ruby-on-rails - 彪马“提前解雇工人”调查困难重重
问题描述
我只更新了我的应用程序的 gem 并移至 Rails 6.1.0.rc1，现在无法运行puma. 我看到许多消息说[7XXXX] Early termination of worker。
我可以通过运行在本地复制它，但在其他环境的日志中或任何其他环境的日志bundle exec puma -p 3000 -e production中都看不到任何其他输出。log/production.log
此时，除了等待新的 Rails rc 之外，我不确定如何找到问题的根源。bundle exec puma -C config/puma.rb -p 3000如果我运行或也没有问题bundle exec rails s。
额外细节
在Gemfile
gem 'puma', '~> 4.1'
并且在Gemfile.lock
puma (4.3.6)
我更新到 Puma 5.0.4，所以我可以使用Puma 可以运行的控制/状态服务器thread-backtraces上的方法。它能够给我以下输出，稍后我将深入研究，但它似乎并没有提供确凿的证据。
[
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/cluster.rb:412:in `select'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/cluster.rb:412:in `run'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/launcher.rb:171:in `run'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/cli.rb:80:in `run'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/bin/puma:10:in `<top (required)>'",
"~/.rvm/gems/ruby-2.7.2/bin/puma:23:in `load'",
"~/.rvm/gems/ruby-2.7.2/bin/puma:23:in `<main>'",
"~/.rvm/gems/ruby-2.7.2/bin/ruby_executable_hooks:24:in `eval'",
"~/.rvm/gems/ruby-2.7.2/bin/ruby_executable_hooks:24:in `<main>'"
],
"name": "Thread: TID-5ic"
},
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/plugin/tmp_restart.rb:21:in `sleep'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/plugin/tmp_restart.rb:21:in `block in start'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/plugin.rb:68:in `block (2 levels) in fire_background'"
],
"name": "Thread: TID-5iw puma plugin background 0"
},
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/reactor.rb:70:in `select'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/reactor.rb:70:in `select_loop'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/reactor.rb:34:in `block in run'"
],
"name": "Thread: TID-5jg puma reactor"
},
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/thread_pool.rb:308:in `sleep'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/thread_pool.rb:308:in `block in start!'"
],
"name": "Thread: TID-5k0 puma threadpool reaper"
},
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/thread_pool.rb:308:in `sleep'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/thread_pool.rb:308:in `block in start!'"
],
"name": "Thread: TID-5kk puma threadpool trimmer"
},
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/server.rb:323:in `select'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/server.rb:323:in `handle_servers'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/server.rb:258:in `block in run'"
],
"name": "Thread: TID-5l4 puma server"
},
{
"backtrace": [
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/launcher.rb:219:in `backtrace'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/launcher.rb:219:in `block in thread_status'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/launcher.rb:215:in `each'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/launcher.rb:215:in `thread_status'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/app/status.rb:59:in `call'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/request.rb:76:in `block in handle_request'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/thread_pool.rb:337:in `with_force_shutdown'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/request.rb:75:in `handle_request'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/server.rb:430:in `process_client'",
"~/.rvm/gems/ruby-2.7.2/gems/puma-5.0.4/lib/puma/thread_pool.rb:145:in `block in spawn_thread'"
],
"name": "Thread: TID-5og puma threadpool 001"
}
]
解决方案
意外！
pumactl并且有一个control-url帮助但我的朋友提出了我只希望更明显的最佳想法，
你是在不同的服务器上抛出错误吗？
我跑了gem install thin，RAILS_ENV=production thin start终于向我展示了我正在寻找的错误！
事实证明，我不应该使用非公共方法，add_template_helper因为ActionMailer::Base可能并不总是获得ActionController::Base. 我在开发中没有看到这个错误，因为 Rails 不会急切地加载你的所有类。