Errno::ECHILD: No child processes error when starting SolidQueue with rake solid_queue:start

[BrijeshSajeev]

When attempting to start SolidQueue using bundle exec rake solid_queue:start, an Errno::ECHILD: No child processes error is raised. The error occurs during the waitpid2 system call within the reap_and_replace_terminated_forks method in the fork_supervisor.rb file. This seems to indicate an issue with handling child processes in the forking supervisor.

Environment

• Ruby Version: 3.1.6
  

• Rails Version: 7.1.0
  

• SolidQueue Version: 0.5.0
  

• RVM/Gemset: ruby-3.1.6@rg_rails_7-rb-3.1.6
  

[BrijeshSajeev]

[rosa]

Huh, this is very strange. Could you share a bit more about your setup? Or the full log since it starts until it triggers that error? At that point it should have forked some process so there should be at least a child process... I wonder if there's no configuration at all and no forks are being created... I might not be handling this case well 🤔

[BrijeshSajeev]

Set Up :

• OS: Ubuntu 22.04.4 LTS x86_64
• Ruby Version: 3.1.6
• Rails Version: 7.1.0
• SolidQueue Version: 0.5.0

[rosa]

And could you share your solid queue configuration as well?

[BrijeshSajeev]

sure thing

[rosa]

Thanks! And are you testing this in development environment? Or production?

[BrijeshSajeev]

i'm testing this in dev, but somehow it takes the config for production

[BrijeshSajeev]

I noticed that when using the following configuration in the development environment:

When I run bundle exec rake solid_queue:start, it appears to be doing nothing—no logs, no errors, no noticeable activity. The process seems to hang without any feedback. This might indicate that no child processes are being forked or that the configuration isn’t being handled as expected.

Let me know if you need any additional details or if there's anything specific I should check!

[rosa]

Ok, so when you added the configuration in development above, you stopped getting the Errno::ECHILD error, right? This is something I need to handle better, when you don't have any workers or dispatchers configured at all 🤔 I'll fix that to finish gracefully, and to use the default configuration if the file is present but doesn't have any configuration for the given environment.

Now, to the second issue:

When I run bundle exec rake solid_queue:start, it appears to be doing nothing—no logs, no errors, no noticeable activity. The process seems to hang without any feedback.

By default it'll use Rails.logger to log, so you might have to check log/development.log to see if it's logging anything. Otherwise, you could add this to your development.rb environment:

config.solid_queue.logger = ActiveSupport::Logger.new(STDOUT)

And see if that shows the workers and dispatcher starting and running.

[BrijeshSajeev]

yeah, it's working now

but something is wrong with my server

[rosa]

Hmm yes... that one is new to me... Maybe this helps?

[BrijeshSajeev]

It's working fine.
Thank you so much for your quick response and for helping me out with this issue. I really appreciate your time and great talk.

[rosa]

Nice! 👏 So glad to hear it's working! I'll ship an improvement for the configuration problem and the Errno::ECHILD error when the configuration is empty.