fix(cache-warm): unblock the nightly lock refresh (gitlab API 403) and the x1 flatpak build #59

Closed
hermes wants to merge 1 commit from hermes/cache-warm-gitlab403-flatpak into main AGit
Contributor

Two independent breaks made cache-warm fail every night since 2026-10-06/09
and froze the fleet's flake.lock at 603b74b (06.10):

  1. simple-nixos-mailserver followed a gitlab: input. GitLab now answers the
    anonymous /repository/commits API — the call nix uses to resolve a branch
    HEAD — with a Cloudflare challenge (403 for every project and every UA;
    verified 11.10 from hermes and reproduced with a single 'nix flake update'
    here). The weekly-update run died before any build, so no lock was pushed.
    The API archive endpoint still serves anonymously, so the input now follows
    the same nixos-26.05 HEAD as an API-archive tarball. The transitive
    'blobs' input is rev-pinned and resolves via that endpoint, unchanged.

  2. The flatpak overlay carried a diff anchored at profile/flatpak.fish:9;
    flatpak 1.18.4 (nixpkgs-unstable since 29.09) reworked that function, so
    'Hunk #1 FAILED' killed flatpak -> xdg-desktop-portal -> the whole
    nixos-system-x1 closure. Applied via postPatch sed anchored on the line
    itself instead (version-agnostic, guarded no-op if upstream ever unsets
    the variable); built flatpak 1.18.4 with the overlay after the change.

PR #36 (push-before-exit) stays closed: the gating is intended, Harald
decided 24.09 on the PR; this fixes the reasons the run fails, not the gate.

Two independent breaks made cache-warm fail every night since 2026-10-06/09 and froze the fleet's flake.lock at 603b74b (06.10): 1. simple-nixos-mailserver followed a gitlab: input. GitLab now answers the anonymous /repository/commits API — the call nix uses to resolve a branch HEAD — with a Cloudflare challenge (403 for every project and every UA; verified 11.10 from hermes and reproduced with a single 'nix flake update' here). The weekly-update run died before any build, so no lock was pushed. The API archive endpoint still serves anonymously, so the input now follows the same nixos-26.05 HEAD as an API-archive tarball. The transitive 'blobs' input is rev-pinned and resolves via that endpoint, unchanged. 2. The flatpak overlay carried a diff anchored at profile/flatpak.fish:9; flatpak 1.18.4 (nixpkgs-unstable since 29.09) reworked that function, so 'Hunk #1 FAILED' killed flatpak -> xdg-desktop-portal -> the whole nixos-system-x1 closure. Applied via postPatch sed anchored on the line itself instead (version-agnostic, guarded no-op if upstream ever unsets the variable); built flatpak 1.18.4 with the overlay after the change. PR #36 (push-before-exit) stays closed: the gating is intended, Harald decided 24.09 on the PR; this fixes the reasons the run fails, not the gate.
fix(cache-warm): unblock the nightly lock refresh (gitlab API 403) and the x1 flatpak build
Some checks failed
nix flake check / flake-check (pull_request) Failing after 3s
9c441ef4af
Two independent breaks made cache-warm fail every night since 2026-10-06/09
and froze the fleet's flake.lock at 603b74b (06.10):

1. simple-nixos-mailserver followed a gitlab: input. GitLab now answers the
   anonymous /repository/commits API — the call nix uses to resolve a branch
   HEAD — with a Cloudflare challenge (403 for every project and every UA;
   verified 11.10 from hermes and reproduced with a single 'nix flake update'
   here). The weekly-update run died before any build, so no lock was pushed.
   The API archive endpoint still serves anonymously, so the input now follows
   the same nixos-26.05 HEAD as an API-archive tarball. The transitive
   'blobs' input is rev-pinned and resolves via that endpoint, unchanged.

2. The flatpak overlay carried a diff anchored at profile/flatpak.fish:9;
   flatpak 1.18.4 (nixpkgs-unstable since 29.09) reworked that function, so
   'Hunk #1 FAILED' killed flatpak -> xdg-desktop-portal -> the whole
   nixos-system-x1 closure. Applied via postPatch sed anchored on the line
   itself instead (version-agnostic, guarded no-op if upstream ever unsets
   the variable); built flatpak 1.18.4 with the overlay after the change.

PR #36 (push-before-exit) stays closed: the gating is intended, Harald
decided 24.09 on the PR; this fixes the reasons the run fails, not the gate.
hermes closed this pull request 2026-10-11 18:51:49 +02:00
Some checks failed
nix flake check / flake-check (pull_request) Failing after 3s

Pull request closed

Sign in to join this conversation.
No reviewers
No labels
No milestone
No project
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
harald/nixcfg!59
No description provided.