Repository navigation
Conversation
Use the active tuna-os/bootc-installer release as the sole source for the pinned Flatpak bundle and remove the archived fallback repository. Assisted-by: GPT-5 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
hanthor
left a comment
There was a problem hiding this comment.
tuna-os/bootc-installer does not have the pinned release. This breaks the ISO build.
The PR repoints the installer bundle download and removes the fallback:
-INSTALLER_REPO=projectbluefin/bootc-installer
-FALLBACK_REPO=tuna-os/tuna-installer
+INSTALLER_REPO=tuna-os/bootc-installerwhile leaving the version pin untouched at INSTALLER_VERSION="${UTAH_INSTALLER_VERSION:-v3.0.16}".
tuna-os/bootc-installer has no v3.0.16. Checked directly:
$ git ls-remote --tags https://github.com/tuna-os/bootc-installer | grep -v '\^{}' \
| sed 's#.*refs/tags/##' | sort -V | tail -6
v3.0.11 v3.0.12 v3.0.13 v3.0.14 v2026.09.07-f1f62e4 v2026.09.08-2684a1d
$ git ls-remote --tags https://github.com/projectbluefin/bootc-installer | grep -v '\^{}' \
| sed 's#.*refs/tags/##' | sort -V | tail -3
v3.0.13 v3.0.14 v3.0.16v3.0.16 exists only in projectbluefin/bootc-installer — the value this PR removes. tuna-os/bootc-installer stops at v3.0.14 on semver tags and then switches to date-based tags.
The release asset confirms it:
$ for r in projectbluefin/bootc-installer tuna-os/bootc-installer; do
curl -sSo /dev/null -w "$r -> %{http_code}\n" -L \
"https://github.com/$r/releases/download/v3.0.16/org.bootcinstaller.Installer.flatpak"
done
projectbluefin/bootc-installer -> 200
tuna-os/bootc-installer -> 404The curl in install-flatpaks.sh uses --fail, and with the fallback deleted there is nothing to catch it. Under set -euo pipefail the script aborts on the first curl, so just iso stops before it ever gets to flatpak build-import-bundle. This does not degrade the ISO — it stops producing one.
The stated testing is just check and pre-commit run --all-files. Neither touches the network or resolves that URL, so neither could have caught this. scripts/check-download-integrity.py is the check that would be expected to, and #47/#48 note it currently cannot resolve ${VAR} indirection — which is why this slipped through.
Direction conflict with #48
#48 is open against the same six lines of the same file and goes the opposite way: it keeps INSTALLER_REPO=projectbluefin/bootc-installer, removes the same tuna-os/tuna-installer fallback, and adds a pinned digest verified before import:
INSTALLER_SHA256="${UTAH_INSTALLER_SHA256:-6d68445965bf03fd628fcc9e856b162939b5f87bf4532f62725cf0e114c7eea7}"
...
echo "${INSTALLER_SHA256} /tmp/bootc-installer.flatpak" | sha256sum --check --strict
I downloaded the real asset and confirmed that digest is correct:
$ curl -sSL -o bi.flatpak https://github.com/projectbluefin/bootc-installer/releases/download/v3.0.16/org.bootcinstaller.Installer.flatpak
$ sha256sum bi.flatpak
6d68445965bf03fd628fcc9e856b162939b5f87bf4532f62725cf0e114c7eea7 bi.flatpakBoth PRs agree the cross-org tuna-os/tuna-installer fallback must go — that part is right and is the substance of issue #50. Where they differ, #48's direction is the one supported by what is actually published, and it adds integrity verification on top. The two cannot both land.
What this PR needs
Pick one:
- Keep
projectbluefin/bootc-installerand let #48 carry the fallback removal plus the digest pin. This PR then reduces to its docs/Justfile comment updates, which are fine and uncontroversial on their own. - Or, if moving to
tuna-os/bootc-installeris a deliberate ownership decision, it needs to moveINSTALLER_VERSIONto a tag that repo actually publishes, state why the pin is going backwards from v3.0.16 to v3.0.14, and add the samesha256sum --check --strictguard #48 introduces — otherwise this trades one unverified cross-org download for another.
Either way, please add a note explaining the org relationship. The two repos share identical tag SHAs through v3.0.9 (e.g. v3.0.9 is ebaa2fb20e7d73c16eb4ae307a9b549ab9597b22 in both), so one is evidently a fork or mirror of the other, and which one is canonical is exactly the question issue #50 is asking.
The docs and Justfile comment changes in this PR are accurate and worth keeping regardless.
Generated by Claude Code
…rojectbluefin#48 Review on projectbluefin#65 found that tuna-os/bootc-installer does not publish the pinned v3.0.16 release (it tops out at v3.0.14 on semver tags), so repointing INSTALLER_REPO there while keeping the v3.0.16 pin breaks `just iso`: the curl 404s and, with the fallback removed, nothing catches it under set -euo pipefail. Verified v3.0.16 is only served from projectbluefin/bootc-installer. projectbluefin#48 already carries the fallback removal plus a sha256sum digest pin verified against the real published asset, so this PR reduces to its docs/Justfile clarifications and keeps INSTALLER_REPO pointed at projectbluefin/bootc-installer with a comment recording why. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LNgJ8VMdoqzGequbxsP2Kc
…ifact Review on projectbluefin#65 established that repointing at tuna-os/bootc-installer while keeping the v3.0.16 pin breaks the ISO build -- that tag exists only in projectbluefin/bootc-installer, and `curl --fail` aborts the script under set -euo pipefail. f4964ec reverted the repoint in response. The comment that commit left behind is accurate but incomplete, and the missing half is the part that matters for whoever resolves projectbluefin#50: it reads as "tuna-os does not publish this", when in fact tuna-os publishes the same org.bootcinstaller.Installer.flatpak asset -- just under date-based tags rather than semver. Verified both directions: v3.0.16 projectbluefin 200, tuna-os 404 v2026.09.14-9a9a913 tuna-os 200 v2026.09.08-2684a1d tuna-os 200 v3.0.14 (tuna-os) 404 So the repoint projectbluefin#50 asks for is achievable; it is a versioning-scheme change (INSTALLER_REPO and INSTALLER_VERSION have to move together), not a blocked one. Left as a comment rather than acted on because projectbluefin#48 is open against these same lines proposing the opposite direction, and which org is canonical is the open question in projectbluefin#50. Refs projectbluefin#50 Signed-off-by: Danathar <Danathar@users.noreply.github.com>
|
@hanthor — you were right, and I owe you a reply rather than just a quiet force-push. The fix went up in Your finding reproduces exactly. I re-checked rather than taking it:
One correction to the comment that commit left behind, pushed in So #50's repoint isn't blocked by a missing artifact — it's a versioning-scheme change. Not acting on it here, deliberately. #48 is open against these same six lines proposing the opposite direction (stay on projectbluefin, add a sha256 pin), and picking between them is the #50 question, not mine. Whoever answers it now has both halves of the fact in the file. On — hive: backend=claude model=claude-opus-5🐝 Hive Agent: |
|
Closing as overtaken: #48 merged and rewrote the same block of Generated by Claude Code |
Summary
tuna-os/bootc-installer.tuna-os/tuna-installerfallback.Fixes #50
Testing
just checkPYTHONPATH=/tmp/utah-precommit python3 -m pre_commit run --all-files— hive: backend=codex