Repository navigation
Bug hunt ledger: Bundler (RubyGems) #316
Replies: 99 comments
|
[agent] 2026-09-30: Bundler (RubyGems) bug-hunt run This is the first run, so the ledger started empty and there were no Tested: main Setup: the sandbox can't reach the Socket patch API. I used a temporary, uncommitted copy of Cells
Issues
False positives ruled out
Next
|
|
[agent] 2026-09-30: Bundler (RubyGems) bug-hunt run Tested: main Setup: a temporary, uncommitted copy of Cells
Issues
False positives ruled out
Next
|
|
[agent] 2026-10-01: maintainer note: test global ( This is a maintainer request, not a run report. Add it to the top of the backlog and keep it there until the cells below are covered. Ask: make sure we correctly scan global installs when Where Bundler (RubyGems) puts global installs: What to check (prove each with a real global install, not by reading source):
Add OS × Bundler (RubyGems) version cells for |
|
[agent] 2026-10-01: Bundler (RubyGems) bug-hunt run Tested: main Setup: for agent and Re-triage
Cells (global mode, maintainer checklist)Probe run: https://github.com/SocketDev/socket-patch/actions/runs/36816092864 (ubuntu, macos and windows × Ruby 2.7 / 3.3 / 3.4).
Other cells
Issues
False positives ruled out
Blocked
Next
|
|
[agent] 2026-10-01: handover from the Composer bug-hunt routine (lead, unverified for gem) While I was confirming #438 (Composer, Windows), the root cause turned out to be Composer evidence: #438 and the probe run https://github.com/SocketDev/socket-patch/actions/runs/36827949605 |
|
[agent] 2026-10-01: Bundler (RubyGems) bug-hunt run Tested: main Setup: a hold-open copy of Re-triage
Cells
Issues
False positives ruled out
Blocked
Next
|
|
[agent] 2026-10-01: Bundler (RubyGems) bug-hunt run Tested: main Setup: a copy of Re-triage
Cells
Issues
False positives ruled out
Next
|
|
[agent] 2026-10-02: Bundler (RubyGems) bug-hunt run Tested: main Setup: uncommitted probe tests added to a copy of Re-triage
Cells
Issues
False positives ruled out
Next
|
|
[agent] 2026-10-02: Bundler (RubyGems) bug-hunt run Tested: main Setup: uncommitted probes in copies of Re-triage
Cells
Issues
False positives ruled out
Next
|
|
[agent] 2026-10-02: Bundler (RubyGems) bug-hunt run Tested: main Setup: uncommitted probe tests in a copy of Re-triage
Cells
Issues
False positives ruled out
Next
|
|
[agent] Janitor: ledger drift. The coverage matrix still lists these issues as
This is a heads-up only. The janitor never edits ledgers. Generated by Claude Code |
|
[agent] Janitor: ledger drift. This ledger still lists these issues as failing, but they are now closed:
Please re-check them and update the matrix on your next run. Generated by Claude Code |
|
[agent] 2026-10-02: Bundler (RubyGems) bug-hunt run Tested: main Setup: a ~30-line Python mock of the patch API serving Re-triage
Cells
Issues
False positives ruled out
Probe branches
Next
|
|
[agent] 2026-10-03: Bundler (RubyGems) bug-hunt run Tested: main Setup: a temporary Re-triage
Cells
Issues
False positives ruled out
Next
|
|
[agent] 2026-11-13: Bundler (RubyGems) bug-hunt run Tested: main Harness: the sandbox reaches rubygems.org but not the Socket patch API, so I used a standalone mock (Python) of the patch API and the patch registry next to the real rubygems.org: real Re-triage: nothing to do. #1109 is Cells
Filed
Commented / closed: nothing. Ruled out
Not filed, worth a look later: the |
|
[agent] 2026-11-14: Bundler (RubyGems) bug-hunt run Tested: main Harness: a standalone Python mock of the patch API and registry next to the real rubygems.org, with real Re-triage: #2069 and #2070 are open and claimed; main has not moved, nothing to re-check. #1109 is Cells
Filed
Commented / closed: nothing. Ruled out
Not filed, worth a look later: |
|
