Ctrl-C with WEBrick crashes with SystemExit

[denisdefreyne]

Steps to reproduce

1. Create a Nanoc site with a Gemfile not containing puma
2. Run bundle exec nanoc view
3. Type ctrl+C

Expected behavior

Nanoc shuts down without error message.

Actual behavior

[2024-04-12 18:56:54] FATAL SystemExit: exit followed by a stack trace.

Details

See #1695 (reply in thread).

Crash log

n/a

[stefanos82]

I have the impression this behavior is related to ruby/webrick#102, because based on my limited investigation deciphering this thrown error message output

stefanos@debian:tutorial $ bin/nanoc view
View the site at http://127.0.0.1:3000/
[2024-04-12 22:07:51] INFO  WEBrick 1.8.1
[2024-04-12 22:07:51] INFO  ruby 3.1.2 (2022-04-12) [x86_64-linux-gnu]
[2024-04-12 22:07:51] INFO  WEBrick::HTTPServer#start: pid=13337 port=3000
^C
[2024-04-12 22:07:53] FATAL SystemExit: exit
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/error_handler.rb:41:in `exit'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/error_handler.rb:41:in `block (2 levels) in handle_while'
        /usr/share/rubygems-integration/all/gems/webrick-1.8.1/lib/webrick/server.rb:173:in `select'
        /usr/share/rubygems-integration/all/gems/webrick-1.8.1/lib/webrick/server.rb:173:in `block in start'
        /usr/share/rubygems-integration/all/gems/webrick-1.8.1/lib/webrick/server.rb:32:in `start'
        /usr/share/rubygems-integration/all/gems/webrick-1.8.1/lib/webrick/server.rb:160:in `start'
        /home/stefanos/gems/gems/rackup-2.1.0/lib/rackup/handler/webrick.rb:34:in `run'
        /home/stefanos/gems/gems/adsf-1.5.0/lib/adsf/server.rb:37:in `run'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/commands/view.rb:37:in `run'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/command_runner.rb:15:in `block in call'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/error_handler.rb:57:in `handle_while'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/error_handler.rb:15:in `handle_while'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/command_runner.rb:14:in `call'
        /home/stefanos/gems/gems/cri-2.15.11/lib/cri/command_dsl.rb:294:in `block in runner'
        /home/stefanos/gems/gems/cri-2.15.11/lib/cri/command.rb:362:in `run_this'
        /home/stefanos/gems/gems/cri-2.15.11/lib/cri/command.rb:298:in `run'
        /home/stefanos/gems/gems/cri-2.15.11/lib/cri/command.rb:316:in `run'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli.rb:88:in `block in run'
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/error_handler.rb:57:in `handle_while'                                                                                         
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli/error_handler.rb:15:in `handle_while'                                                                                         
        /home/stefanos/gems/gems/nanoc-cli-4.12.20/lib/nanoc/cli.rb:86:in `run'                                                                                                                
        /home/stefanos/gems/gems/nanoc-4.12.20/bin/nanoc:18:in `<top (required)>'                                                                                                              
        bin/nanoc:27:in `load'                                                                                                                                                                 
        bin/nanoc:27:in `<main>'                                                                                                                                                               
[2024-04-12 22:07:53] INFO  going to shutdown ...                                                                                                                                              
[2024-04-12 22:07:53] INFO  WEBrick::HTTPServer#start done.

seems like the server does not have the time to stop / kill all running spawned processes, which throws this message.

Most probably though I've got things wrong, but it's a plausible hypothesis.