This error in Capistrano (> 3) prevents me from deploying my Rails application. My Capistrano config works fine on another environment.
DEBUG [6c7d7bed] Command: if test ! -d /opt/my_app/releases/20140429175529; then echo "Directory does not exist '/opt/my_app/releases/20140429175529'" 1>&2; false; fi
cap aborted!
NoMethodError: undefined method `recv' for #<IO:fd 49>
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/buffered_io.rb:65:in `fill'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/packet_stream.rb:86:in `next_packet'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:178:in `block in poll_message'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:173:in `loop'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:173:in `poll_message'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:210:in `block in wait'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:208:in `loop'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:208:in `wait'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh/transport/session.rb:87:in `initialize'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh.rb:200:in `new'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/net-ssh-2.8.0/lib/net/ssh.rb:200:in `start'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/connection_pool.rb:24:in `create_or_reuse_connection'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:173:in `ssh'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:126:in `block in _execute'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:123:in `tap'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:123:in `_execute'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:66:in `execute'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/abstract.rb:75:in `within'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/capistrano-3.2.1/lib/capistrano/tasks/deploy.rake:207:in `block (3 levels) in <top (required)>'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:54:in `instance_exec'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/backends/netssh.rb:54:in `run'
/Users/such/.rvm/gems/ruby-2.1.1@my_app/gems/sshkit-1.4.0/lib/sshkit/runners/parallel.rb:12:in `block (2 levels) in execute'
Tasks: TOP => deploy:set_current_revision
(See full trace by running task with --trace)
The deploy has failed with an error: #<NoMethodError: undefined method `recv' for #<IO:fd 49>>
I'm using a proxy to connect to my remote servers (on both environment). SSH connection is working perfectly.
set :ssh_options, proxy: Net::SSH::Proxy::Command.new('ssh user@gateway -W %h:%p')
Has anyone ever experienced this issue?
Workaround
My weak understanding of the error is that after the ssh negotiation, sometimes the socket returned to net-ssh is a file descriptor which is unexpected. I have added:
gem 'sshkit', github: 'msimonin/sshkit', branch: 'gateway'
to my Gemfile so that the ssh connection is only established once.
Related configuration is:
SSHKit.config.backend.configure do |ssh|
ssh.gateway = 'user@gateway'
end
It works fine this way. I can't wait for the pull request to be merged.