Skip to content

fix(packages): record nss-mdns .local resolution parity gap as unavailable - #569

Merged
castrojo merged 2 commits into
projectbluefin:mainfrom
kylerankin:fix/nss-mdns-parity
Oct 4, 2026
Merged

castrojo merged 2 commits into
projectbluefin:mainfrom
kylerankin:fix/nss-mdns-parity

Conversation

@kylerankin

@kylerankin kylerankin commented Oct 3, 2026 •

Copy link
Copy Markdown
Contributor

Closes #499.

Bluefin ships nss-mdns (baselines/bluefin/rpms.tsv, 0.15.1-28.fc44) so
ping foo.local resolves over mDNS. Utah does not, and Hummingbird's
repository does not carry it either -- it appears in that repodata only as a
Recommends of cups (a virtual provide), never as an installable <name>.
Enabling avahi-daemon restores mDNS service browsing but not .local name
resolution, which is what this module provides; the two are separate (the
issue is out-of-scope for the automated gap check per #411).

Records nss-mdns as an image-level parity gap in packages/utah.toml
[unavailable]. No baselines/triage.toml entry is added: nss-mdns ships
no paths in baselines/bluefin/surface.tsv, so scripts/image-baseline.py
never reports it as a gap and a triage entry would be flagged as stale
("no longer gaps, drop from triage.toml").
Bumps the genuinely-unavailable count 7 -> 8 and syncs the README parity table,
site/data/packages.json (regenerated), and docs/skills/package-contract.md
so just check stays green.

Files

  • packages/utah.toml: prose comment for the gap + nss-mdns in the [unavailable] list
  • README.md: "Genuinely unavailable" 7 -> 8
  • site/data/packages.json: regenerated (python3 scripts/generate-site-data.py)
  • docs/skills/package-contract.md: count 7 -> 8; nss-mdns named alongside nvtop

Verification

  • python3 scripts/check-doc-counts.py: 61 Bluefin contract, 105 Utah additions, 8 genuinely unavailable
  • python3 scripts/generate-site-data.py --check: current
  • python3 scripts/image-baseline.py check: 281 gaps, all triaged
  • TOML validity and 102-script syntax gate green

— hive: backend=pi model=lemonade/Ornith-1.5-35B-A3B-GGUF-Q6_K


🐝 Hive Agent: contributor | SHA: ee97ce8

…lable

Bluefin ships nss-mdns (baselines/bluefin/rpms.tsv, 0.15.1-28.fc44) so
`ping foo.local` resolves over mDNS; Utah does not, and Hummingbird's
repository does not carry it either -- it appears in that repodata only as
a Recommends of cups (a virtual provide), never as an installable <name>.
Enabling avahi-daemon restores mDNS service browsing but not .local name
resolution, which is what this module provides (the two are separate;
reviewed against projectbluefin#411).

Records nss-mdns as an image-level parity gap in packages/utah.toml
[unavailable] and baselines/triage.toml ([package.nss-mdns], status
unavailable, area network-fs, issue projectbluefin#499), mirroring the nvtop precedent.
Bumps the genuinely-unavailable count 7 -> 8 and updates the README parity
table, site/data/packages.json (regenerated), and docs/skills/package-contract.md
in lockstep so just check stays green.

Hive-Run: projectbluefin#499
Hive-Plan: record nss-mdns as an unavailable image-level parity gap in utah.toml and triage.toml, bump the unavailable count 7 to 8 and sync README, packages.json, package-contract.md
Hive-Spec: package-contract.md#[unavailable]
Signed-off-by: hive <hive@kubestellar.io>

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

correctness

  • baselines/triage.toml:1028 adds [package.nss-mdns], but scripts/image-baseline.py:126-133 derives gaps from baselines/bluefin/surface.tsv only, and nss-mdns owns no surface path there (it ships /usr/lib64/libnss_mdns*.so; grep ^nss-mdns baselines/bluefin/surface.tsv is empty — unlike nvtop, surface.tsv:3464-3466). At head ee97ce8, python3 scripts/image-baseline.py check prints 281 gaps, all triaged; no longer gaps, drop from triage.toml: nss-mdns (exit 0). The PR body's "281 gaps, all triaged" omits that tail, and triage.toml:10 ("Drop an entry once it stops being a gap") says the entry should not exist. write_report() (image-baseline.py:156-160) iterates only found gaps, so the entry is also invisible in the gap report. Either drop the triage entry (the packages/utah.toml:400 [unavailable] list + site/data/packages.json already carry the debt), or, if triage tracking is the goal, extend surface extraction to cover NSS modules so the checker actually sees it.

docs-currency

  • docs/skills/package-contract.md:78 now cites nss-mdns alongside nvtop as an example of a gap "triaged in baselines/triage.toml" — per the finding above the triage checker never classifies nss-mdns as a gap, so the example is misleading; keep nvtop only if the triage entry is dropped (also "are the current example" → "examples").

