[bug] Windows: memory writes and statusline fail on a fresh install (POSIX-only path…
Los mantenedores suelen responder en 5 días
Nadie ha tomado este issue todavía.
- #2802 de @isihussain — cerrado sin fusionar
Evaluación
- Dificultad
- 4/5
- Tiempo estimado
- 3-5 días
- Aptitud para principiantes
- 48/100
- Tipo de issue
- Error
- Claridad
- Bien especificado
- Estado de actividad
- Activo
- Stack tecnológico
- bun, typescript
Línea de trabajo
Start with TOOLS/lib/Boundary.ts, TOOLS/MemoryWriter.ts, STATUSLINE/Admission.ts, and STATUSLINE/State.ts; trace the seven Windows path and permission failures described in the issue. Run the listed memory health check and pipe session JSON into LIFEOS_StatusLine.sh on Windows to reproduce them. Done means the health check has no critical findings, ledger verification succeeds, and the statusline renders every row.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
What happened
On Windows every memory write is refused and the statusline crashes on every render, because several tools assume POSIX paths and permission bits
Summary
Environment
- LifeOS 8.761.0, Windows 11, Bun 1.4.4 (Windows x64)
- Claude Code 2.1.303, hooks run through Git Bash
- The user-data folder is a directory junction from the install tree into the config tree
Summary
On Windows, every memory write is refused and the statusline crashes on every render. Both come from code that assumes / separators, a / filesystem root, or POSIX permission bits. Seven defects, each hidden behind the previous one.
Memory (health check reports "Latest reviewer run is failed")
-
TOOLS/lib/Boundary.ts,confinedPath: an absolute walk starts atsep, so the first component it stats is a backslash followed by the drive letter, which never exists (ENOENT). Every absolute path is refused, including the real file inside the user-data repo. Relative paths pass. Error:EWRITE_FAILED: memory write refused at the system/user boundary ... does not resolve inside the USER_DATA repo. Fix: start fromparse(p).rootand join withcur.endsWith(sep). -
TOOLS/lib/Boundary.ts,atomicWrite:invariant((fstatSync(fd).mode & 0o777) === mode)always fails, because Windows reports 0o666 whatever mode was requested. -
TOOLS/MemoryWriter.ts,snapshotBeforeWrite:absPath.split("/").pop()returns the whole Windows path, which trips the invariantsnapshot target ... must sit directly in ... memory-snapshots. -
TOOLS/MemoryWriter.ts: the invariantsnapshot dir ... is wider than 0700fails for the same reason as (2). -
TOOLS/MemoryWriter.ts(4 sites):abs.replace(CLAUDE_ROOT + "/", "")leaves the full drive-letter path as the ledger row'sfilekey.CortexLedger.ts verifythen reportsentries-drifton files the writer itself just wrote.
Statusline (prints nothing)
-
STATUSLINE/Admission.ts(3 sites):startsWith(this.directory + "/render-")throwsINVARIANT VIOLATION: render lock must be scoped to inflight. Fix: usesep. -
STATUSLINE/State.ts,expandRoot: converts a drive-letter install root to the MSYS form (slash, drive letter, slash). Bun on Windows cannot open that form, so every read understate.rootreturns null. Visible effect: blank STATE row and no doctor line. Fix: keep the native path whenprocess.platform === "win32".
Smaller items found on the way
TOOLS/MergeSettings.tswith a relative--output settings.jsonfails withEEXIST: file already exists, mkdir '.'. An absolute output path works.TOOLS/Inference.tsline 483:spawnSync("id", ["-un"]).stdout.trim()throws when the USER variable is unset andidis not on PATH (PowerShell).TOOLS/ParkedReplay.tsreplays onlyEDRIFTitems, so writes parked withEWRITE_FAILEDare never replayed after the cause is fixed.- The header's flag emoji draws as two letters, because Windows fonts have no flag glyphs.
Repro
- Fresh install on Windows 11; let one session end so the memory reviewer fires.
- Run
bun LIFEOS/TOOLS/MemoryHealthCheck.ts: overall is critical. - Pipe any session JSON into
LIFEOS/LIFEOS_StatusLine.sh: invariant violation.
Verification of the fixes
confinedPath: 12 accept and refuse cases pass, including..escapes through the junction and targets in the system tree.- After all seven: the health check has 0 critical findings,
CortexLedger.ts verifyis OK, a reviewer run wrote entries unaided, and the statusline renders every row.
The patches are small, about 25 changed lines in four files. Happy to open a PR.
Details
- Kind: bug
- LifeOS version: 8.761.0
Filed with the LifeOS /feedback command. The reporter approved this exact text before it was sent; paths, secrets and personal details were stripped locally.
- Lenguaje dominante
- TypeScript
- Estrellas
- 19.9k
- Forks
- 2.5k
- Métricas de merge de PR
- Sin PR fusionados en 30 d
Preparar el entorno
Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de danielmiessler/LifeOS
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 67/100
danielmiessler/LifeOS#2894 ·
Los mantenedores suelen responder en 5 días
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 82/100
danielmiessler/LifeOS#2891 ·
Los mantenedores suelen responder en 5 días
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 85/100
danielmiessler/LifeOS#2888 ·
Los mantenedores suelen responder en 5 días
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 78/100
danielmiessler/LifeOS#2886 ·
Los mantenedores suelen responder en 5 días
-
Dificultad 1/5 Menos de una hora Aptitud para principiantes 68/100
danielmiessler/LifeOS#2848 ·
Los mantenedores suelen responder en 5 días
Todos los issues de danielmiessler/LifeOS
Issues similares
-
ready-for-agent
Dificultad 1/5 Menos de una hora Aptitud para principiantes 86/100
LucasSantana-Dev/Lucky#3257 ·
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 75/100
XiaomiMiMo/MiMo-Code#2849 · 1 comentario ·
Los mantenedores suelen responder en 1 día
-
enhancement frontend good first issue help wanted testing-infrastructure
Dificultad 2/5 1-3 horas Aptitud para principiantes 76/100
Los mantenedores suelen responder en 2 días
-
enhancement follow-up
Dificultad 2/5 1-3 horas Aptitud para principiantes 76/100
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 82/100
obsidian-tasks-group/obsidian-tasks#4155 ·
Los mantenedores suelen responder en 1 día