procsd 0.5.4 → 0.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '09e6c1faceca84dc0337311001b11397aa338b25bc285960175e4537d3a38f71'
-  data.tar.gz: e1fc39db9a119b15d2042558b093ec222aed03baa67d75f824c3da8b99464e72
+  metadata.gz: ecf9ac7491a94c05d51ac93c055b54da7bd03d2c049179eb467c0af1762162d6
+  data.tar.gz: 6eae45fdc21ee57016a19c830933664ae142e3f4a2d77537faff22e9cd888dae
 SHA512:
-  metadata.gz: d321a3cb39f9a5c442a6c088682b020fafa6fc59e6f03b892bd438df4a10b0fcc3490e5651e20eb060e920de22b7364425da07368f73305b089d63a791f4246c
-  data.tar.gz: ce218eae99a2bd36cbc771e3ba199e0ac9b68cb8736202f0ec171c74f36cef22bca84d89f7704d356c2ee29fb9cb2d7ac935e98cf76079500bd8ed6799067033
+  metadata.gz: 23f82ef2670831ae7d452e5f14a820225b02f964ec7009f8f067f3ae4c938c15d89491d53f7af10205aba479840c2e9e2dd7260998ac33ddccedfdbf52f315db
+  data.tar.gz: 9afd0665731d122bb1406a1bd4f1fcbeb16d092cf584ca66c1365931db75ea17482893774dd3829cdcce8745f885b408e7f9ff6bab3542a48c0a74ba42078555

data/CHANGELOG.md

@@ -1,4 +1,17 @@
 # CHANGELOG
+## 0.6.0
+* **Breaking change:** Require Ruby >= 3.2.0 (dropping support for EOL versions)
+* Fix: Use YAML.safe_load instead of deprecated YAML.load
+* Fix: Use new `trim_mode` keyword argument for ERB.new
+* Change: Relax bundler/rake version constraints to allow modern versions
+* Add: Test coverage for generator output
+
+Thanks to [@botandrose](https://github.com/botandrose) (Micah Geisel) for contributions!
+
+## 0.5.5
+* Add: Allow to start/stop/restart particular service in the app target (example: `$ procsd start web`)
+* Add: RuntimeMaxSec option for process
+
 ## 0.5.4
 * Add: information how to use SSL integration with Cloudflare CDN enabled
 * Add: procsd config certbot_command command

data/README.md

@@ -1,18 +1,14 @@
 # Procsd
 
-I do like the way how simple is managing of application processes in production on Heroku with [Procfile](https://devcenter.heroku.com/articles/procfile). How easily can be accessed application logs with [heroku logs](https://devcenter.heroku.com/articles/logging) command. Just type `heroku create` and you're good to go.
+I really like how simple it is to manage application processes in production on Heroku with [Procfile](https://devcenter.heroku.com/articles/procfile). How easily you can access application logs with the [heroku logs](https://devcenter.heroku.com/articles/logging) command. Just type `heroku create` and you're good to go.
 
-Can we have something similar on the cheap Ubuntu VPS from DigitalOcean? Yes we can, all we need is a **systemd wrapper** which allows to export application processes from Procfile to system services, and control them/check status/access logs using simple commands.
+Can we have something similar on a cheap Ubuntu VPS from DigitalOcean? Yes we can, all we need is a **systemd wrapper** that exports application processes from Procfile to system services, and controls them/checks status/accesses logs using simple commands.
 
-> These days most of Linux distributions (including Ubuntu) has systemd as a default system processes manager. That's why it is a good idea to use systemd for managing application processes in production (for simple cases).
+> These days most Linux distributions (including Ubuntu) have systemd as their default system processes manager. That's why it is a good idea to use systemd for managing application processes in production (for simple cases).
 
 ## Getting started
 
-> **Note:** latest version of Procsd is `0.5.3`. Since version `0.4.0` there are some breaking changes. Check the [CHANGELOG.md](CHANGELOG.md). To update to the latest version, run `$ gem update procsd` or `$ bundle update procsd` (if you have already installed procsd).
-
-> **Note:** Procsd works best with Capistrano integration: [vifreefly/capistrano-procsd](https://github.com/vifreefly/capistrano-procsd)
-
-Install `procsd` first: `$ gem install procsd`. Required Ruby version is `>= 2.3.0`.
+Install `procsd` first: `$ gem install procsd`. Required Ruby version is `>= 3.2.0`.
 
