Repository navigation
feat: prepare security key sign-in packages - #776
Conversation
Import the pam-u2f source family and record its two binary outputs as recipe-covered. Hosted build and authentication evidence remain outstanding. Hive-Run: projectbluefin#478 Hive-Plan: vpn package source families Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Codex via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: Karibhou <157208370+Karibhou@users.noreply.github.com>
Hive-Run: projectbluefin#478 Hive-Plan: vpn package source families Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Codex via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: Karibhou <157208370+Karibhou@users.noreply.github.com>
|
@Karibhou this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (6 waiting): fixes are handed out before new work as soon as you have capacity. |
|
@Karibhou the hive has adopted this PR ( |
There was a problem hiding this comment.
Summary: This change adds the build recipe for a security-key sign-in component and its setup tool, copied from Fedora. I checked it from five angles. The copied files match Fedora exactly and the source checks are wired the same way as the other recipes, so nothing blocks it. The only note is that the new sentence in the backlog guide was placed under an unrelated heading.
style
docs/skills/factory-build-backlog.md:719-721— the newpam-u2f/pamu2fcfgparagraph is appended under### GNOME apps batch #484: terminal outputs and publication reconciliation(line 702), separated by a double blank line (717-718). The Yubico source-family guidance it belongs with is at line 392. Low; move it beside that text or give it its own###heading.
Verified (not findings): packages/pam-u2f/{pam-u2f.spec,sources,yubico-release-gpgkeys.asc,.gitignore} are byte-identical to src.fedoraproject.org/rpms/pam-u2f at 5b713d39; the .sig Source1 is locked in packages/pam-u2f/sources:2 and fetched from the lookaside by tools/source_pipeline.py:162-190 (bundled_sources), matching how other recipes handle non-upstream sources.
No findings from: correctness, security, intent-alignment, docs-currency.
Confidence: 5/5 (safe)
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
The hive adopted this PR ( Blocker: lane reason Moving to |
|
hive/scanner: conflicts with main; not resolved by hive because PR lacks stage/074-adopted-by-hive — contributor please merge main. 🐝 Hive Agent: — hive: agent=scanner backend=copilot model=claude-opus-5.5 copilot=1.0.88 |
Signed-off-by: Karibhou <157208370+Karibhou@users.noreply.github.com>
|
hive-base-moved: merged main at |
Signed-off-by: Karibhou <157208370+Karibhou@users.noreply.github.com>
|
Moved the pam-u2f guidance beside the Yubico source guidance in a79112e. Both — hive: backend=codex |
|
@Karibhou this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (4 waiting): fixes are handed out before new work as soon as you have capacity. |
|
@Karibhou the hive has adopted this PR ( |
Assisted-by: GPT-6 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: Karibhou <157208370+Karibhou@users.noreply.github.com>
|
hive-base-moved: merged main at 5454aad and resolved the shared configuration conflicts. |
|
@Karibhou this PR has a new head without confirmed hive push evidence, so the hive has handed this PR back to you ( |
Assisted-by: GPT-6 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: Karibhou <157208370+Karibhou@users.noreply.github.com>
|
@Karibhou this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (6 waiting): fixes are handed out before new work as soon as you have capacity. |
There was a problem hiding this comment.
Summary: Re-review of the changes since the last look at this security-key sign-in packaging change. The new commits only merge the main branch in and add one clarifying comment; the packaging itself is unchanged. One thing blocks merging: the main branch moved again right after that merge, so the branch now conflicts and its package-count figures in the architecture document are one short.
What I need from you: Merge main once more and bump the counts to 658 (653 rawhide).
Reviewed delta a79112e…9a054bba (merge of main through 5454aad plus one comment in config/factory-build-backlog.toml:310). No packages/pam-u2f/* file changed; the previous approval of the recipe stands.
docs-currency
docs/architecture.md:81,133,137-139,144,163set the recipe count to 657 (652 rawhide). That was right against merge-base 5454aad, but main has since landedlibpinyin(#796, main995f68b0) and already reads 657 — GitHub reports this PR asCONFLICTING. After the next merge of main the figures must be 658 / 653 rawhide; the same six files touched by #796 (.packit.yaml,config/*.toml|json,reports/factory-build-backlog.json) will need conflict resolution.(and the 657→658 in the table rows and the-asserts the package list is non-empty, which a list of 657 satisfies while +asserts the package list is non-empty, which a list of 658 satisfies while -validated 657 source RPMs (652 rawhide, 5 upstream) +validated 658 source RPMs (653 rawhide, 5 upstream)
94 of 657line).
