consuldemocracy/installer

Capistrano Deployment Error

FabriceHecht opened this issue · 15 comments

FabriceHecht commented

Hi,

Unfortunately I have some problems deploying with Capistrano. I followed the instructions and got to the point:

branch=stable cap production deploy

But unfortunately the whole thing gives an error message afterwards. It seems to me that Capistrano is not using the login credentials as requested.

00:00 rvm1:hook
      01 mkdir -p undefined/rvm1scripts/
#<Thread:0x0000562b38f31c68@/home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/runners/parallel.rb:10 run> terminated with exception (report_on_exception is true):
Traceback (most recent call last):
	17: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/runners/parallel.rb:12:in `block (2 levels) in execute'
	16: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/abstract.rb:31:in `run'
	15: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/abstract.rb:31:in `instance_exec'
	14: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/rvm1-capistrano3-1.4.0/lib/rvm1/tasks/capistrano3.rake:5:in `block (3 levels) in <top (required)>'
	13: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/abstract.rb:80:in `execute'
	12: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/abstract.rb:148:in `create_command_and_execute'
	11: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/abstract.rb:148:in `tap'
	10: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/abstract.rb:148:in `block in create_command_and_execute'
	 9: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/netssh.rb:130:in `execute_command'
	 8: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/netssh.rb:177:in `with_ssh'
	 7: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/connection_pool.rb:63:in `with'
	 6: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/backends/connection_pool.rb:63:in `call'
	 5: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/net-ssh-6.1.0/lib/net/ssh.rb:251:in `start'
	 4: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/net-ssh-6.1.0/lib/net/ssh.rb:251:in `new'
	 3: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/net-ssh-6.1.0/lib/net/ssh/transport/session.rb:73:in `initialize'
	 2: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:631:in `tcp'
	 1: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:227:in `foreach'
/home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:227:in `getaddrinfo': getaddrinfo: Servname not supported for ai_socktype (SocketError)
	1: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/runners/parallel.rb:11:in `block (2 levels) in execute'
/home/fabrice/.rbenv/versions/2.6.6/lib/ruby/gems/2.6.0/gems/sshkit-1.21.0/lib/sshkit/runners/parallel.rb:15:in `rescue in block (2 levels) in execute': Exception while executing as undefined@undefined: getaddrinfo: Servname not supported for ai_socktype (SSHKit::Runner::ExecuteError)
(Backtrace restricted to imported tasks)
cap aborted!
SSHKit::Runner::ExecuteError: Exception while executing as undefined@undefined: getaddrinfo: Servname not supported for ai_socktype


Caused by:
SocketError: getaddrinfo: Servname not supported for ai_socktype

Tasks: TOP => rvm1:hook
(See full trace by running task with --trace)
javierm commented

Hi @FabriceHecht 😄.

Could you provide the contents of your config/deploy-secrets.yml file?

FabriceHecht commented

HI @javierm

fabrice@fabrice-VirtualBox:~/consul/config$ cat deploy-secrets.yml
production:
deploy_to: "/home/fabrice/consul"
server1: "192.168.178.85"
user: "fabrice"
db_server: "localhost"
server_name: "192.168.178.85"
ssh_port: "22"
full_app_name: "consul"


fabrice@fabrice-VirtualBox:~/consul/config$ cat deploy.rb
# config valid only for current version of Capistrano
lock "~> 3.14.1"

def deploysecret(key)
  @deploy_secrets_yml ||= YAML.load_file("config/deploy-secrets.yml")[fetch(:stage).to_s]
  @deploy_secrets_yml.fetch(key.to_s, "undefined")
end

set :rails_env, fetch(:stage)
set :rvm1_map_bins, -> { fetch(:rvm_map_bins).to_a.concat(%w[rake gem bundle ruby]).uniq }

set :application, "consul"
set :deploy_to, deploysecret(:deploy_to)
set :ssh_options, port: deploysecret(:ssh_port)

set :repo_url, "https://github.com/FabriceHecht/consul.git"

set :revision, `git rev-parse --short #{fetch(:branch)}`.strip

set :log_level, :info
set :pty, true
set :use_sudo, false

set :linked_files, %w[config/database.yml config/secrets.yml]
set :linked_dirs, %w[.bundle log tmp public/system public/assets public/ckeditor_assets]

set :keep_releases, 5

set :local_user, ENV["USER"]

set :puma_conf, "#{release_path}/config/puma/#{fetch(:rails_env)}.rb"

set :delayed_job_workers, 2
set :delayed_job_roles, :background

set :whenever_roles, -> { :app }

namespace :deploy do
  Rake::Task["delayed_job:default"].clear_actions
  Rake::Task["puma:smart_restart"].clear_actions

