Repository navigation
Issue & discussion janitor log #576
Mikola Lysenko (mikolalysenko)
started this conversation in
General
Replies: 1 comment
|
[agent] Janitor: bridge test. The janitor/ledger workflow posted this comment on the routine's behalf. Hourly runs log here from now on. Generated by Claude Code |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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] Janitor: the issue and discussion janitor rewrites this log each run. It shows the last run, the actions it took with a reason for each, a rolling list of recent actions, deferred candidates, and anything that needs a human. The routine writes it to the
janitor/ledgerbranch, and a workflow on that branch applies it here.Last run
2026-11-19 13:22Z (Thursday) on origin/main
4fe08592. Reviewed the 17 merges since89c482d8(#2100, #2116, #2118–#2122, #2124, #2126–#2128, #2130, #2131, #2134, #2136, #2138, #2141), issues opened, updated or closed since 2026-11-18 13:00Z, claimed issues, and a fresh discussion snapshot (25 discussions, generated 13:23:05Z).This run
vexrefuses 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, Vendored gem scan on a stock Rails Gemfile fails 5 of 7 gems (group blocks, native extensions) and none of the errors says what to do, thoughgroup:at the top level and--mode hostedboth work #2091, 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, Vendoring a gem that agent mode already patched drops its manifest record but keeps the in-place patch, so afterrollback/remove/vendor --revertthe app still loads patched code and no command can undo it #2093, Agent re-scan that replaces a gem patch with a newer one leaves the old patch's other files patched:apply --checkandvexpass, and afterrollbackthe app still loads patched code that no command lists #2096, Nightly: gem_vendor_fresh_checkout_bundle_install_and_revert fails under bundler 1.17.3 (bundle cleanleaves rack-3.2.8) #2115, 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, Agent-modescanfrom a monorepo root skips the Bundler projects below it without a word: "No packages found. Run your package manager's install first." or a success summary for the root alone #2125, apply reports a passenger gem patch as applied while the downloaded agent it runs is unchanged; on main the rebuild runs 16 minutes and its product is discarded silently #2144, Cargo VEX: an ignored [patch] Socket pin still attests a transitive crate; bypass remedies name ambiguous or incomplete commands #2045, Decide "is this source crates.io" in one helper: CargoLock::entries uses a substring test that accepts index.crates.io.evil.example #1933 and Slow scan: cargo hosted 3.8x median ms/pkg (per-patch Cargo.lock re-parse) #761, and their PRs had already closed all of them. Pin a Maven patch only where Maven selects its base version #2100 only refs Tracking: scope project-mode cache crawls to what the project resolves #595 and Fix Bun defects: lock version 3, bundled trees, JSONC manifests, default-trusted packages #2120 only refs Decide: should hosted VEX require installed evidence by default instead of attesting from lockfile wiring? #1099 (agent:needs-human); Test Maven hosted, vendored and agent modes against production #2118, Take the root alone for a pnpm workspace pattern that names the root #2121, Vendor patched npm packages in Rush monorepos #2134 and Refer to Socket's backend generically across code, docs and CI notes #2141 cite no open issue.agent:claimedfrom 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, Vendored gem scan on a stock Rails Gemfile fails 5 of 7 gems (group blocks, native extensions) and none of the errors says what to do, thoughgroup:at the top level and--mode hostedboth work #2091, 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, Vendoring a gem that agent mode already patched drops its manifest record but keeps the in-place patch, so afterrollback/remove/vendor --revertthe app still loads patched code and no command can undo it #2093, Agent re-scan that replaces a gem patch with a newer one leaves the old patch's other files patched:apply --checkandvexpass, and afterrollbackthe app still loads patched code that no command lists #2096, Nightly: gem_vendor_fresh_checkout_bundle_install_and_revert fails under bundler 1.17.3 (bundle cleanleaves rack-3.2.8) #2115, 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 and Agent-modescanfrom a monorepo root skips the Bundler projects below it without a word: "No packages found. Run your package manager's install first." or a success summary for the root alone #2125: they are closed.rollback/removetell the user to re-run with--gem-refresh-shared, but the re-run exits 1 with "nothing to roll back" and the patched gem stays in the shared gem home #2086 (PR Fix gem unwind warning naming a re-run that cannot work (#2086) #2156), Hosted to vendored switch of a gem fails with "no installed package found on disk" once a newer patch is published, though the gem is installed and--dry-runpreviews success #2129 (Fix vendoring a hosted gem whose patch was superseded (#2129) #2143) and Hosted gem scan sends a gemspec-declared gem to--mode vendored, which refuses it for the same reason and names no remedy #2151 (Fix gem remedy naming a mode that refuses too (#2151) #2160) all have open PRs, so none is stale. No bughunt issue lacks apm:*label.4fe0859). It already records the fixes as fixed or not yet re-run, so there is no drift. socket-patch implementation review (living document, file by file): abstraction, duplication, mechanisms, tests #1830 changed but belongs to the paused architecture routine.commit.gpgsign, ssh), without-c user.*overrides. Commits made through the contents API are rejected too.Recent actions (rolling, newest first; no-op runs left out)
agent:claimedfrom closed 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, Vendored gem scan on a stock Rails Gemfile fails 5 of 7 gems (group blocks, native extensions) and none of the errors says what to do, thoughgroup:at the top level and--mode hostedboth work #2091, 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, Vendoring a gem that agent mode already patched drops its manifest record but keeps the in-place patch, so afterrollback/remove/vendor --revertthe app still loads patched code and no command can undo it #2093, Agent re-scan that replaces a gem patch with a newer one leaves the old patch's other files patched:apply --checkandvexpass, and afterrollbackthe app still loads patched code that no command lists #2096, Nightly: gem_vendor_fresh_checkout_bundle_install_and_revert fails under bundler 1.17.3 (bundle cleanleaves rack-3.2.8) #2115, 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, Agent-modescanfrom a monorepo root skips the Bundler projects below it without a word: "No packages found. Run your package manager's install first." or a success summary for the root alone #2125agent:claimedfrom closed Rolling back a gem that went from hosted to vendored leaves the patched installed copy invendor/bundle: the app keeps loading patched code andbundle pristine/--redownloadfail on the restored checksum #2083, Hosted gem scan over an agent-patched project leaves the upstream.gemin the gem home cache, sobundle pristine/--redownloadfail on the lock checksum and remove the gem (or silently reinstall it unpatched) #2089, 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 #2105applycan 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 (Fetch gems that get built over https only; refuse paths that cannot be probed #211289c482dchanged the docs only; argon2 rebuild still fails), left openagent:claimedfrom closedscan --mode agent --no-gem-native-rebuild(andSOCKET_NO_GEM_NATIVE_REBUILD=1) still rebuilds the gem native extension andvexattests it; onlyapplyandgethonor the opt-out #2069, A hosted-patched gem disappears frombundle outdatedandbundle updatecan no longer move it, and nothing in the scan output or docs says so #2070, 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, Windows: hot_patch test compares a Path::join result against a /-joined string (main CI red since #2068) #2080, 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 #20907c5240e)9bac91ffixed items 1–2 only), left openpm:pnpmto pnpm reader follow-ups from #2009's review #2044$expansion fixed by Fix the defects found by the 5.1.0 full review: do no harm, Bundler hosted mode, VEX, monorepo scans #2052354ab998; JSON escaping,parse_vendor_pathcheck and tests remain), left openagent:claimedfrom closed On a go 1.16 go.mod, Go apply, apply --check and vex trust a stale go.sum zip line left by a downgrade, so they report an inert replace as applied (or refuse a selected version) #1862overridesforces to a remote tarball URL;npm ciinstalls it unpatched, and vendoredvex/vendor --checkstill pass #1965) and Bug hunt ledger: Go modules #317 (On a go 1.16 go.mod, Go apply, apply --check and vex trust a stale go.sum zip line left by a downgrade, so they report an inert replace as applied (or refuse a selected version) #1862)26e39db5,refusals.rs:79)go mod tidydrop the replace but not restore the module's go.sum lines, so the next defaultgo buildfails with "missing go.sum entry" while rollback exits 0 with no warning #549 as completed (193efefvia Sweep core utils http, mod and hatch #1914)go mod tidydrop the replace but not restore the module's go.sum lines, so the next defaultgo buildfails with "missing go.sum entry" while rollback exits 0 with no warning #549)2e3e30ec)agent:claimedfrom 43 closed issuesDeferred / unsure
