Hi,
After recent upgrades of the kernel and Incus on our ubuntu / Noble host, VMs with attached physical NIC devices fail to start with the message
Failed setting up device via monitor: Failed setting up device "mce0": Failed adding NIC device: Monitor command timed out: "device_add" after 30s
Earlier, this has been due to incompatibilities between Incus and Qemu. So I wonder if that is the case again.
Best regards, Fredrik
VM devices section (for FreeBSD/pfsense VM with Mellanox 25Gb/s NIC)
devices:
mce0:
name: mce0
nictype: physical
parent: e3d0
type: nic
mce1:
name: mce1
nictype: physical
parent: e3d1
type: nic
Journalctl on host:
Aug 29 11:48:29 gojira kernel: mlx5_core 0000:01:00.1: MLX5E: StrdRq(1) RqSz(8) StrdSz(2048) RxCqeCmprss(0 enhanced)
Aug 29 11:48:29 gojira kernel: mlx5_core 0000:01:00.1 e3d1: renamed from eth0
Aug 29 11:48:29 gojira kernel: mlx5_core 0000:01:00.1: is_dpll_supported:213:(pid 610): Missing SyncE capability
Instance Qemu log
[2026-08-29T11:47:58Z] QUERY: {"execute":"device_add","arguments":{"addr":"00.0","bootindex":1,"bus":"qemu_pcie4","driver":"vfio-pci","host":"0000:01:00.0","id":"dev-incus_mce0"},"id":28}
$ incus --version
7.4$sudo /opt/incus/bin/qemu-system-x86_64 --version
/opt/incus/bin/qemu-system-x86_64: error while loading shared libraries: liburing.so.2: cannot open shared object file: No such file or directory