machine-id in Talos world

[tuxtof]

Is there a document or a link that explains how Talos generates the /etc/machine-id and the logic behind it?

In the Linux world, for operating systems using systemd, this behavior is typically documented here:

systemd machine ID documentation

Systemd attempts to retrieve the D-Bus machine ID from `/var/lib/dbus/machine-id`, 
the value of the kernel command line option `container_uuid`, the KVM DMI `product_uuid`, 
or the device tree `vm.uuid` on KVM systems. It also considers the Xen hypervisor UUID, and finally,
if none of these options are available, it generates a random UUID.

In a virtual machine, we usually expect the UUID of the VM to be present in /etc/machine-id, but in Talos, this is not the case; instead, a random ID is generated, which can be problematic in certain situations.

Are there any plans to improve this aspect or any way to change the default behavior?

Thank you.

[smira]

No, we don't plan to.

In Kubernetes world, both values (SMBIOS UUID) and machine-id are published:

System Info:
  Machine ID:                 308e9eaa496eca97a7b497f9e57ece00
  System UUID:                90dc6a4f-4f8d-4c21-89e0-10bb6e22c03d
  Boot ID:                    b82e8fa6-64fa-4f35-95dd-5606fd429e07

So you can consume the one you need.

Machine ID will change if the machine is wiped and re-joins the cluster, but it's guaranteed to be random.

[tuxtof]

the main issue , following k8s recommendation, System UUID usage is not clear, This field is normally specific to Red Hat hosts !!!

[smira]

See documentation:

       The /etc/machine-id file contains the unique machine ID of the
       local system that is set during installation or boot. The machine
       ID is a single newline-terminated, hexadecimal, 32-character,
       lowercase ID. When decoded from hexadecimal, this corresponds to a
       16-byte/128-bit value. This ID may not be all zeros.

       The machine ID is usually generated from a random source during
       system installation or first boot and stays constant for all
       subsequent boots. Optionally, for stateless systems, it is
       generated during runtime during early boot if necessary.
       
       The machine ID does not change based on local or network
       configuration or when hardware is replaced.

I don't how SMBIOS UUID is related to Red Hat, I'm lost.

[tuxtof]

i talk from Kubernetes perspective

kubectl explain node.status.nodeInfo

machineID -required-
MachineID reported by the node. For unique machine identification in the
cluster this field is preferred. Learn more from man(5) machine-id:
http://man7.org/linux/man-pages/man5/machine-id.5.html

systemUUID -required-
SystemUUID reported by the node. For unique machine identification MachineID
is preferred. This field is specific to Red Hat hosts
https://access.redhat.com/documentation/en-us/red_hat_subscription_management/1/html/rhsm/uuid

[smira]

I can't speak for Kubernetes documentation.

systemUUID is SMBIOS UUID, and it might not be unique due to buggy hypervisor/hardware

machineID is generated by Talos and it is actually unique.

I don't see where Talos is doing anything contradicting the documentation.

[tuxtof]

Helo Andrey , never talk about contradiction :-) was just trying to understand why i have a different behaviour between Talos and other Linux and how to adapt

in all case thanks for your response