GitHub Documentation

Redis Presence: process id still remains after shutting down instances

  • Hi guys,
    I'm using redis presence for my colyseus instances. The problem is when one instance is shutdown, its process id still remain in redis, so the presence db will get bigger over time. I'm afraid this will increase query time. Is there any solution to clear unused process ids in redis presence ?
    Many thanks

    alt text

  • administrator

    This is a good point, I'm also facing a similar issue on production, but it mostly happens when the node process is forcibly shut down due to some exception.

    It would be nice if we can auto-heal colyseus:nodes automatically. The proxy won't be able to communicate with the dead nodes listed on Redis.


  • administrator

    As a workaround - you can kill all your nodes and clear the entire Redis database (flushdb) - and then spawn all nodes again

  • Thanks for workaround. I think that the only way for now


© 2020 Endel Dreyer