 Let's say you have following application's Procfile:
 
@@ -20,7 +16,7 @@ Let's say you have following application's Procfile:
 web: bundle exec rails server -p $PORT
 worker: bundle exec sidekiq -e $RAILS_ENV
 ```
-and you want to have one instance of web process && two instances of worker process. Create inside application directory `procsd.yml` config file:
+and you want to have one instance of the web process and two instances of the worker process. Create a `procsd.yml` config file inside the application directory:
 
 ```yaml
 app: sample_app
@@ -50,7 +46,7 @@ Create: /etc/systemd/system/sample_app-web.1.service
 Create: /etc/systemd/system/sample_app-worker.1.service
 Create: /etc/systemd/system/sample_app-worker.2.service
 Create: /etc/systemd/system/sample_app.target
-Reloaded configuraion (daemon-reload)
+Reloaded configuration (daemon-reload)
 Created symlink /etc/systemd/system/multi-user.target.wants/sample_app.target → /etc/systemd/system/sample_app.target.
 Enabled app target sample_app.target
 App services were created and enabled. Run `start` to start them
@@ -62,21 +58,22 @@ deploy ALL=NOPASSWD: /bin/systemctl start sample_app.target, /bin/systemctl stop
 You can provide additional options for `create` command:
 * `--user` - name of the user, default is current _$USER_ env variable
 * `--dir` - application's working directory, default is current _$PWD_ env variable
-* `--path` - $PATH to include to the each service. Default is current _$PATH_ env variable
-* `--add-to-sudoers` - if option present, procsd will create sudoers rule file `/etc/sudoers.d/app_name` which allow to start/stop/restart app services without a password prompt (passwordless sudo).
-* `--or-restart` - if option provided and services already created, procsd will skip creation and call instead `restart` command. Otherwise (if services are not present), they will be created and (in additional) started. It's useful option for deployment tools like Capistrano, Mina, etc.
+* `--path` - $PATH to include in each service. Default is current _$PATH_ env variable
+* `--add-to-sudoers` - if this option is present, procsd will create a sudoers rule file `/etc/sudoers.d/app_name` which allows starting/stopping/restarting app services without a password prompt (passwordless sudo).
+* `--or-restart` - if this option is provided and services are already created, procsd will skip creation and call the `restart` command instead. Otherwise (if services are not present), they will be created and additionally started. This is a useful option for deployment tools like Capistrano, Mina, etc.
 
 
 ### Start application
 > Other control commands: `stop` and `restart`
 
+> You can start/stop/restart a particular process by providing its name, e.g.: `$ procsd restart worker`
+
 ```
 deploy@server:~/sample_app$ procsd start
 
 Started app services (sample_app.target)
 ```
 
-
 ### Check the status
 > You can filter processes, like `$ procsd status worker` (show status only for worker processes) or `$ procsd status worker.2` (show status only for worker.2 process)
 
@@ -205,7 +202,7 @@ Puma starting in single mode...
 Use Ctrl-C to stop
 ```
 
-`procsd exec` requres all the environment variables defined in `environment` section of `procsd.yml` config file.
+`procsd exec` requires all the environment variables defined in `environment` section of `procsd.yml` config file.
 
 Sometimes in development mode you need different environment configuration. For that you can add additional environment section `dev_environment` and require it as well using `--dev` flag, example:
 
@@ -226,13 +223,13 @@ dev_environment:
 deploy@server:~/sample_app$ procsd exec web --dev
 ```
 
-> In case if `dev_environment` has env variable with the same name like in `environment`, this variable will be rewritten with value from `dev_environment`.
+> If `dev_environment` has an env variable with the same name as one in `environment`, it will be overwritten with the value from `dev_environment`.
 
 
 ### Nginx integration (with automatic HTTPS)
-> Before make sure that you have Nginx installed `sudo apt install nginx` and running `sudo systemctl status nginx`.
+> First, make sure that you have Nginx installed (`sudo apt install nginx`) and running (`sudo systemctl status nginx`).
 