[agent] 2026-11-14: Bundler (RubyGems) bug-hunt run Tested: main Harness: the standalone Python mock again (patch API on Re-triage: #2069 still reproduces on Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-15: Bundler (RubyGems) bug-hunt run Tested: main Harness: the standalone Python mock again, rebuilt from the state file's notes ( Re-triage: main has not moved; #2069, #2070, #2075, #2076, #2083 and #2084 are open and unchanged, nothing to re-check. #1109 is Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-15: Bundler (RubyGems) bug-hunt run Tested: main Harness: a Re-triage: main has not moved, so nothing to re-check. #1109 is Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-16: Bundler (RubyGems) bug-hunt run Tested: main Harness: the standalone Python mock again, with real Re-triage: main has not moved, so nothing to re-check. #1109 is Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-16: Bundler (RubyGems) bug-hunt run Tested: main Harness: the standalone Python mock, rebuilt from the state file's notes, with Re-triage: main has not moved, so nothing to re-check. #1109 is Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-17: Bundler (RubyGems) bug-hunt run Tested: main Harness: no mock this run. Agent records were written by hand ( Re-triage: not done. #2083 and #2086 (the two oldest without Cells (agent mode,
Filed
Commented / closed: nothing. Found, not filed (3-issue cap; first backlog item after re-triage)
Ruled out
|
|
[agent] 2026-11-17: Bundler (RubyGems) bug-hunt run Tested: main Harness: the standalone Python mock rebuilt from the state file's notes ( Re-triage (the two oldest without
Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-18: Bundler (RubyGems) bug-hunt run Tested: main Harness: a Python mock of the patch API and the patch registry (13 gems), plus Re-triage
Cells
Filed
Commented / closed: nothing. Ruled out
|
|
[agent] 2026-11-18: Bundler (RubyGems) bug-hunt run Tested: main Harness: the Python mock of the patch API and registry (14 gems for the Rails 8 app), restarted on the same port with new uuids for superseding patches; Re-triage: main has not moved since the last run re-checked #2086 (still reproduces) and #2087 (claimed), so neither was re-run and nothing was commented. Cells
Filed
Commented / closed: nothing. Ruled out
Blocked: Ruby 3.4 and 4.0 are untested. The sandbox has 3.1 to 3.3, |
|
[agent] 2026-11-19: Bundler (RubyGems) bug-hunt run Tested: main Harness: a Python mock of the patch API and the patch registry (rack 2.2.8 / 2.2.25, puma, rack-test, base64, json), plus a second compact index standing in for a private gem server. Re-triage
Cells
Filed
Commented / closed: nothing. Ruled out
Blocked: Ruby 3.4 and 4.0 are still untested (no such Ruby in the sandbox). |
|
[agent] 2026-11-19: Bundler (RubyGems) bug-hunt run Tested: main Harness: a Python mock of the patch API and the patch registry (rack 2.2.8, puma 6.4.3, rack-test 2.1.0, base64 0.2.0, json 2.7.2), plus a compact index standing in for a private gem server ( Re-triage: main has not moved since the last run re-checked #2086 and #2106 (both still reproduce), so neither was re-run and nothing was commented. Cells
Filed
Commented / closed: nothing. Ruled out
Blocked: Ruby 3.4 and 4.0 are still untested (no such Ruby in the sandbox). |
|
[agent] 2026-11-20: Bundler (RubyGems) bug-hunt run Tested: main Harness: the repo's own stand-in for the patch service, Re-triage
Cells (all on a precompiled nokogiri unless noted)
Filed
Commented / closed: one comment on #2174 ( Ruled out
Blocked: agent mode beside a platform gem (needs hand-written blobs; next run). Ruby 3.4 / 4.0 and JRuby: none in the sandbox. |
|
[agent] 2026-11-20: Bundler (RubyGems) bug-hunt run Tested: main Harness: one loopback server in front of several instances of the repo's stand-in ( Re-triage: main had not moved since the last run re-checked #2152, and no commit touches #2106, so neither was re-run and nothing was commented. Cells (precompiled platform gems, backlog item 1)
Filed
Commented / closed: nothing. Ruled out
Blocked: #2174's fix on |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
[agent] Progress ledger for the scheduled Bundler (RubyGems) bug-hunt routine (label pm:bundler).
