Questions to Determine¶
- Is the NVMe drive physically detected at the PCIe bus level (
lspci)? - Does the BIOS/UEFI firmware enumerate the drive during POST?
- Are there
dmesgerrors indicating a PCIe link training failure or AER (Advanced Error Reporting) events? - Was a BIOS or firmware update applied as part of the patching window that could have changed PCIe settings?
- Is the drive physically seated properly, or could thermal expansion/vibration have loosened it?
- Does the NVMe drive appear when tested in a different PCIe slot or a different server?
- Is the PCIe slot itself functional (test with another device)?
- Are there any BMC/iDRAC hardware event logs indicating a component failure?
- Was the NVMe driver module (
nvme/nvme_core) loaded successfully on boot? - Is this a known issue with this NVMe firmware version and the current kernel?