-If one of your application processes is a web process, you can automatically setup Nginx (reverse proxy) config for it. Why? For example to serve static files (assets, images, and all other files located in `public` folder or another customly defined folder) directly using fast Nginx, rather than application server. Or to enable SSL support (see below).
+If one of your application processes is a web process, you can automatically set up an Nginx (reverse proxy) config for it. Why? For example, to serve static files (assets, images, and all other files located in `public` folder or another custom folder) directly using Nginx, rather than the application server. Or to enable SSL support (see below).
 
 Add to your procsd.yml `nginx` section with `server_name` option defined:
 
@@ -266,7 +263,7 @@ Create: /etc/systemd/system/sample_app-web.1.service
 Create: /etc/systemd/system/sample_app-worker.1.service
 Create: /etc/systemd/system/sample_app-worker.2.service
 Create: /etc/systemd/system/sample_app.target
-Reloaded configuraion (daemon-reload)
+Reloaded configuration (daemon-reload)
 Created symlink /etc/systemd/system/multi-user.target.wants/sample_app.target → /etc/systemd/system/sample_app.target.
 Enabled app target sample_app.target
 App services were created and enabled. Run `start` to start them
@@ -320,7 +317,7 @@ Everything is done. Start app services (`procsd start`) and go to `http://my-dom
 
 #### Auto SSL using Certbot
 
-To generate Nginx config with free SSL certificate (from [Let’s Encrypt](https://letsencrypt.org/)) included, you need to install [Certbot](https://certbot.eff.org/) on the remote server first:
+To generate Nginx config with free SSL certificate (from [Let’s Encrypt](https://letsencrypt.org/)) included, you need to install [Certbot](https://certbot.eff.org/) on the remote server first. For Ubuntu 18.04 (check here instructions for other versions https://certbot.eff.org/lets-encrypt/ubuntufocal-nginx):
 
 ```bash
 $ sudo apt install software-properties-common
@@ -344,7 +341,7 @@ Configuration is done. **Make sure that all domains defined in procsd (nginx.ser
 
 > By default Certbot obtaining certificate from _Let's Encrypt_ without a contact email. If you want to provide contact email, define env variable `CERTBOT_EMAIL` with your email in the `.env` file.
 
-<details/>
+<details>
   <summary>Output</summary>
 
 ```
@@ -355,7 +352,7 @@ Create: /etc/systemd/system/sample_app-web.1.service
 Create: /etc/systemd/system/sample_app-worker.1.service
 Create: /etc/systemd/system/sample_app-worker.2.service
 Create: /etc/systemd/system/sample_app.target
-Reloaded configuraion (daemon-reload)
+Reloaded configuration (daemon-reload)
 Created symlink /etc/systemd/system/multi-user.target.wants/sample_app.target → /etc/systemd/system/sample_app.target.
 Enabled app target sample_app.target
 App services were created and enabled. Run `start` to start them
@@ -408,7 +405,7 @@ Successfully installed SSL cert using certbot
 That's it. Start app services (`procsd start`) and go to `https://my-domain.com` where you'll see your application proxying with Nginx and SSL enabled.
 
 
-<details/>
+<details>
   <summary>Note about using Cloudflare CDN</summary><br>
 
 If you use Cloudflare CDN, that means the process of obtaining Let's Encrypt SSL Certificate will fail. To fix it, install `python-certbot-dns-cloudflare` package:
@@ -489,15 +486,15 @@ Commands:
 
 ## Difference with Foreman
 
-[Foreman](http://ddollar.github.io/foreman/) itself designed for _development_ (not production) usage only and doing it great. Yes, Foreman allows to [export](http://ddollar.github.io/foreman/#EXPORTING) Procfile to the Systemd, but that's all. After export you have to manually use `systemctl` and `journalctl` to manage/check exported services. Procsd not only exports application, but provides [simple commands](#all-available-commands) to manage exported target.
+[Foreman](http://ddollar.github.io/foreman/) itself is designed for _development_ (not production) usage only and does it great. Yes, Foreman allows you to [export](http://ddollar.github.io/foreman/#EXPORTING) Procfile to Systemd, but that's all. After export you have to manually use `systemctl` and `journalctl` to manage/check exported services. Procsd not only exports the application, but provides [simple commands](#all-available-commands) to manage the exported target.
 
