mchus/logpile
1
0
Fork 0
Code Issues 3 Pull Requests Actions Packages Projects Releases 24 Wiki Activity

fix(collector): recover GPUs/NICs dropped on xFusion G5500 Redfish exports

Browse Source 
looksLikeGPU now falls back to resolving VendorId/DeviceId through the
pci.ids database when the BMC leaves Name/Model/Manufacturer/ClassCode
empty, so GPUs identifiable only by raw PCI IDs (e.g. NVIDIA H100 SXM5
0x10de/0x2330) are no longer misclassified as generic PCIe devices.

The replay pipeline's "backed by canonical NIC" dedup used to trust a
PCIeDevice's Links.NetworkDeviceFunctions reference at face value and
drop the device, assuming a NetworkAdapters record existed elsewhere.
On BMCs that expose resource IDs with characters (parentheses) that
404 on fetch, that canonical NIC never gets captured, so the device
carrying its actual hardware identity vanished from the export
entirely. hasResolvableLinkedMember now verifies the linked resource
is actually present in the snapshot before treating it as authoritative.

Also normalize PartNumber through normalizeRedfishIdentityField in the
GPU/PCIe parsers so a BMC-supplied literal "null" string doesn't leak
into exports verbatim.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
Mikhail Chusavitin
2026-08-18 14:17:25 +03:00
co-authored by Claude Sonnet 5
parent 2c3072cf10
commit a3567dd5f6
3 changed files with 146 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files
internal/collector/redfish.go
+28 -2
View File
@@ -4355,7 +4355,7 @@ func parseGPUWithSupplementalDocs(doc map[string]interface{}, functionDocs []map
Model: firstNonEmpty(asString(doc["Model"]), asString(doc["Name"])),
Manufacturer: asString(doc["Manufacturer"]),
SerialNumber: findFirstNormalizedStringByKeys(doc, "SerialNumber"),
PartNumber: asString(doc["PartNumber"]),
PartNumber: normalizeRedfishIdentityField(asString(doc["PartNumber"])),
Firmware: asString(doc["FirmwareVersion"]),
Status: mapStatus(doc["Status"]),
Details: redfishPCIeDetailsWithSupplementalDocs(doc, functionDocs, supplementalDocs),
@@ -4439,7 +4439,7 @@ func parsePCIeDeviceWithSupplementalDocs(doc map[string]interface{}, functionDoc
BDF: sanitizeRedfishBDF(asString(doc["BDF"])),
DeviceClass: asString(doc["DeviceType"]),
Manufacturer: asString(doc["Manufacturer"]),
PartNumber: asString(doc["PartNumber"]),
PartNumber: normalizeRedfishIdentityField(asString(doc["PartNumber"])),
SerialNumber: findFirstNormalizedStringByKeys(doc, "SerialNumber"),
VendorID: asHexOrInt(doc["VendorId"]),
DeviceID: asHexOrInt(doc["DeviceId"]),
@@ -5084,6 +5084,32 @@ func looksLikeGPU(doc map[string]interface{}, functionDocs []map[string]interfac
}
}
// Some BMCs (e.g. xFusion) leave Name/Model/Manufacturer/ClassCode empty on
// the PCIeDevice and its PCIeFunctions, exposing only raw VendorId/DeviceId.
// Resolve those through the pci.ids database so GH100/GA100/etc. GPUs are
// still recognized even without vendor-supplied model text.
vendorID := asHexOrInt(doc["VendorId"])
deviceID := asHexOrInt(doc["DeviceId"])
for _, fn := range functionDocs {
if vendorID == 0 {
vendorID = asHexOrInt(fn["VendorId"])
}
if deviceID == 0 {
deviceID = asHexOrInt(fn["DeviceId"])
}
}
if vendorID != 0 || deviceID != 0 {
resolvedText := strings.ToLower(strings.Join([]string{
pciids.VendorName(vendorID),
pciids.DeviceName(vendorID, deviceID),
}, " "))
for _, hint := range gpuHints {
if strings.Contains(resolvedText, hint) {
return true
}
}
}
return false
}