bb-open/sextant-overlay-bbopen
1
0
Fork 0
Code Issues Pull requests 2 Projects Releases Packages Wiki Activity Actions

Follow DAWO-Core main on Codeberg, now that 0.1.2 carries printing #1

Merged
bbadmin merged 1 commit from chore/core-follows-main into main 2026-08-18 07:13:17 +00:00
Conversation 0 Commits 1 Files changed 2 +61 -82
bbadmin commented 2026-08-17 23:01:54 +00:00
Owner
Copy link

The overlay tracked feat/printing-driver-set-and-discovery because core main declared the same dawo.printing options and would not evaluate for the workplace class. Core 0.1.2 (06b0d7d) ends that: modules/services/printing.nix is byte-identical to the f250cd4 the lock held, and main is otherwise ahead.

The branch pin also broke the console. The upstream watcher reads the remote HEAD, so it compared our revision against a ref it had never seen and staged an update that would have removed the printing work. Lock and watcher now follow the same thing.

The URL moves to Codeberg with it: the core is run as a public project there (ADR 0024), and both forges carry 06b0d7d today, so this is a home address and not a content change.

Following main means taking the core's own pins, so this change carries them: nixpkgs 4 July to 9 August, plus nixos-hardware, home-manager and rust-overlay. That closes the pin age the fleet SBOM measured.

Measured

Both device classes evaluate on this tree:

dawo-inspoelstraat -> nixos-system-dawo-inspoelstraat-26.05.20260809.fcb8fcd
e2e5               -> nixos-system-e2e5-26.05.20260809.fcb8fcd

The production gate-runner (0.86.0) validated this ref against both hosts and returned ok:true. A control run with a ref that does not exist failed with couldn't find remote ref, so the ref is genuinely fetched and merged rather than ignored.

That pair is also the evidence the 1.0 gate asks for in item 6: a change carrying a flake.lock bump, evaluated for both device classes through the remote gate. Recorded in DAWO/DAWO-Sextant#36.

Before merging

Two things follow this, and neither is in this PR:

  • The console's upstreamRepo still points at code.overheid. It has to move to Codeberg too, or the watcher and this lock disagree the first time one forge moves.
  • This moves every device's nixpkgs by five weeks. Rings are promoted separately, so merging does not move a device by itself, but the first ring promotion after this is a large rebuild.
The overlay tracked `feat/printing-driver-set-and-discovery` because core main declared the same `dawo.printing` options and would not evaluate for the workplace class. Core 0.1.2 (`06b0d7d`) ends that: `modules/services/printing.nix` is byte-identical to the `f250cd4` the lock held, and main is otherwise ahead. The branch pin also broke the console. The upstream watcher reads the remote HEAD, so it compared our revision against a ref it had never seen and staged an update that would have removed the printing work. Lock and watcher now follow the same thing. The URL moves to Codeberg with it: the core is run as a public project there (ADR 0024), and both forges carry `06b0d7d` today, so this is a home address and not a content change. Following main means taking the core's own pins, so this change carries them: nixpkgs 4 July to 9 August, plus nixos-hardware, home-manager and rust-overlay. That closes the pin age the fleet SBOM measured. ## Measured Both device classes evaluate on this tree: ``` dawo-inspoelstraat -> nixos-system-dawo-inspoelstraat-26.05.20260809.fcb8fcd e2e5 -> nixos-system-e2e5-26.05.20260809.fcb8fcd ``` The production gate-runner (0.86.0) validated this ref against both hosts and returned `ok:true`. A control run with a ref that does not exist failed with `couldn't find remote ref`, so the ref is genuinely fetched and merged rather than ignored. That pair is also the evidence the 1.0 gate asks for in item 6: a change carrying a `flake.lock` bump, evaluated for both device classes through the remote gate. Recorded in `DAWO/DAWO-Sextant#36`. ## Before merging Two things follow this, and neither is in this PR: - The console's `upstreamRepo` still points at code.overheid. It has to move to Codeberg too, or the watcher and this lock disagree the first time one forge moves. - This moves every device's nixpkgs by five weeks. Rings are promoted separately, so merging does not move a device by itself, but the first ring promotion after this is a large rebuild.
chore(core): follow main again, now that 0.1.2 carries printing
Some checks failed
ci / evaluate (push) Failing after 1m12s
Details
ci / update-owned-inputs (push) Has been skipped
Details
ci / evaluate (pull_request) Failing after 1s
Details
ci / update-owned-inputs (pull_request) Has been skipped
Details
19c3562268 
The overlay tracked feat/printing-driver-set-and-discovery because core main
declared the same dawo.printing options and would not evaluate for the
workplace class. Core 0.1.2 (06b0d7d) ends that: modules/services/printing.nix
is byte-identical to the f250cd4 the lock held, and main is otherwise ahead.

A branch pin also broke the console. The upstream watcher reads the remote
HEAD, so it compared our revision against a ref it had never seen and staged
an update that would have removed the printing work. Lock and watcher now
follow the same thing.

Following main means taking the core's own pins, so this carries them:
nixpkgs 4 July to 9 August, nixos-hardware, home-manager, rust-overlay and the
rest. That closes the pin age the fleet SBOM measured, and it is why this
change is worth evaluating rather than merging on sight.

Evaluated both device classes on this tree:
  dawo-inspoelstraat -> nixos-system-dawo-inspoelstraat-26.05.20260809.fcb8fcd
  e2e5               -> nixos-system-e2e5-26.05.20260809.fcb8fcd
bbadmin force-pushed chore/core-follows-main from 19c3562268
Some checks failed
ci / evaluate (push) Failing after 1m12s
Details
ci / update-owned-inputs (push) Has been skipped
Details
ci / evaluate (pull_request) Failing after 1s
Details
ci / update-owned-inputs (pull_request) Has been skipped
Details
to a58e04c394
Some checks failed
ci / evaluate (push) Failing after 46s
Details
ci / update-owned-inputs (push) Has been skipped
Details
ci / evaluate (pull_request) Failing after 1s
Details
ci / update-owned-inputs (pull_request) Has been skipped
Details
2026-08-17 23:09:57 +00:00 Compare
bbadmin changed title from Follow core main again, now that 0.1.2 carries printing to Follow DAWO-Core main on Codeberg, now that 0.1.2 carries printing 2026-08-17 23:11:38 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
bb-open/sextant-overlay-bbopen!1
No description provided.