-* Foreman systemd export uses [dymamic](https://fedoramagazine.org/systemd-template-unit-files/) services templates and as a result generates quite a lot of files/folders in the systemd directory even for a simple application.
+* Foreman systemd export uses [dynamic](https://fedoramagazine.org/systemd-template-unit-files/) services templates and as a result generates quite a lot of files/folders in the systemd directory even for a simple application.
 
-* Services generated using Foreman contain [$PORT variable](http://ddollar.github.io/foreman/#PROCFILE) in their names (and it's [undocumented](http://ddollar.github.io/foreman/#SYSTEMD-EXPORT) logic). For example for Procfile and formation `web=1,worker=2` (from example above), exported services with Foreman will be: `sample_app-web@2500.service`, `sample_app-worker@2600.service` and `sample_app-worker@2601.service`. My opinion about this approach: it's complicated. Why is there required PORT variable in the services names? Procsd following one rule: simplicity. For export it uses static service files (that means for each process will be generated it's own service file) and services names have predictable, Heroku-like names.
+* Services generated using Foreman contain the [$PORT variable](http://ddollar.github.io/foreman/#PROCFILE) in their names (and it's [undocumented](http://ddollar.github.io/foreman/#SYSTEMD-EXPORT) logic). For example, for Procfile and formation `web=1,worker=2` (from the example above), exported services with Foreman will be: `sample_app-web@2500.service`, `sample_app-worker@2600.service` and `sample_app-worker@2601.service`. My opinion about this approach: it's complicated. Why is the PORT variable required in the service names? Procsd follows one rule: simplicity. For export it uses static service files (that means for each process its own service file will be generated) and service names are predictable, Heroku-like.
 
 * Procsd export can provide additional stop/restart commands for each service (see _Notes_ below).
 
-* To delete existing app services from Systemd, there is `procsd destroy` command. It is doing the following: stop services if they are running, delete all required systemd files from systemd directory, and restart systemd (`daemon-reload`). This command especially useful while testing, when you need frequently create/update configuration.
+* To delete existing app services from Systemd, there is the `procsd destroy` command. It does the following: stops services if they are running, deletes all required systemd files from the systemd directory, and restarts systemd (`daemon-reload`). This command is especially useful while testing, when you need to frequently create/update configuration.
 
 
 ## Notes
@@ -519,9 +516,9 @@ Execute: journalctl --no-pager --no-hostname --all --output short-iso -n 3 --uni
 
 * You can use extended format of processes commands inside `procsd.yml` to provide additional restart/stop commands for each process:
 
-> All possible options: `ExecStart`, `ExecReload` and `ExecStop`
+> All possible options: [ExecStart](https://www.freedesktop.org/software/systemd/man/systemd.service.html#ExecStart=) (default command to start a process), [ExecReload](https://www.freedesktop.org/software/systemd/man/systemd.service.html#ExecReload=), and [ExecStop](https://www.freedesktop.org/software/systemd/man/systemd.service.html#ExecStop=).
 
