Repository navigation
Conversation
d253482 to
3462416
Compare
[scanner] CI root cause: every srpm job fails on a pruned packit image digest (
|
[scanner] CI root cause identified:
|
[scanner] CI red is systemic, two clusters: (1) all
|
|
[scanner] Root cause of the rebuild/srpm failures: RPM transaction conflict between ruby3.3/ruby3.4/ruby4.0 default-gems (file conflicts on /usr/share/gems/specifications/default/*). Unblocked by utah-packages#83 (exclude legacy ruby default-gems from Hummingbird repo) — after #83 merges, rebase this branch onto main and CI should clear. scanner agent (ACMM L5 — hold-gated mode)🐝 Hive Agent: — hive: agent=scanner backend=copilot model=kimi-k3 copilot=1.0.78 |
hanthor
left a comment
There was a problem hiding this comment.
The declared change is fine. The undeclared two-thirds of the diff is the problem — including one hunk that is arguably the most valuable fix in the whole open-PR queue and is buried where nobody will find it.
What the body describes vs. what the diff does
The body describes four things, all in the anaconda-live / runtime-contract area. Three-dot diff against main (54cc8d1):
config/runtime-contract.toml | 1 + <- declared
tools/import_bluefin_rawhide.py | 5 + <- declared
config/upstream-sources.json | 3 ++- <- NOT declared
packages/dracut/dracut.spec | 2 +- <- NOT declared
packages/fish/fish.spec | 9 ++- <- NOT declared
packages/mozjs140/mozjs140.spec | 5 ++ <- NOT declared
(reports/bluefin-rawhide-resolution.json and docs/skills/runtime-contract-exceptions.md are mentioned in the body but do not appear in the diff against current main.)
The fish hunk is the real fix for a failure three other PRs are arguing about
%check
export CI=1
-%cmake_build --target fish_run_tests
+if [ "$(id -u)" -eq 0 ]; then
+ useradd --system --create-home fish-tests
+ chown -R fish-tests:fish-tests %{_vpath_builddir}
+ runuser -u fish-tests -- env USER=fish-tests LOGNAME=fish-tests HOME=/home/fish-tests \
+ cmake --build %{_vpath_builddir} --target fish_run_tests
+else
+ %cmake_build --target fish_run_tests
+fiThis is correct, and I can show it is aimed at the right target. From the fish job on the latest scheduled run of unmodified main (run 34571911239 @ 54cc8d1, job 103175696704):
193 / 196 passed (44 skipped)
Failed tests:
tests/checks/cd.fish
tests/checks/fish_config.fish
tests/checks/path.fish
FAILED: [code=1] CMakeFiles/fish_run_tests ...
error: Bad exit status from /var/tmp/rpm-tmp.56yZAa (%check)
Three root-sensitive tests, everything else green. Your hunk fixes exactly that. Meanwhile #79 is open claiming fish fails in the buildroot for want of crate RPMs, rewrites the spec to vendor ~120 crates, does not touch %check at all, and removes fish from [unavailable] on the strength of it. #79 will not turn that job green; this hunk will. But because it is unmentioned here, #79's author had no way to know it existed.
The mozjs140 hunk is in three PRs at once
Byte-identical in #56, #57 and #58, and it is the headline claim of #30:
+ %if "%{toolchain}" == "clang"
+ export CC="/work/tools/sccache clang"
+ export CXX="/work/tools/sccache clang++"
+ %else
Four PRs, one fix. Someone needs to own it.
The dracut changelog hunk is shared with #56 and fixes a real error
-* Mon aug 10 2026 Hans de Goede <johannes.goede@oss.qualcomm.com> - 111-2
+* Mon Aug 10 2026 Hans de Goede <johannes.goede@oss.qualcomm.com> - 111-2Also byte-identical in #56. And it is not cosmetic — the dracut build log on main (job 103175696635) emits:
error: bad date in %changelog: Mon aug 10 2026 Hans de Goede <johannes.goede@oss.qualcomm.com> - 111-2
right before the buildroot transaction. Worth landing on its own with that log line as justification, rather than riding along unexplained in two different PRs.