--mode vendored, which refuses it for the same reason and names no remedy #2151 (hosted scan sends a gemspec gem to vendored), Hosted gemrollback/removesays it rebuilds a native gem "from its patched gem" while it installs the upstream build #2152 (hosted gem unwind says it rebuilds "from its patched" tree) and Nightly: maven_reactor_implicit_subprojects fails on Maven 4.0.0-rc-7 ("the root is pinned") since 4fe0859 #2158 (Maven 4.0.0-rc-7 reactor nightly) are distinct defects. Hosted gemrollback/removesays it rebuilds a native gem "from its patched gem" while it installs the upstream build #2152 and Hosted gemrollback/removetell the user to re-run with--gem-refresh-shared, but the re-run exits 1 with "nothing to roll back" and the patched gem stays in the shared gem home #2086 both concern hosted unwind messages but name different wrong remedies, so they are not duplicates.applycan 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: Never take a native gem build that compiled nothing of a patch for a patched build; test agent mode on real nokogiri #2136 doesn't touch argon2'sMakefile.real(no match onorigin/mainoutside a test comment), and ledger Bug hunt ledger: Bundler (RubyGems) #316 re-confirmed it on4fe0859. Close when an argon2 rebuild fix lands.rollback/removetell the user to re-run with--gem-refresh-shared, but the re-run exits 1 with "nothing to roll back" and the patched gem stays in the shared gem home #2086 (shared gem home re-run) and Hosted to vendored switch of a gem fails with "no installed package found on disk" once a newer patch is published, though the gem is installed and--dry-runpreviews success #2129 (hosted→vendored after a superseded patch) are separate defects with open PRs.COMPOSER_VENDOR_DIREnvGuard race) and Flake: an_interrupt_pass_ends_the_whole_build_group fails on macOS ("the pass reaped the build's child itself") #2088 (macOSan_interrupt_pass_ends_the_whole_build_groupflake) are open ci-janitor flakes. They differ from closed Crawler unit tests read the developer's real HOME, YARN_NODE_LINKER and VIRTUAL_ENV: 12 pkg_managers/pnpm_layout tests fail under a user nodeLinker setting #1919.set_varguards and the production golang/deno legs remain. Close once those land.lock_cli.rs:18), Move npm-family workspace resolution out of hosted::governing_root: vendor, vex, scan and the crawler all import it #1974, Read gradle-wrapper.properties through one parser: the hosted and vendored Gradle 6.8 gates read the wrapper version two different ways and disagree ondistributionUrl = …#1963, Write every --json envelope through ui::print_json: 43 production sites print println!("{}", env.to_pretty_json()) instead #1925, Interpolate Maven ${...} properties through one formats::maven helper: five copies drifted on recursion, undefined names and the length cap #1887, Route the remaining hand-rolled child-process deadlines through utils::process #1067, Tracking: drive the seven npm-family vendor backends through one generic driver #920.send_json_requeststill loops besideretry_vendor; child 3 open. Consolidate remaining BOM stripping after the pnpm reader failures were fixed #905: inline'\u{feff}'stripping remains. Delete the vendored and hosted-vlt helpers left without a production caller by the v5 consolidation #782: the named helpers still exist.vendorandvendor --checkrun from an npm workspace member directory exit 0 without looking at the root lock: the eject says "nothing to vendor" and the check passes over a tampered artifact #1921/Vendored scan/get run from an npm workspace member directory refuses with "no package-lock.json … run your package manager's install first", a remedy that loops; hosted names the workspace root #1904, Bundler's path tier rule is written twice in ruby_crawler, and store discovery ignores BUNDLE_DISABLE_SHARED_GEMS: "false" #1931/Move Bundler settings parsing