-> If procsd.yml has `processes:` option defined, then content of Procfile will be ignored
+> If procsd.yml has `processes:` option defined, then content of Procfile (if it exists) will be ignored.
 
 ```yml
 app: sample_app
@@ -534,6 +531,18 @@ processes:
 
 Why? For example default Ruby on Rails application server [Puma](http://puma.io/) supports [Phased or Rolling restart](https://github.com/puma/puma/blob/master/docs/restart.md#normal-vs-hot-vs-phased-restart) feature. If you provide separate `ExecReload`command for a process, then this command will be called while executing `$ procsd restart` by systemd instead of just killing and starting process again.
 
+* Another option you can provide for each process is [RuntimeMaxSec](https://www.freedesktop.org/software/systemd/man/systemd.service.html#RuntimeMaxSec=). It is used to automatically restart a process every N period of time. Could be useful for worker types of processes (like Sidekiq) where process memory could increase while running:
+
+> Example values for RuntimeMaxSec: `30s` (30 seconds), `5m` (5 minutes), `3h` (3 hours), `1d` (1 day).
+
+```yml
+app: sample_app
+processes:
+  web:
+    ExecStart: bundle exec rails server -p $PORT
+    RuntimeMaxSec: 12h
+```
+
 * If you use Nginx integration but default Nginx requests timeout (60s) is too small for you, [you can set a custom timeout](https://serverfault.com/a/777753) in the global Nginx config.
 
 
@@ -542,7 +551,7 @@ Why? For example default Ruby on Rails application server [Puma](http://puma.io/
 https://github.com/vifreefly/capistrano-procsd
 
 
-## ToDo
+## TODO
 * Add `procsd update` command to quickly update changed configuration (application units, nginx config, etc), instead of calling two separate commands (`procsd destroy` and `procsd create`)
 
 

data/lib/procsd/cli.rb

@@ -58,7 +58,7 @@ module Procsd
         end
 
         if execute %w(sudo systemctl daemon-reload)
-          say("Reloaded configuraion (daemon-reload)", :green)
+          say("Reloaded configuration (daemon-reload)", :green)
         end
         say("App services were stopped, disabled and removed", :green)
 
@@ -105,44 +105,67 @@ module Procsd
     end
 
     desc "start", "Start app services"
-    def start
+    def start(service_name = nil)
       preload!
       say_target_not_exists and return unless target_exist?
 
-      say "Note: app target #{target_name} already started/active" if target_active?
-      if execute %W(sudo systemctl start #{target_name})
-        say("Started app services (#{target_name})", :green)
+      if service_name
+        full_name = to_full_name(service_name)
+        say "Note: app service #{full_name} already started/active" if service_active?(full_name)
+        if execute %W(sudo systemctl start #{full_name} --all)
+          say("Started app service (#{full_name})", :green)
+        end
+      else
+        say "Note: app target #{target_name} already started/active" if target_active?
+        if execute %W(sudo systemctl start #{target_name})
+          say("Started app services (#{target_name})", :green)
+        end
       end
     end
 
     desc "stop", "Stop app services"
-    def stop
+    def stop(service_name = nil)
       preload!
       say_target_not_exists and return unless target_exist?
 
-      say "Note: app target #{target_name} already stopped/inactive" if !target_active?
-      if execute %W(sudo systemctl stop #{target_name})
-        say("Stopped app services (#{target_name})", :green)
+      if service_name
+        full_name = to_full_name(service_name)
+        say "Note: app service #{full_name} already stopped/inactive" if !service_active?(full_name)
+        if execute %W(sudo systemctl stop #{full_name} --all)
+          say("Stopped app service (#{full_name})", :green)
+        end
+      else
+        say "Note: app target #{target_name} already stopped/inactive" if !target_active?
+        if execute %W(sudo systemctl stop #{target_name})
+          say("Stopped app services (#{target_name})", :green)
+        end
       end
     end
 
     desc "restart", "Restart app services"
-    def restart
+    def restart(service_name = nil)
       preload!
       say_target_not_exists and return unless target_exist?
 
-      # If one of the child services of a target has `ExecReload` and `ReloadPropagatedFrom`
-      # options defined, then use `reload-or-restart` to call all services (not the main target)
-      # because of systemd bug https://github.com/systemd/systemd/issues/10638
-      success =
-        if has_reload?
-          execute %W(sudo systemctl reload-or-restart #{app_name}-* --all)
-        else
-          execute %W(sudo systemctl restart #{target_name})
+      if service_name
+        full_name = to_full_name(service_name)
+        if execute %W(sudo systemctl reload-or-restart #{full_name} --all)
+          say("Restarted app service (#{full_name})", :green)
         end
+      else
+        # If one of the child services of a target has `ExecReload` and `ReloadPropagatedFrom`