Last run: 2026-11-20 (second run), main
6579162(6af8cd6, the #2174 fix, landed during the run and is untested), release 5.1.0, newest Bundler 4.1.0 stable (no newer prerelease). Tested so far: Bundler 1.17.3, 2.3.27, 2.4.19, 2.5.22/23, 2.6.9, 4.0.18, 4.0.24, 4.1.0.beta3, 4.1.0 on Ruby 3.1.6, 3.2.6, 3.3.6.Harness notes: the sandbox reaches rubygems.org but not the Socket patch API. Use the repo's own stand-in,
tests/nokogiri/patch_service.rb:PatchService.new(name, version, platforms, uuid, ghsa, workdir, mark)downloads the real gems, appendsmarktolib/<name>.rb, and serves the patches API, the compact index and the artifacts for ONE gem on a loopback port (orgnokogiri-e2e). An empty platform list gives a ruby-platform patch; vendored mode then also needs agem-stub-gemspecartifact (add it topackage_resultand serve the gem'sspec.to_ruby). For several gems, a superseding patch or agent mode, put one loopback server in front of several instances: override each one'surlwith the front's, send a path that holds a uuid to that instance'srespond, mergebatchandpackage, and serve/patches/blob/<git-sha256>from its@before/@after. Pass--api-url <mock> --patch-server-url <mock> --api-token fake --org nokogiri-e2eto every command. Git-ignore**/vendor/bundle/in test repos. Rubies are in/opt/rbenv/versions(3.1.6: Bundler 2.3.27, 2.6.9; 3.2.6: 2.4.19, 4.0.18; 3.3.6: 2.5.22, 4.0.18, 4.1.0);gem install bundler -v <v>works. Select a Bundler withBUNDLER_VERSION. A realistic project:gem install rails -v 8.0.2into a scratchGEM_HOME, thenrails new app --skip-git --skip-bundle --skip-javascript --skip-hotwire --skip-docker. A release build takes over 10 minutes.vexwrites with--output(-ois--org);removetakes one purl. No Ruby 3.4 / 4.0 here (ruby-build's downloads are blocked).Coverage matrix
Linux only. A cell is
passwith the Bundler versions and the date,fail #N,n/a, oruntested.bundle install, app loads patched code,vex, rollback,bundle installagainupdate --all, rollback byte-exact (11-18); refusals name their remedy (#2091 fixed, 11-19)?platform=rubypurl, fail #2182BUNDLE_PATH)GEM_HOME(11-15); native rebuild and rollback byte-exact in a sharedGEM_HOME(11-17); two projects with their own records for one native gem (11-18)scanandrollbackwith--gem-refresh-shared(11-15); two projects on one gem (11-18); precompiled nokogiri on 2.5.22 with the printedrm … && bundle installremedy (#2086 fixed, 11-20)bundle config set path vendor/bundleBUNDLE_DEPLOYMENT,BUNDLE_FROZEN,.bundle/config)GEM_HOME=BUNDLE_APP_CONFIG, deployment,without) 4.0.18 (11-17)BUNDLE_WITHOUTtoo, 4.0.18 (11-14)-x86_64-linux)vendor/cache:pristinethenapply, clone, rollback 4.0.18 (11-20)vendor/cache), 2.5.22, 4.0.18, 4.1.0; ffi and sqlite3 pass 4.0.18, superseding patch too: frozen install, deployment clone, lock stable,vex,update --all,add, rollback byte-exact; a missing prebuild and arubyentry beside the builds are refused with remedies that work (11-20); the refused gem is missing from the closing totals, fail #2181vendor/cache,rubyentry beside the builds); ffi and sqlite3 pass 4.0.18: superseding patch, one-gem rollback,pristine;cache_all_platformsarchives swapped and restored,install --localclone; per-gemforce_ruby_platform: truebuilds the patched ruby gem; rollback byte-exact (11-20); a missing prebuild is refused and--remove-platformfixes it, but the closing totals say "Nothing left to patch", fail #2181; standalonevexwas fail #2087 (fixed, not re-run)vex,pristinethen re-apply, rollback byte-exact; a killedapply,--offlineand no compiler leave the gem intact. #2104, #2105, #2117 fixed (not re-run); argon2 never rebuilds, fail #2106Bundler.requireloads it was fail #2084 (fixed on3b14afc, not re-run)bundle update/bundle add/bundle remove/bundle outdatedafter patching--redownloadthenapply(11-14)cache(11-14)lockflags,update <other gem>,clean,doctor,check4.0.18 (11-16);outdated,update --allon the Rails 8 app (11-18; #2070 fixed)GemfileandGemfile.lockbyte for byteremove,vendor --revertpass 4.0.18 (11-14)vendor/cachearchives too (11-16);removeof a transitive and a direct gem out of 12 (11-18); nocache/dir 2.6.9 (#2092 fixed),--add-checksumson a patched lock (11-19)vexreports only what is really appliedbundle pristinetoo (11-13)vendored_tree_out_of_syncwas fail #2076 (fixed on3b14afc, not re-run)gems.rb/gems.lockedspellinggemfile_not_loaded), not re-run--cwdto use (#2125 fixed, 11-19)list, rollback from the root (11-16)list, rollback from the root (11-14); two locks with different platform sets 2.5.22: one wired, one refused, remedy, rollback from the root (11-20)gemspecin the Gemfile), runtime and development dependencies patchedsource … doblock: pass 4.0.18, app,vex, rollback tree byte-exact (11-19)bundle lockstable,vex, rollback (11-19)sourcelines (Bundler 2.x): the warning was fail #2162 (fixed, not re-run)vex, rollback tree byte-exact (11-19)vendor/bundleand an emptyGEM_HOME,vex, rollback (11-19)gemfiles/*.gemfile(BUNDLE_GEMFILE)BUNDLE_GEMFILE; a root scan wires the root pair only and the appraisal lock still installs (11-19)setupand the Bundler plugin it installsgem_bundler_plugin_staledetection is left, untested=begin,__END__, …)--jsonoutputpristinepasses (#2083 fixed, 11-18);get <purl> --mode vendoredover one of three hosted gems, thenremove,vendor --revert,rollback --offlinepass 4.0.18 (11-18); the same for a precompiled platform gem was fail #2174 (fixed on6af8cd6, not re-run); vendored to hosted is refused with a remedy (11-20)vex, rollback pass (11-18); #2090 fixed, not re-runOVERRIDESlock, and on RubyGems 4.1 with a native rebuild (11-16)OVERRIDESlock and the new default lock that lists therubyvariant beside platform builds (11-16)OVERRIDES,global_gem_cache,cooldown, RubyGems 4.1 native refresh (11-16);CONTENT ADDRESSESuntestedGEMsection, deployment clone,update(11-17)Backlog
vendor/bundle, andbundle pristinethen fails on the restored checksum #2174's repro on6af8cd6or later (hosted, then vendored, then rollback, on a precompiled nokogiri).1b. Precompiled platform gems, the rest: Bundler 2.6.9 and 4.1.0 for ffi / sqlite3,
force_ruby_platformas a config under a platform-only lock, ajava/ JRuby lock if a JRuby appears,vendor_prebuilt_platform_pendingon a gem that is already vendored, google-protobuf (its main file is notlib/<name>.rb, so the stand-in needs a tweak).SOCKET_RUBYGEMS_URL, a private source, a Rust (rb_sys) gem, binary-only platform gems (sass-embedded), eventmachine (C++), a.h-only patch; Never take a native gem build that compiled nothing of a patch for a patched build; test agent mode on real nokogiri #2136 on real nokogiri with--use-system-libraries.--dry-runpreviews success #2129, Hosted gem scan sends a gemspec-declared gem to--mode vendored, which refuses it for the same reason and names no remedy #2151, Hosted gem scan on a Gemfile with two top-levelsourcelines blames a "bundler < 2.2" lock and prescribesbundle lock, which changes nothing #2162, then Standalonevexrefuses a hosted gem asnot_appliedbecause of a leftover<gem>-<version>-x86_64-linuxdirectory that has no gemspec, after the scan's ownrm -f … && bundle installremedy #2087, Hosted gemrollback/removeleave the patched gem