virtualisation:proxmox_ve:vms_cts:initial_debugging_vm_created_best_practices
Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| virtualisation:proxmox_ve:vms_cts:initial_debugging_vm_created_best_practices [2026/04/11 22:27] – removed - external edit (Unknown date) 127.0.0.1 | virtualisation:proxmox_ve:vms_cts:initial_debugging_vm_created_best_practices [2026/04/12 09:31] (current) – thum | ||
|---|---|---|---|
| Line 1: | Line 1: | ||
| + | ======= Initial debugging of VMs created with best practices ======= | ||
| + | The following tips may be useful if you have created a VM [[vm_best_practices_modern_setup|via this article]]. | ||
| + | |||
| + | ====== Proxmox VE - common dmesg Warnings in KVM/q35 VMs ====== | ||
| + | |||
| + | These warnings appear in the guest dmesg of a properly configured Proxmox VM | ||
| + | (q35, OVMF/UEFI, VirtIO SCSI as per the above " | ||
| + | |||
| + | ===== SATA/ATA "link down" Messages ===== | ||
| + | |||
| + | ata1: SATA link down | ||
| + | ata3: SATA link down | ||
| + | ata4: SATA link down | ||
| + | ata5: SATA link down | ||
| + | ata6: SATA link down | ||
| + | |||
| + | The q35 chipset always emulates an AHCI SATA controller with 6 ports — | ||
| + | even when only VirtIO SCSI is used. The unused ports simply report | ||
| + | "no device found" | ||
| + | |||
| + | '' | ||
| + | |||
| + | **Verdict: | ||
| + | |||
| + | ===== shpchp: Slot Initialization Failed ===== | ||
| + | |||
| + | shpchp 0000: | ||
| + | shpchp 0000: | ||
| + | |||
| + | The Linux Standard Hot-Plug PCI driver tries to register itself on bridges | ||
| + | that QEMU/q35 only partially emulates. This causes no functional problems | ||
| + | and is a well-known cosmetic issue in KVM virtual machines. | ||
| + | |||
| + | **Verdict: | ||
| + | |||
| + | ===== powernow-k8: | ||
| + | |||
| + | powernow-k8: | ||
| + | |||
| + | The guest kernel identifies the CPU as "AMD QEMU Virtual CPU" and loads | ||
| + | the legacy '' | ||
| + | scaling is not supported inside the VM. This is expected behavior when | ||
| + | not using the '' | ||
| + | |||
| + | **Verdict: | ||
| + | |||
| + | ===== snd_hda_intel: | ||
| + | |||
| + | snd_hda_intel 0000: | ||
| + | |||
| + | No audio device is configured in the VM. The driver loads anyway but finds | ||
| + | nothing to attach to. Either ignore this, or disable the audio device in | ||
| + | the VM configuration to suppress the message. | ||
| + | |||
| + | **Verdict: | ||
| + | |||
| + | ===== What you should also see ===== | ||
| + | |||
| + | * '' | ||
| + | * UEFI/Secure Boot active, EFI disk detected correctly | ||
| + | * Disk recognized via VirtIO SCSI with Write Cache + FUA/DPO support | ||
| + | * Clean boot in ≤ 3 seconds on a managed Vikings Proxmox VE Ceph Cluster with Hardware post 2025 sans cloud-init | ||
| + | |||
| + | ===== Summary ===== | ||
| + | |||
| + | ^ Warning ^ Cause ^ Action ^ | ||
| + | | '' | ||
| + | | '' | ||
| + | | '' | ||
| + | | '' | ||