Verified independently: Hummingbird repodata/primary.xml has 0 <name>nss-mdns</name> entries, matching packages/utah.toml:487-489.

No findings from: security, intent-alignment, style.

Confidence: 3/5 (needs attention) — 1 medium finding; 2 perspectives requested changes

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

nss-mdns ships no path in baselines/bluefin/surface.tsv, so
image-baseline.py check never derives it as a gap and flagged the
entry as one to drop. The parity debt stays recorded in
packages/utah.toml [unavailable] and site/data/packages.json.
Restore nvtop as the sole triage.toml example in package-contract.md.

Hive-Fix-Attempt: 1/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <scanner@hive.kubestellar.io>

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

intent-alignment

  • (low) PR body says it adds baselines/triage.toml [package.nss-mdns] (status unavailable, area network-fs, issue #499) "mirroring the nvtop precedent". The diff touches only README.md, docs/skills/package-contract.md, packages/utah.toml, site/data/packages.json; at head 638e2f6 baselines/triage.toml has no nss-mdns entry (the nvtop one is at baselines/triage.toml:761). CI does not notice because scripts/image-baseline.py:126-133 derives gaps from surface.tsv, where nss-mdns never appears, so check still prints "281 gaps, all triaged". Either drop that paragraph/bullet from the body (adding the entry would only make check emit "no longer gaps, drop from triage.toml: nss-mdns", image-baseline.py:205-206), or add it knowingly.

Verified at head: generate-site-data.py --check, check-doc-counts.py (8 unavailable), image-baseline.py check all green; current Hummingbird primary.xml.gz contains no <name>nss-mdns</name>, so the --check-unavailable probe (scripts/check-repo-availability.py:154-158) should report "still unavailable".

No findings from: correctness, security, style, docs-currency.

Confidence: 3/5 (needs attention) — 1 perspective requested changes

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@Danathar Danathar left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Read through. Records nss-mdns as unavailable with the reason and tracking issue #499, and the counts in README, package-contract.md and packages.json move together. CI green.

@castrojo
castrojo merged commit ce484fa into projectbluefin:main Oct 4, 2026
14 checks passed
castrojo added a commit to arteeh/utah that referenced this pull request Oct 6, 2026
Request the Intel VA-API iHD driver and both Media SDK/oneVPL runtime
generations (intel-gmmlib, libva-intel-media-driver, intel-mediasdk,
intel-vpl-gpu-rt, libvpl) in [parity], and replace the
[multimedia_overrides] narrative with the media package scope and the
remaining projectbluefin#383 gaps.

Rebased onto main as a single commit: PR projectbluefin#397's eight commits (including
two merge-integration commits) are squashed and the conflicts with main's
bluez-cups/projectbluefin#403/projectbluefin#569 changes are resolved, keeping both sides. The libva,
libheif and mesa-* names are described as not requested rather than as
Fedora's builds, and the counts and site data are regenerated.

Refs: projectbluefin#383
Co-authored-by: arteeh <arteeh@users.noreply.github.com>
Co-authored-by: castrojo <castrojo@users.noreply.github.com>
Co-authored-by: scanner <scanner@hive.kubestellar.io>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: arteeh <arteeh@users.noreply.github.com>
Signed-off-by: castrojo <castrojo@users.noreply.github.com>
hanthor pushed a commit that referenced this pull request Oct 6, 2026
Request the Intel VA-API iHD driver and both Media SDK/oneVPL runtime
generations (intel-gmmlib, libva-intel-media-driver, intel-mediasdk,
intel-vpl-gpu-rt, libvpl) in [parity], and replace the
[multimedia_overrides] narrative with the media package scope and the
remaining #383 gaps.

Rebased onto main as a single commit: PR #397's eight commits (including
two merge-integration commits) are squashed and the conflicts with main's
bluez-cups/#403/#569 changes are resolved, keeping both sides. The libva,
libheif and mesa-* names are described as not requested rather than as
Fedora's builds, and the counts and site data are regenerated.

Refs: #383

Signed-off-by: arteeh <arteeh@users.noreply.github.com>
Signed-off-by: castrojo <castrojo@users.noreply.github.com>
Co-authored-by: castrojo <castrojo@users.noreply.github.com>
Co-authored-by: arteeh <arteeh@users.noreply.github.com>
Co-authored-by: scanner <scanner@hive.kubestellar.io>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

agent/scanner Filed or owned by the scanner agent.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Review backlog (correctness): nss-mdns parity gap remains

3 participants