Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Ensure the machine-id operation is the last one #6

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

cjeanner
Copy link

We must ensure the modified image doesn't have any content in
/etc/machine-id file. This will then trigger the usual
systemd-machine-id-setup during the first boot of the image, ensuring
we'll get an as unique as possible ID.

This is mandatory, since the tools are generating a machine-id at
different stages, for instance in order to ensure packages are
installing without any issue[1]

Of course, if the operator wants to get an actual machine-id, they can
just disable the "machine-id" operation. Note that it's enabled by
default, so the issue is more about ordering than anything.

[1] For instance here: https://github.com/rwmjones/guestfs-tools/blob/3a498512f58bc431db490e96cdb712b19389bda4/customize/customize_run.ml#L170-L188

We must ensure the modified image doesn't have any content in
/etc/machine-id file. This will then trigger the usual
systemd-machine-id-setup during the first boot of the image, ensuring
we'll get an as unique as possible ID.

This is mandatory, since the tools are generating a machine-id at
different stages, for instance in order to ensure packages are
installing without any issue[1]

Of course, if the operator wants to get an actual machine-id, they can
just disable the "machine-id" operation. Note that it's enabled by
default, so the issue is more about ordering than anything.

[1] For instance here: https://github.com/rwmjones/guestfs-tools/blob/3a498512f58bc431db490e96cdb712b19389bda4/customize/customize_run.ml#L170-L188
@rwmjones
Copy link
Member

rwmjones commented Mar 25, 2022

Patch sort of looks reasonable, although running it after the --customize operations [edit: which have order 99] might be wrong since those operations might be needed to fix any earlier built-in operation. However I asked on the bug for a reproducer so I can understand the exact sequence of steps that is being done and what goes wrong, because without that I cannot really tell if the bug has been fixed or not.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants