Exit code 1

[najamelan]

Just gave it a try. At first it worked. I just wanted to run it again after I changed something in the code and ever since I get "Debugger exited without connecting (exit code 1)". I tried restarting vscode, but no avail.

Any idea what error code 1 is?

There is no output in any of the terminals. I don't use bundler.

[castwide]

The most common cause of exit code 1 is that the program couldn't be found. If you're using something like rvm or rbenv, is it possible you switched to a version of Ruby that doesn't have the gem installed?

[najamelan]

I'll try to debug it a bit more, but I don't use rvm or rbenv. I just have stock ruby installed with pacman on arch, then all needed gems installed globally with gem install as root. I'm trying to debug a ruby script that is in a gem like directory, with a bin script that loads the rest of the app with require_relative. I tried putting the scripts path as an absolute path rather than a ${workspaceRoot} relative, but that didn't fix it.

Would it be the debugger program that couldn't be found? I will do a reboot, since I did an system upgrade since, but yesterday i literally just changed a debug statement in the code and since it wouldn't work anymore, so I'm a bit puzzled.

[najamelan]

Ok, I figured a workaround. There is a lingering readapt process. When I kill that with the system monitor, I can launch debug again.

This is probably unrelated, but I do get an error in the debug console:\

#<Thread:0x0000557647b1f478 /usr/lib/ruby/gems/2.7.0/gems/backport-1.1.2/lib/backport/client.rb:101 run> terminated with exception (report_on_exception is true):
/usr/lib/ruby/gems/2.7.0/gems/readapt-1.0.0/lib/readapt/output.rb:20:in `to_json': "\\xC3" from ASCII-8BIT to UTF-8 (Encoding::UndefinedConversionError)
	from /usr/lib/ruby/gems/2.7.0/gems/readapt-1.0.0/lib/readapt/output.rb:20:in `send_event'
	from /usr/lib/ruby/gems/2.7.0/gems/readapt-1.0.0/lib/readapt/output.rb:8:in `receiving'
	from /usr/lib/ruby/gems/2.7.0/gems/backport-1.1.2/lib/backport/client.rb:63:in `tick'
	from /usr/lib/ruby/gems/2.7.0/gems/backport-1.1.2/lib/backport/server/stdio.rb:18:in `update'
	from /usr/lib/ruby/2.7.0/observer.rb:197:in `block in notify_observers'
	from /usr/lib/ruby/2.7.0/observer.rb:196:in `each'
	from /usr/lib/ruby/2.7.0/observer.rb:196:in `notify_observers'
	from /usr/lib/ruby/gems/2.7.0/gems/backport-1.1.2/lib/backport/client.rb:121:in `read_input'
	from /usr/lib/ruby/gems/2.7.0/gems/backport-1.1.2/lib/backport/client.rb:102:in `block in run_input_thread'

Don't really know where it's coming from, as all of my source files are valid UTF-8. My default locale is en_US.UTF-8. It doesn't keep me from debugging, but it might cause the process to linger?

[castwide]

Yeah, that could be the problem. If it's making the server adapter hang, it might stop the process from exiting properly. Thanks for tracking it down. I'll look into it.

[qortex]

Ran into the same situation. When something fails and readapt hangs, then it does exit 1 and should be killed through the terminal to work again.

[rodcul]

Happened to me too, lost around an hour till I found this thread - thanks guys for the solution!

[cubytz]

@najamelan Thanks a lot!! Worked for me!

[jarrodldavis]

I'm having the same issue, but only on GitHub Codespaces. Everything works fine at first, but restarting or stopping then starting the debug session results in either an exit code 1 error or the debug session briefly starting but Rails/Puma complaining about an existing server.pid file.

Locally, using the Remote - Containers extension, restarting debug sessions works fine.

[jarrodldavis]

Okay, I figured out the problem with Codespaces. It looks like something with how auto-forwarded ports does not play well with Readapt's TCP server. Turning off automatic port forwarding fixes the issue:

// settings.json
{
  "remote.autoForwardPorts": false
}

[DaVicki]

@najamelan ran into the same problem. Thanks for the solution.