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

fix(collector): stop NVIDIA enrichment from clobbering PCIe status

Browse Source 
enrichPCIeWithNVIDIAData unconditionally overwrote dev.Status after
collectPCIe() had already flagged a Warning/Critical (e.g. PCIe link
speed degraded), so a clean ECC/remap/reset readout silently downgraded
that finding back to OK while leaving the stale ErrorDescription behind.

Add a severity-ordered merge (OK/Unknown < Warning < Critical) shared
via mergeDeviceStatus in contract.go, and route both the NVIDIA status
calculation and the driver-unavailable fallback through it.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
Mikhail Chusavitin
2026-08-04 17:28:47 +03:00
co-authored by Claude Sonnet 5
parent 6c8be629d2
commit a34e823f82
3 changed files with 153 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
audit/internal/collector/nvidia.go
+7 -7
View File
@@ -88,23 +88,24 @@ func enrichPCIeWithNVIDIAData(devs []schema.HardwarePCIeDevice, gpuByBDF map[str
}
status := statusOK
desc := ""
if info.ECCUncorrected != nil && *info.ECCUncorrected > 0 {
status = statusWarning
devs[i].ErrorDescription = stringPtr("GPU reports uncorrected ECC errors")
desc = "GPU reports uncorrected ECC errors"
}
if info.RemapUncorrectable != nil && *info.RemapUncorrectable > 0 {
status = statusWarning
devs[i].ErrorDescription = stringPtr("GPU has uncorrectable row remap events (bad HBM cell repair scheduled)")
desc = "GPU has uncorrectable row remap events (bad HBM cell repair scheduled)"
}
if info.RemapFailure != nil && *info.RemapFailure {
status = statusCritical
devs[i].ErrorDescription = stringPtr("GPU row remap failed to commit to InfoROM (XID 64)")
desc = "GPU row remap failed to commit to InfoROM (XID 64)"
}
if info.ResetRequired != nil && *info.ResetRequired {
status = statusCritical
devs[i].ErrorDescription = stringPtr("GPU requires a reset")
desc = "GPU requires a reset"
}
devs[i].Status = &status
mergeDeviceStatus(&devs[i], status, desc)
injectNVIDIATelemetry(&devs[i], info)
enriched++
}
@@ -326,8 +327,7 @@ func isNVIDIADevice(dev schema.HardwarePCIeDevice) bool {
}
func setPCIeFallback(dev *schema.HardwarePCIeDevice) {
status := statusUnknown
dev.Status = &status
mergeDeviceStatus(dev, statusUnknown, "")
}
func injectNVIDIATelemetry(dev *schema.HardwarePCIeDevice, info nvidiaGPUInfo) {