Bound every pipeline lock wait: fail the stage on timeout, name the holder, add a no-progress watchdog

Abierto
#2,205 0 comentarios 0 reacciones 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
35/100
Tipo de issue
Nueva funcionalidad
Claridad
Bastante claro
Estado de actividad
Activo
Stack tecnológico
shell
Área
ci-cd, tooling

Línea de trabajo

The issue covers every lock taken in a pipeline stage, so start by finding where stages acquire locks, such as flock calls in the stage scripts, and where the global run deadline is enforced. Add a bounded wait that fails the stage on timeout and prints the holder's PID, command and start time when readable. The done check is a test that holds the lock, runs the stage, and confirms it fails within the bound and names the holder. The no-progress watchdog is a separate part and should be scoped with maintainers first.

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

Descripción

Problem. An unbounded lock wait in a pipeline stage looks like a slow run. It burns the whole run budget, then fails with a generic deadline error that names no cause.

Failure scenario. A stage re-enters a script that takes the same non-reentrant lock its parent holds. It blocks at near-zero CPU with no output until the global deadline, and the log never says who held the lock.

Proposed practice.

  1. Every lock in a pipeline path uses a bounded wait (for example flock -w <seconds>).
  2. On timeout, fail that stage at once and print the holder (PID, command, start time) when readable.
  3. Add a no-progress watchdog: no output and near-zero CPU for a set window aborts the run with a stated reason.

Done when. A test that holds the lock and runs the stage fails within the bounded wait and names the holder. Related: #483, #2166.

Lenguaje dominante
Python
Estrellas
2
Forks
1
Merge medio
1 h 29 min
PR fusionados (30 d)
567

Preparar el entorno

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 protocol/perun

Todos los issues de protocol/perun

Issues similares

Más issues de Python

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.