The evolution-ews stage bump directly contradicts #70
"filename": "evolution-ews-3.61.3.tar.xz",
- "sha512": "617369f9...cf813"
+ "sha512": "617369f9...cf813",
+ "stage": 2evolution-ews requires libebook/libecal/libedata-* >= 3.61.3, which come from evolution-data-server. On main that builds at stage 1. This PR moves evolution-ews to stage 2 — one stage later, which is the minimum that could work. #70 moves evolution-data-server to stage 9 and evolution-ews to stage 10 instead. Those two are incompatible orderings of the same dependency, sitting in two PRs neither of which mentions the other. If this lands first and #70 later, the merge resolves silently to whichever side git prefers and the stage graph breaks in a way that only shows up hours into a matrix run. evolution-ews is currently a confirmed failure on main (job 103175696652), so this matters.
The declared change itself is good
anaconda-live in [unavailable] plus the [unavailable] filter in import_bluefin_rawhide.py is the right shape — it keeps a package Utah does not use from both the runtime transaction and the Rawhide import, and it uses the mechanism the repo already has rather than inventing one. I confirmed the effect: merged into main, the contract drops from 72 to 71 packages, and all gates stay green:
PR#57 fc=0 val=0 pytest=0 | validated 193 source RPMs | 72 passed in 0.66s
PR#57 quoting=0 runtime_contract=0 unittest=0 | validated runtime contract: 71 packages
What I would ask for
- Split the
fish%checkhunk into its own PR, with the log excerpt above as the justification, and cross-link it from #79. It is valuable and it is currently invisible. - Drop the mozjs140 hunk here; let one PR own it.
- Split the dracut changelog fix out (or coordinate with #56, which has the same line).
- Resolve the
evolution-ewsstage question with #70 before either merges. - Leave this PR as the anaconda-live change it says it is — that part is ready.
Generated by Claude Code
|
[scanner] CI root cause: both failure classes here are systemic, not from this PR.
Once #83 and #87 merge and this branch is rebased, CI should validate this PRs actual changes. Note: the hive App token currently lacks the scanner agent (ACMM L5 — hold-gated mode)🐝 Hive Agent: — hive: agent=scanner backend=copilot model=kimi-k3 copilot=1.0.78 |
|
[scanner] Root-cause triage of the red checks — none are caused by this PR's diff:
Merging #85+#83+#81 and re-running should clear this PR without changes here.🐝 Hive Agent: — hive: agent=scanner backend=copilot model=kimi-k3 copilot=1.0.78 |
Signed-off-by: sec-check[bot] <sec-check[bot]@users.noreply.github.com>
8ba2cee to
029b00f
Compare
…e import, and contract (#71, #72, #56, #57) (#92) * test(quality): unit coverage for validate.py and recalculate_hummingbird_gaps.py * fix(ci): verify cosign binary sha256 checksum in rebuild-rpms Verify the downloaded cosign v2.4.1 binary against its release SHA-256 checksum prior to installation and keyless signing. Closes #46 Closes #66 Closes #68 * fix(tools): validate package and branch inputs in import_rawhide (#56) * fix(contract): exclude anaconda-live from runtime transaction (#57) --------- Co-authored-by: castrojo <castrojo@users.noreply.github.com>
Utah uses
bootc-installer, not Anaconda WebUI or Slitherer (commit 68829f8).anaconda-liveinconfig/runtime-contract.tomlunder[unavailable]alongsideslitherer.[unavailable]packages intools/import_bluefin_rawhide.pyto prevent automatically importing unused installer stack sources.reports/bluefin-rawhide-resolution.json.docs/skills/runtime-contract-exceptions.mddocumenting the convention.