, snapshot map[string]interface{}, discovered DiscoveredResources, _ MatchSignals) {
|
|
if snapshotHasGPUProcessor(snapshot, discovered.SystemPaths) && snapshotHasPathPrefix(snapshot, "/redfish/v1/Chassis/GPU") {
|
|
plan.Directives.EnableProcessorGPUFallback = true
|
|
plan.Directives.EnableMSIProcessorGPUChassisLookup = true
|
|
plan.Directives.EnableMSIGhostGPUFilter = true
|
|
addAnalysisLookupMode(plan, "msi-index")
|
|
addAnalysisNote(plan, "msi analysis enables processor-gpu fallback from discovered GPU chassis")
|
|
addAnalysisNote(plan, "msi ghost-gpu filter enabled: GPUs with temperature=0 on powered-on host are excluded")
|
|
}
|
|
},
|
|
}
|
|
}
|