installed when the gem home has nocache/dir (Bundler 4.1prune,rm -rf ruby/*/cache), and the "next package-manager install" note does not fix it #2092 under Bundler 4.1prune, Agentapplybuild failure for a precompiled gem (pg without libpq-dev) cuts off the cause, points at anmkmf.logit has deleted, and says to install a compiler that is already there #2117, Vendored gem scan exits 1 with "Gemfile.lock GEM specs has no entry" for a gem in the shared gem home that the project does not lock, where hosted mode skips it with a warning #2075,vexwarnsvendored_tree_out_of_syncfor every vendored gem afterbundle install, and the remedy it names never clears it because Bundler keeps the unused registry copy #2076, Vendored mode declares a transitive gem withoutrequire: false, soBundler.requirestarts loading it at boot (hosted mode adds the option) #2084, Reverting a vendored gem while another gem is hosted puts its spec in the hosted gem'sGEMsection: freshbundle installfails androllbackcan no longer undo either gem #2090, Agentrollbackof a gem that builds its library in its own tree (scrypt, ffi-compiler gems) restores the source but leaves the patched binary loaded, and no command lists it afterwards #2104, Agentapplyreports a digest-crc C patch applied andvexattests it, but the gem's Rakefile build is a no-op and Ruby keeps loading the unpatched extension #2105.scan,list,show,get,rollbackandvexon the common path as a first-time user;--jsonshapes unchanged.sourceblock (same name and version as a public gem), in all three modes; a default gem in the Ruby's own gem home (noGEM_HOME); an Appraisal lock in agent mode;bundle install --standalone; a scoped mirror (BUNDLE_MIRROR__HTTPS://RUBYGEMS.ORG/) under hosted mode.get(notscan) onto a newer agent patch; a superseding agent patch for a gem with a native extension.scan --mode vendoredexits 1 whenever one gem is refused, with next steps that read like success; a rollback of a vendored gem prints "1 unwired package keeps its patched bytes in installed trees" though the installed copy was never patched (seen again 11-20);remove <gem>from a monorepo root doesn't name the sub-projects that pin it;docs/usage.mdnever mentions Ruby; theredirect_gem_platform_unsupportedwarning names no remedy itself (the docs do); a rollback after agent then hosted downloads the before blob though the file is already original, and leaves an empty.socket/manifest.json.scan --prune;apply --dry-runstarts no Ruby; a relative crawl root (Rebuild gem native extensions from a relative crawl root #2048)./opt. Bundler 4.1CONTENT ADDRESSES.socket-patch rollback pkg:gem/<gem>@<version>?platform=rubyunquoted, which zsh rejects and a nullglob bash turns into a barerollback#2182, Hosted gemrollback/removesays it rebuilds a native gem "from its patched gem" while it installs the upstream build #2152 (reproduced on6579162), Agentapplycan never rebuild argon2, the gem the docs name as rebuilt from its installed tree: its one-shotextconf.rbfails onMakefile.real, and the remedy says to install a compiler #2106 (no commit touches it).Known non-bugs
The full list is in
known-non-bugs/bundler.mdon this branch. Search it withgit grepbefore filing, and append one line when you rule something out. Don't read it whole. Most likely to be re-filed:gem buildhas a different gemspec. Repack with the originalmetadata.gz.vex/listreport no patches on a hosted project whose patch URLs are on loopback: pass--patch-server-url.vexexits 2 asking for a product when the project has no git origin, gemspec or other manifest: pass--product.All reactions