out of ruby_crawler into formats/gem/settings.rs, and read .bundle/config keys with one parser (two drifted today) #1935 and PnpmLock::parse inventories a pnpm 11 env-lockfile document's configDependencies as project packages, while InstalledPackages and the upstream evidence reader drop it #1949/"pnpm lock importers list a non-root member" is written twice and drifted on comment lines: keep formats::pnpm::workspace, delete the utils::pnpm_workspace copy #1952/Read pnpm YAML settings through one formats::pnpm scalar reader: three top-level-value readers drifted on quotes and empty values, and .modules.yaml is parsed twice #1918 are different code paths. None is a duplicate.# socket-patch vendor:comment:vendor --revertdrift-keeps forever (exit 0), and the suggested "re-vendor" remedy is a no-op or adds a duplicate line #977 (pip half fixed by Fix vendored requirements revert on edited comment (#977) #1825) and Perf regression: yarn-berry/rescan wall +21% (a43cf327..17793173; #1321/#1323/#1397 candidates) #1607 areagent:needs-human. Perf regression: yarn-berry/rescan wall +21% (a43cf327..17793173; #1321/#1323/#1397 candidates) #1607 is labeledpm:yarnwhile other berry issues usepm:yarn-berry.failmarks are stale.scan --check: a read-only exit status for "patches are available but not applied" #2051 (scan --check) areagent:needs-human. pnpm reader follow-ups from #2009's review #2044 has an open design question from 2026-11-09.de81f1b5. No exact duplicates found. Next Monday sweep: 2026-11-23.Needs a human
agent:needs-human.agent:needs-humandecisions: Maven CI matrix has no Windows leg, a stale 4.0 RC, and no legs at the resolver boundaries #267, Hosted Composer vex attests not_affected for an unpatched package when config.vendor-dir is absolute or uses ~/ or $HOME/ #686, Vendored yarn classic wiring breaks every install run from a workspace member directory: yarn resolves thefile:./.socket/vendor/…tarball against the member dir #691, Vendored requirements.txt can't be reverted once the user touches the# socket-patch vendor:comment:vendor --revertdrift-keeps forever (exit 0), and the suggested "re-vendor" remedy is a no-op or adds a duplicate line #977, Port the Composer transport-options drop to depscan's composer-lock-members.ts #1052, Self-update and install.sh trust an unsigned SHA256SUMS from the same release #1065, Decide: should hosted VEX require installed evidence by default instead of attesting from lockfile wiring? #1099, Hosted gem stale-install guard still flags an unused system gem-home copy under Bundlerdeploymentor the.bundledefault path, soscan --mode hosted --vexfails withno_applicable_patcheson fresh checkouts #1109, Decide: should hosted rollback keep an originals sidecar, or restore only formats whose original is a pure function of registry data? #1130, Decide: keep the in-memory hosted engine and napi addon as a supported product, or delete them #1200, CI perf (settings): merge queue builds 5 entries at a time — a 13-PR burst took 100 min to drain on a 21-min CI run (~21 min per PR past the 5th) #1248, Decide (post-v5): fix/undo/sync command model and exit 0 when there is nothing to undo #1352, CI perf (settings): slow Depot 16-core hosts since 10-13 15:20 add ~1.5–2.5 min to the merge-queue critical path (~0.6k Linux job-min/day) #1512, Perf regression: yarn-berry/rescan wall +21% (a43cf327..17793173; #1321/#1323/#1397 candidates) #1607, Depot runners lost mid-job evicted 2 merge-queue entries in 27h #1660, CI perf (settings): 90-min merge-queue check timeout let one stuck check freeze main for 77 min #1670, CI perf (coverage tradeoff): Windows-only path bugs reach main becausetest (windows-latest, *)runs only nightly #1720, Decide: one hostedredirectJSON block for disk scan and the in-memory engine (today {mode, rewrittenFiles} vs a seven-key block) #1881, Merge queue evicts entries on CLI_CONTRACT.md conflicts (3 PRs, 4 evictions on 11-07) #2036, Addscan --check: a read-only exit status for "patches are available but not applied" #2051.overridesforces to a remote tarball URL;npm ciinstalls it unpatched, and vendoredvex/vendor --checkstill pass #1965 is closed (fixed by Fix npm overrides to a URL treated as registry (#1965) #19661e63b45f) but still carriesagent:claimedandagent:needs-human. A human should drop both labels if the decision is settled.scan/applystill skip the package aspackage_not_installedwith exit 0 (composer_crawler.rs:821).Generated by Claude Code
All reactions