  after :updating, "rvm1:install:rvm"
  after :updating, "rvm1:install:ruby"

  after "deploy:migrate", "add_new_settings"

  after :publishing, "setup_puma"

  after :published, "deploy:restart"
  before "deploy:restart", "puma:restart"
  before "deploy:restart", "delayed_job:restart"
  before "deploy:restart", "puma:start"

  after :finished, "refresh_sitemap"

  desc "Deploys and runs the tasks needed to upgrade to a new release"
  task :upgrade do
    after "add_new_settings", "execute_release_tasks"
    invoke "deploy"
  end
end

task :refresh_sitemap do
  on roles(:app) do
    within release_path do
      with rails_env: fetch(:rails_env) do
        execute :rake, "sitemap:refresh:no_ping"
      end
    end
  end
end

task :add_new_settings do
  on roles(:db) do
    within release_path do
      with rails_env: fetch(:rails_env) do
        execute :rake, "settings:add_new_settings"
      end
    end
  end
end

task :execute_release_tasks do
  on roles(:app) do
    within release_path do
      with rails_env: fetch(:rails_env) do
        execute :rake, "consul:execute_release_tasks"
      end
    end
  end
end

desc "Create pid and socket folders needed by puma"
task :setup_puma do
  on roles(:app) do
    with rails_env: fetch(:rails_env) do
      execute "mkdir -p #{shared_path}/tmp/sockets; true"
      execute "mkdir -p #{shared_path}/tmp/pids; true"
    end
  end
end
javierm commented

@FabriceHecht I take the production server is on the same local network as your computer; is that right? Did you use the installer to install CONSUL with the default deploy user? If not, which user did you create on the production server?

FabriceHecht commented

@javierm

I take the production server is on the same local network as your computer; is that right?

Yes, the production server runs (for testing purposes) on an old laptop and the local installation is on my usual working laptop.
They both are in the same network. SSH it self works fine and without password request. I also tried to deploy my fork from my local installation to a server in our data center, which also doesn't work with the same error.

Did you use the installer to install CONSUL with the default deploy user? If not, which user did you create on the production server?

I used the installer for the production installation. The name of my user is fabrice. As I mentioned, I also tried it on a second machine in the datacenter. For this second Installation I used the default deploy user, but with the same result.

javierm commented

In the second installation, were you also using deploy_to: "/home/fabrice/consul" and user: "fabrice" in the deploy-secrets.yml fiel or did you modified those lines so deploy was used?

FabriceHecht commented

@javierm
Yes, I changed it to deploy in the second installation.

javierm commented

Could you paste the error messages you got in the second installation? Right now I'm not sure on which machine the error is taking place since on both machines the user is fabrice 🤔.

FabriceHecht commented

@javierm

00:00 rvm1:hook
      01 mkdir -p undefined/rvm1scripts/
#<Thread:0x0000562028b9d310@/home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:10 run> terminated with exception (report_on_exception is true):
Traceback (most recent call last):
	17: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:12:in `block (2 levels) in execute'
	16: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:31:in `run'
	15: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:31:in `instance_exec'
	14: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/rvm1-capistrano3-1.4.0/lib/rvm1/tasks/capistrano3.rake:5:in `block (3 levels) in <top (required)>'
	13: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:80:in `execute'
	12: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:148:in `create_command_and_execute'
	11: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:148:in `tap'
	10: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:148:in `block in create_command_and_execute'
	 9: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/netssh.rb:130:in `execute_command'
	 8: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/netssh.rb:177:in `with_ssh'
	 7: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/connection_pool.rb:63:in `with'
	 6: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/backends/connection_pool.rb:63:in `call'
	 5: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/net-ssh-6.1.0/lib/net/ssh.rb:251:in `start'
	 4: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/net-ssh-6.1.0/lib/net/ssh.rb:251:in `new'
	 3: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/net-ssh-6.1.0/lib/net/ssh/transport/session.rb:73:in `initialize'
	 2: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:631:in `tcp'
	 1: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:227:in `foreach'
/home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:227:in `getaddrinfo': getaddrinfo: Servname not supported for ai_socktype (SocketError)
	1: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:11:in `block (2 levels) in execute'
/home/fabrice/.rvm/gems/ruby-3.0.0/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:15:in `rescue in block (2 levels) in execute': Exception while executing as undefined@undefined: getaddrinfo: Servname not supported for ai_socktype (SSHKit::Runner::ExecuteError)
(Backtrace restricted to imported tasks)
cap aborted!
SSHKit::Runner::ExecuteError: Exception while executing as undefined@undefined: getaddrinfo: Servname not supported for ai_socktype


