bhyve Ubuntu Guest won't start after update to FreeBSD 12.3

[Nugman]

Hi Folks,

today I've updated my server from FreeBSD 12.2 to 12.3. After the reboot one of my Bhyve guests does not start up anymore. I have a couple of similar Linux Guest VMs on the same machine (all Ubuntu 20.04 LTS) which do work fine, but this one does not.

To manage the VMs, I use vm-bhyve.
When I try to start the guest machine with 'vm start influxdb', bhyve-grub dumps core and quits.
The logfiles says:

März 07 17:10:47: initialising
März 07 17:10:47:  [loader: grub]
März 07 17:10:47:  [cpu: 1]
März 07 17:10:47:  [memory: 1024M]
März 07 17:10:47:  [hostbridge: standard]
März 07 17:10:47:  [com ports: com1]
März 07 17:10:47:  [uuid: ffef441d-a827-11ea-9e0c-bc5ff4cb7409]
März 07 17:10:47:  [utctime: yes]
März 07 17:10:47:  [debug mode: no]
März 07 17:10:47:  [primary disk: disk0]
März 07 17:10:47:  [primary disk dev: sparse-zvol]
März 07 17:10:47: initialising network device tap3
März 07 17:10:47: adding tap3 -> vm-public (public addm)
März 07 17:10:47: bring up tap3 -> vm-public (public addm)
März 07 17:10:47: booting
März 07 17:10:47: create file /mnt/vm/influxdb/device.map
März 07 17:10:47:  -> (hd0) /dev/zvol/zroot/vm/influxdb/disk0
März 07 17:10:47: /usr/local/sbin/grub-bhyve -c /dev/nmdm-influxdb.1A -m /mnt/vm/influxdb/device.map -M 1024M -r hd0,2 -d /grub -g grub.cfg influxdb
März 07 17:10:48: fatal; loader returned error 139
März 07 17:10:48: destroying network device tap3
März 07 17:10:48: stopped

The guest configuration looks like this:

loader="grub"
grub_run_partition="2"
grub_run_dir="/grub"
grub_run_file="grub.cfg"
guest="linux"
cpu=1
memory=1024M
network0_type="virtio-net"
network0_switch="public"
disk0_type="virtio-blk"
disk0_name="disk0"
disk0_dev="sparse-zvol"
uuid="ffef441d-a827-11ea-9e0c-bc5ff4cb7409"
network0_mac="58:9c:fc:09:da:fa"

Any hints on what to do to revive the VM?

 

[Cath O'Deray]

Might the issue be coincidental? At a glance I can't see anything relevant in release notes.

Are packages up-to-date following the OS upgrade?

 

[Nugman]

Might be coincidental, yes. Strange thing is, that the other VMs that are installed and configured exactly the same (except for the services they run) work fine. Beside the upgrade to 12.3 I didn't change anything since the last security updates in January. After installing those, I rebooted the machine and everything came up fine.

OFC I did an 'pkg upgrade' before and after the system upgrade to 12.3.
I simply don't know how to debug this. The log doesn't say anything relevant beside that a core was dumped. And to analyse a coredump is beyond me.

In the meantime I quickly installed the InfluxDB in a new VM and restored the data from the backup. So I just lost the measurements from today while the VM was down.