+        # options defined, then use `reload-or-restart` to call all services (not the main target)
+        # because of systemd bug https://github.com/systemd/systemd/issues/10638
+        success =
+          if has_reload?
+            execute %W(sudo systemctl reload-or-restart #{app_name}-* --all)
+          else
+            execute %W(sudo systemctl restart #{target_name})
+          end
 
-      if success
-        say("Restarted app services (#{target_name})", :green)
+        if success
+          say("Restarted app services (#{target_name})", :green)
+        end
       end
     end
 
@@ -159,7 +182,7 @@ module Procsd
         command = %w(systemctl status --no-pager --output short-iso --all)
       end
 
-      command << (options["target"] ? target_name : "#{app_name}-#{service_name}*")
+      command << (options["target"] ? target_name : to_full_name(service_name))
       execute command, type: :exec
     end
 
@@ -266,7 +289,7 @@ module Procsd
       generator.generate_units(save: true)
 
       if execute %w(sudo systemctl daemon-reload)
-        say("Reloaded configuraion (daemon-reload)", :green)
+        say("Reloaded configuration (daemon-reload)", :green)
       end
 
       enable
@@ -376,10 +399,18 @@ module Procsd
       system "systemctl", "is-active", "--quiet", target_name
     end
 
+    def service_active?(service_name)
+      system "systemctl", "is-active", "--quiet", service_name
+    end
+
     def target_name
       "#{app_name}.target"
     end
 
+    def to_full_name(service_name)
+      "#{app_name}-#{service_name}*"
+    end
+
     def app_name
       @config[:app]
     end
@@ -389,7 +420,7 @@ module Procsd
 
       raise ConfigurationError, "Config file procsd.yml doesn't exists" unless File.exist? "procsd.yml"
       begin
-        procsd = YAML.load(ERB.new(File.read "procsd.yml").result)
+        procsd = YAML.safe_load(ERB.new(File.read "procsd.yml").result)
       rescue => e
         raise ConfigurationError, "Can't read procsd.yml: #{e.inspect}"
       end
@@ -402,7 +433,7 @@ module Procsd
         msg = "Procfile doesn't exists. Define processes in procsd.yml or create Procfile"
         raise ConfigurationError, msg unless File.exist? "Procfile"
         begin
-          procfile = YAML.load_file("Procfile")
+          procfile = YAML.safe_load_file("Procfile")
         rescue => e
           raise ConfigurationError, "Can't read Procfile: #{e.inspect}"
         end

data/lib/procsd/generator.rb

@@ -99,7 +99,7 @@ module Procsd
       b.local_variable_set(:config, conf)
       template_path = File.join(File.dirname(__FILE__), "templates/#{template_name}.erb")
       content = File.read(template_path)
-      ERB.new(content, nil, "-").result(b)
+      ERB.new(content, trim_mode: "-").result(b)
     end
 
     def write_file!(dest_path, content)

data/lib/procsd/templates/service.erb

@@ -18,6 +18,10 @@ ExecStop=/bin/bash -lc '<%= stop %>'
 ExecReload=/bin/bash -lc '<%= reload %>'
 <% end -%>
 
+<% if max_sec = config["commands"]["RuntimeMaxSec"] -%>
+RuntimeMaxSec=<%= max_sec %>
+<% end -%>
+
 Restart=always
 RestartSec=1
 TimeoutStopSec=30

data/lib/procsd/version.rb

@@ -1,3 +1,3 @@
 module Procsd
-  VERSION = "0.5.4"
+  VERSION = "0.6.0"
 end

data/procsd.gemspec

@@ -22,12 +22,12 @@ Gem::Specification.new do |spec|
   spec.bindir        = "exe"
   spec.executables   = "procsd"
   spec.require_paths = ["lib"]
-  spec.required_ruby_version = ">= 2.3.0"
+  spec.required_ruby_version = ">= 3.2.0"
 
   spec.add_dependency "thor"
   spec.add_dependency "dotenv"
 
-  spec.add_development_dependency "bundler", "~> 1.16"
-  spec.add_development_dependency "rake", "~> 10.0"
+  spec.add_development_dependency "bundler", ">= 1.16"
+  spec.add_development_dependency "rake", ">= 10.0"
   spec.add_development_dependency "minitest", "~> 5.0"
 end

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: procsd
 version: !ruby/object:Gem::Version
-  version: 0.5.4
+  version: 0.6.0
 platform: ruby
 authors:
 - Victor Afanasev
-autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-05-12 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: thor
@@ -42,28 +41,28 @@ dependencies:
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '1.16'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '1.16'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '10.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '10.0'
 - !ruby/object:Gem::Dependency
@@ -80,7 +79,6 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '5.0'
-description: 
 email:
 - vicfreefly@gmail.com
 executables:
@@ -109,7 +107,6 @@ homepage: https://github.com/vifreefly/procsd
 licenses:
 - MIT
 metadata: {}
-post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
@@ -117,15 +114,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.3.0
+      version: 3.2.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.0.3
-signing_key: 
+rubygems_version: 4.0.1
 specification_version: 4
 summary: Manage your application processes in production hassle-free like Heroku CLI
   with Procfile and Systemd