style (advisory, carried from previous review — still present)
docs/skills/factory-build-backlog.md:455-460: the Yubico andpam-u2fparagraphs still sit under### Compression measurements on Btrfs(line 439); the next heading is at 462. Give them their own heading:(replacing line 455 only; the### Security key sign-in (Yubico) Yubico C releases (`libyubikey`, `ykclient`, `ykpers`) still declare `http://opensource.yubico.com/...` `Source0` URLs in Fedora Rawhide, and `libyubikey` and `ykclient` still carry legacy MD5 `sources` lines: repin `sources` to `SHA512 (<file>) = <hex>`, lock `https://opensource.yubico.com/...` as the primary `url`, and pair `https://developers.yubico.com/<project>/Releases/<file>` with Fedora's `/md5/<md5>/` lookaside path in `fallback_urls`. `ykpers` builds and links against `libyubikey` (`libyubikey.so.0`), so import `libyubikey` in the same PR. When a recipe carries a committed OpenPGP keyring whose filename does not match `*key*.asc`, `*keyring*.asc`, or `[0-9A-F]*.asc` (`packages/pkcs11-provider/simo_redhat.asc`, `packages/bc/kevin_pizzini.asc`), unignore it explicitly in the root `.gitignore` (`!packages/<pkg>/<name>.asc`) so `git add` tracks `SourceN` for `%{openpgpverify}`. By contrast, `yubikey-manager` (`python3-yubikey-manager`) is a `pyproject` package whose `[project].dependencies` generate runtime `python3dist(...)` `Requires` across multiple backlog areas (`python3-click`, `python3-fido2`, `python3-keyring` plus `secretstorage`/`jeepney`/`jaraco` in `dev-tools`, and `python3-pyscard`, `python3-pskc` in `long-tail`); keep `yubikey-manager` and `python3-yubikey-manager` pending until those Python runtime providers land in the factory.pam-u2fparagraph at 457-460 then follows under the new heading.)
No findings from: correctness, security, intent-alignment.
Confidence: 3/5 (needs attention) — 1 medium finding; 1 perspective requested changes
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
@Karibhou the hive has adopted this PR ( |
Merge origin/main (libpinyin) into the pam-u2f branch, keeping both source-lock entries, and regenerate the derived counts: architecture.md now reports 658 recipes (653 rawhide) and the factory build backlog snapshot is refreshed. Give the Yubico/pam-u2f guidance its own heading instead of leaving it under the Btrfs compression section. Hive-Fix-Attempt: 1/6 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Summary: This prepares software that lets people use a security key when signing in. It adds the build instructions for both the sign-in component and its setup tool. The packages still need a hosted build and a real sign-in test before users can rely on them.
What I need from you: Review the recipe and run the hosted build and authentication checks.
Refs #478 — partial: adds recipe coverage for
pam-u2fandpamu2fcfg; the other outstanding outputs, publication, dependency resolution and authentication proof remain open. Consumer tracking: projectbluefin/utah#389.pam-u2f1.4.0-9 at commit5b713d39c3a349036b81234c96c7a83cfcd1bdbc, retaining its signing keyring, signature verification and tests. The recipe produces both requested binaries.6ee4c5538a3b290748fae3b8c285b20a784fe2c0fdeff3a0bf250ec9b03659c6ddda4ea2d668c93d606f810e4f68cf3fc4130584a8532db9c9833a684d78f51f. The detached signature matches the SHA-512 insourcesand verifies with the imported keyring; the key's identity was not independently verified.libfido2,libfido2-devel,pamandpam-devel; this is metadata evidence, not a dependency solver transaction. Factory imagesha256:20af10a9313f5aaaebbba78ffee1fc4fe2a63e6bd1fb69497cb333cb929318ec(revision4e8f6e81a61a821406291d4f7d2a51f74011916c, run 40667554565) does not contain either requested binary.Validation:
python3 tools/source_pipeline.py pam-u2f: archive and bundled signature accepted against SHA-512; the signature downloaded directly from Yubico matches and verifies against the bundled keyring.python3 tools/factory_build_backlog.py --check,just check, andjust testpassed; each test run reported 1,390 passed, 2 skipped and 2,833 subtests passed.git diff --checkpassed.pre-commitwas unavailable.pamu2fcfginvocation and real security-key authentication remain unverified. No runtime completion is claimed.— hive: backend=codex