[Bug]: Container hub repeatedly refuses startup with unverified proxy ownership

Abierto
  • ¿Algo no está bien?
#8,552 1 comentario 1 reacción 0 asignados Ver en GitHub

Los mantenedores suelen responder en 1 día

Nadie ha tomado este issue todavía.

Evaluación

Dificultad
4/5
Tiempo estimado
3-5 días
Aptitud para principiantes
42/100
Tipo de issue
Error
Claridad
Bastante claro
Estado de actividad
Activo
Stack tecnológico
bun, docker, kubernetes, typescript
Área
backend, devops

Línea de trabajo

Start with src/cli/start-owner-fence.ts and src/server/proxy-liveness.ts, then trace the startup recovery call in src/cli/index.ts. Reproduce or inspect the inconclusive liveness path using the reported PID and runtime-port state; the exact trigger is unknown. Done means startup recovers when the previous owner is proven absent while preserving ownership fencing.

Escrito por el modelo de indexación a partir del texto del issue.

Descripción

bug service
Client or integration

Other

Area

Service lifecycle

Summary

In a Kubernetes hub deployment using the community image nordz0r/opencodex:2.100.0, the process failed repeatedly at recoverStartJournalAfterOwnerRecheck (src/cli/start-owner-fence.ts:14) with the following error. Persistent state included ocx.pid containing 1 and runtime-port.json recording PID 1. The exact cause of the inconclusive liveness observation has not been established; stale records and PID namespace reuse are hypotheses, not a proven mechanism.

Proxy ownership is unverified; startup refused. PID record and recovery journal were preserved. Retry when the owner can be verified.

The pod entered CrashLoopBackOff. The trigger that produced the stale state was not established, nor was the liveness observation kind recorded. The expected behavior is safe recovery when absence of the previous owner can be established, without compromising ownership fencing.

Reproduction
  1. Observe a Kubernetes hub pod running the community image nordz0r/opencodex:2.100.0 with a persistent state volume. In the observed incident it repeatedly exited with the error below; the trigger is unknown.
  2. While the pod was stopped, its state directory contained ocx.pid with 1 and runtime-port.json with PID 1, port 10100 and hostname 0.0.0.0. Other ownership and journal files were present.
  3. The operator stopped the workload and quarantined ocx.pid, run/.c10100.lease, codex-home/opencodex-journal.json and .opencodex-owner.json; on restart the same error remained.
  4. After additionally moving runtime-port.json aside and restarting, the pod became ready and healthz returned 200. This recovery observation does not establish which file or liveness branch caused the failure. A minimal docker kill reproduction has not been verified.
Version

2.100.0 (crash observed); fence code identical in 2.100.2 (src/cli/start-owner-fence.ts, src/server/proxy-liveness.ts unchanged between the two tags).

Operating system

Debian (k3s container, linux/amd64); oven/bun image, proxy as container PID 1.

Provider and model

Not provider-specific.

Logs or error output
error: Proxy ownership is unverified; startup refused. PID record and recovery journal were preserved. Retry when the owner can be verified.
      at recoverStartJournalAfterOwnerRecheck (/home/bun/app/src/cli/start-owner-fence.ts:15:15)
      at recoverStartStateUnderOwnershipLease (/home/bun/app/src/cli/start-owner-fence.ts:45:23)
      at async handleStart (/home/bun/app/src/cli/index.ts:446:41)

kubelet then reports CrashLoopBackOff for the pod on every restart.

Screenshots and supporting files

Observed stale state and recovery sequence are described above. Do not remove ownership records unless the previous owner has been independently proven absent; the exact failure mechanism requires further investigation.

Redacted configuration
{"runtimeRole":"hub","hostname":"0.0.0.0","port":10100}
Checks
  • I searched existing issues and documentation.
  • I removed secrets, tokens, account details, request credentials, and personal data.
Lenguaje dominante
TypeScript
Estrellas
16.9k
Forks
1.3k
Merge medio
1 d 4 h
PR fusionados (30 d)
773

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

  1. Lee el issue completo y luego la guía de contribución del proyecto.
  2. Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
  3. Haz un fork del repositorio y trabaja en una rama.
  4. Abre un pull request que haga referencia al número del issue.

Más de lidge-jun/opencodex

Todos los issues de lidge-jun/opencodex

Issues similares

Más issues de TypeScript

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.