Neighborhoods Kōjō is a distributed task manager.

5.5.0 2021-11-09 18:46 UTC

This package is auto-updated.

Last update: 2024-12-10 02:00:21 UTC


README

工場

A distributed task manager.

Kōjō is a collection of the following components:

  • Distributed task management.
  • Distributed, cooperative, process-aware semaphores and mutexes.
  • Static cron scheduling.
  • Multi-process model.
  • Distributed IPC.

Trying out Kōjō

Real-world use cases for Kōjō and instructions for getting Kōjō up and running in your project can be found at KojoFitness.

Debugging Kōjō

XDebug version greater than xdebug-2.7.0alpha1 is required when trying to debug Kōjō. This version of XDebug resolves issues (https://bugs.xdebug.org/938) caused by the way the Kōjō forks using pcntl.

If you are using PhpStorm and you have more concurrent Kōjō jobs running than the Max. simultaneous connections defined for your XDebug listener, Kōjō will appear to hang. To avoid this, increase your Max. simultaneous connections to the max value of 20 in PhpStorm's preferences under Languages & Frameworks > PHP > Debug, External connections section.

Kōjō appears to be hanging during local development?

  • Stop all local containers and rebuild them
  • Run ps -auxf to see if any processes are still running Kōjō with a title like neighborhoods-kojo: /server[8]/root[22]/job[30]
    • If yes run pkill -9 -f kojo to SIG_KILL all the processes containing the word kojo
  • Run docker-compose exec redis redis-cli monitor to see if Kōjō is creating any activity in redis.
    • If nothing is streaming by like "GET" "/neighborhoods/kojo/area_manager/job_state_changelog_processor.lock" messages then something else is blocking the execution of Kōjō internals. This is typically due to a debugger.
  • Turn off the debugger. Sometimes this will mess with Kojo
  • Check your version of XDebug (version before 2.7.0 won't work and 2.8.2 has seen some odd behavior. Recommend 2.9.2)
  • Checkout the debugging logs Troubleshooting common PHP debugging issues - Help | PhpStorm