Caused by:
SocketError: getaddrinfo: Servname not supported for ai_socktype

Tasks: TOP => rvm1:hook
(See full trace by running task with --trace)
javierm commented 
3: from /home/fabrice/.rvm/gems/ruby-3.0.0/gems/net-ssh-6.1.0/lib/net/ssh/transport/session.rb:73:in `initialize'
2: from /home/fabrice/.rbenv/versions/2.6.6/lib/ruby/2.6.0/socket.rb:631:in `tcp'

@FabriceHecht Are you using both rvm and rbenv on your working laptop? 🤔 Also RVM seems to be using Ruby 3.0.0, while CONSUL depends on Ruby 2.6.6.

FabriceHecht commented

@javierm Yes, I am using both. Should In uninstall one of it? How can I tell RVM to use Ruby 2.6.6?

javierm commented

They might be incompatible, so yes, uninstall one of them unless you're sure you need both for some reason.

How can I tell RVM to use Ruby 2.6.6?

It should do it automatically because it reads the .ruby-version file present in CONSUL. It's probably failing now because both rvm and rbenv are installed 🤔.

javierm commented

By the way, on your local machine, it doesn't matter whether you use rbenv or RVM. On the production server CONSUL expects RVM though, and that's what the installer sets up.

FabriceHecht commented

@javierm I uninstalled rbenv and had to bundle again and set the default ruby version for rvm and now its running into the same error but with rvm instead of rbenv.

fabrice@fabrice-VirtualBox:~/consul$ cap production deploy
00:00 rvm1:hook
      01 mkdir -p undefined/rvm1scripts/
#<Thread:0x00005648a8e0a488@/home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:10 run> terminated with exception (report_on_exception is true):
Traceback (most recent call last):
	17: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:12:in `block (2 levels) in execute'
	16: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:31:in `run'
	15: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:31:in `instance_exec'
	14: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/rvm1-capistrano3-1.4.0/lib/rvm1/tasks/capistrano3.rake:5:in `block (3 levels) in <top (required)>'
	13: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:80:in `execute'
	12: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:148:in `create_command_and_execute'
	11: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:148:in `tap'
	10: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/abstract.rb:148:in `block in create_command_and_execute'
	 9: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/netssh.rb:130:in `execute_command'
	 8: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/netssh.rb:177:in `with_ssh'
	 7: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/connection_pool.rb:63:in `with'
	 6: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/backends/connection_pool.rb:63:in `call'
	 5: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/net-ssh-6.1.0/lib/net/ssh.rb:251:in `start'
	 4: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/net-ssh-6.1.0/lib/net/ssh.rb:251:in `new'
	 3: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/net-ssh-6.1.0/lib/net/ssh/transport/session.rb:73:in `initialize'
	 2: from /home/fabrice/.rvm/rubies/ruby-2.6.6/lib/ruby/2.6.0/socket.rb:631:in `tcp'
	 1: from /home/fabrice/.rvm/rubies/ruby-2.6.6/lib/ruby/2.6.0/socket.rb:227:in `foreach'
/home/fabrice/.rvm/rubies/ruby-2.6.6/lib/ruby/2.6.0/socket.rb:227:in `getaddrinfo': getaddrinfo: Servname not supported for ai_socktype (SocketError)
	1: from /home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:11:in `block (2 levels) in execute'
/home/fabrice/.rvm/gems/ruby-2.6.6/gems/sshkit-1.21.2/lib/sshkit/runners/parallel.rb:15:in `rescue in block (2 levels) in execute': Exception while executing as undefined@undefined: getaddrinfo: Servname not supported for ai_socktype (SSHKit::Runner::ExecuteError)
(Backtrace restricted to imported tasks)
cap aborted!
SSHKit::Runner::ExecuteError: Exception while executing as undefined@undefined: getaddrinfo: Servname not supported for ai_socktype


Caused by:
SocketError: getaddrinfo: Servname not supported for ai_socktype

Tasks: TOP => rvm1:hook
(See full trace by running task with --trace)
javierm commented

Just realized something:

fabrice@fabrice-VirtualBox:~/consul/config$ cat  -secrets.yml
production:
deploy_to: "/home/fabrice/consul"
server1: "192.168.178.85"
user: "fabrice"
db_server: "localhost"
server_name: "192.168.178.85"
ssh_port: "22"
full_app_name: "consul"

Are the file contents literally these ones? I'm asking because every line except the first one should start with two spaces 🤔.

javierm commented

@FabriceHecht We've added more precise instructions about configuring the deploy-secrets.yml file. Hope that helps!