I'm trying to run commands remotely with ruby's net ssh. I need the output as well as the exit code. There are other stackoverflow threads on this topic but the accepted solutions do not work. Some people suggested using net-ssh-shell gem but as I tried it, I got an error saying that this package is in conflict with the version of my net-ssh package...
Here is what I have:
Net::SSH.start(remote_ip, username,:keys => [ssh_key_path], :verbose => :debug) do |ssh|
cmd = "blah bla"
result = ssh.exec!(cmd)
puts result
end
It works but it does not raise an exception if it fails. I've also tried using the channel to retrieve the exit code but it always returns 0:
channel.on_request("exit-status") do |ch,data|
exit_code = data.read_long
end
Please help me out. I've already tried several things based on wrong info on the internet.
If you are using Ruby 1.9+ I would suggest Open3.popen3:
i, o, e, t = Open3.popen3("ssh ... remote_user#remote_host remote_command")
i.write ... # if your command needs input
output = o.read
error = e.read
status = t.value.exitstatus
Related
Is there an Expect equivalent gem for Ruby?
I tried searching on code.google and rubygems.org, but sadly it did not show up.
FYI: Expect is a Unix automation and testing tool, written by Don Libes as an extension to the Tcl scripting language, for interactive applications such as telnet, ftp, passwd, fsck, rlogin, tip, ssh, and others.
Ruby comes with the PTY module for setting up pseudoterminals to drive interactive command line applications. With it comes an expect method that allows you to interact with an application kinda like Expect. For learning how to use expect, I found "What to expect from the Ruby expect library?" helpful.
As far as gems go, maybe checkout greenletters which is supposed to improve upon PTY + expect (although I haven't tried it myself).
I recently spent quite a bit of time struggling with this issue (I am stuck with 1.8.7). I found this question, this blog post and this forum thread really useful.
At the end this is my application code if anyone is interested in a little example (pass the password to rpm when signing packages):
def run_interactive command, password, promt
output = ''
begin
r, w, pid = PTY.spawn(command)
puts r.expect(promt)
sleep(0.5)
w.puts(password)
begin
r.each { |l| output += l }
rescue Errno::EIO
end
$?.exitstatus
Process.wait(pid)
rescue PTY::ChildExited => e
$stderr.puts "The child process #{e} exited! #{$!.status.exitstatus}"
end
output
end
password = "mypassword"
command = "rpm --define '_signature gpg' --define '_gpg_name #{key_id}' --addsign #{package}"
promt = %r{.*: }
expected = %r{good}
output = run_interactive(command, password, promt)
if output.match(expected)
puts output
else
abort "Error: expected: '#{expected}' got '#{output}'"
end
It has little error checking but it was all I needed.
Edit: Update the code with Process.wait(pid) to make sure it finishes before continuing and add comment about this being for 1.8.7.
checkout this rubygem: https://github.com/abates/ruby_expect. It could handle some small task for you. from its official example, it's enough to 'enter password' and login and interactive with local script.
here is an example that update the git code (which is authenticated with password):
require 'rubygems'
require 'ruby_expect'
def update_code
password = 'your password here'
exp = RubyExpect::Expect.spawn('git pull', :debug => true)
exp.procedure do
each do
expect /password: / do
send password
end
end
end
end
update_code
just run the code above, and your will see like this:
$ ruby update_code.rb
shensiwei#gforge.1ver??.net's password:
remote: Counting objects: 133, done.
remote: Compressing objects: 100% (84/84), done.
remote: Total 85 (delta 62), reused 0 (delta 0)
Unpacking objects: 100% (85/85), done.
for more example and details, please dive into its source code.
expect4r seems to do what you are asking for, though it is made specific for connections to Cisco and Juniper devices.
Perhaps even better is yax as this is "yet another expect".
RExpect
From the project's website:
RExpect is a drop in replacement for the expect.rb module in the
standard library that is faster and more robust, cabable of driving
many devices simultaneously.
parley is another one you can try, (written by me). It is inspired by Perl expect.
I tried to use ruby Net::Telnet to connect windows 2008 and execute some commands. But it failed.
if execute
tn = Net::Telnet::new("Host"=>"walnutserver","Port"=>2300,"Prompt"=> /C:.*>/)
tn.login("user","pass")
tn.cmd("dir")
tn.cmd("dir")
the first tn.cmd("dir") is success but the second one throws exceptions.And then subsequent commands all failed. After experimenting,I found that any windows command will cause this.
Exceptions:
Timeout::Error: timed out while waiting for more data
from c:/troy/data/chef/chef-client11/chef/embedded/lib/ruby/1.9.1/net/telnet.rb:558:in `waitfor'
from c:/troy/data/chef/chef-client11/chef/embedded/lib/ruby/1.9.1/net/telnet.rb:697:in `cmd'
from (irb):20
from c:/troy/data/chef/chef-client11/chef/embedded/bin/irb:12:in `<main>'
use sock.sysread() method to read responding, I found that terminal is blocked and display dir\r\n0x00More?
Buf if execute
tn = Net::Telnet::new("Host"=>"walnutserver","Port"=>2300,"Prompt"=> /C:.*>/)
tn.login("user","pass")
tn.cmd("ls")
tn.cmd("uname")
It't running normally. ls、uname are some linux commands brought by chef which installed in target machine.
ruby version：ruby 1.9.3p286 (2012-10-12) [i386-mingw32]
I found someone else
asking the same question on Stackoverflow, but he didn't get the solution. http://www.ruby-forum.com/topic/1516840
Need your help.
solved. the reason is ruby net/telnet library use error newline seperator. Must be EOL(CR+LF) but CR+NULL . But I don't know who make the bug,windows or ruby? I write a monkey patch as below:
class Net::Telnet
def print(string)
string = string.gsub(/#{IAC}/no, IAC + IAC) if #options["Telnetmode"]
if #options["Binmode"]
self.write(string)
else
if #telnet_option["BINARY"] and #telnet_option["SGA"]
self.write(string.gsub(/\n/n, CR))
elsif #telnet_option["SGA"]
self.write(string.gsub(/\n/n, EOL)) ### fix here. reaplce CR+NULL bY EOL
else
self.write(string.gsub(/\n/n, EOL))
end
end
end
end
What is the correct way of dealing with input prompts that are triggered by commands I run through Capistrano?
One example is the iptables-persistent package I install using aptitude. Despite the --no-gui flag, a prompt still comes up asking me to confirm how I want things configured.
Is there a way to pass parameters through command line to avoid such prompts?
I found and was able to implement this very helpful handle_command_with_input method from:
https://github.com/nesquena/cap-recipes/blob/master/lib/cap_recipes/tasks/utilities.rb
def handle_command_with_input(local_run_method, shell_command, input_query, response=nil)
send(local_run_method, shell_command, {:pty => true}) do |channel, stream, data|
if data =~ input_query
if response
logger.info "#{data} #{"*"*(rand(10)+5)}", channel[:host]
channel.send_data "#{response}\n"
else
logger.info data, channel[:host]
response = ::Capistrano::CLI.password_prompt "#{data}"
channel.send_data "#{response}\n"
end
else
logger.info data, channel[:host]
end
end
end
None of the code is mine. Gracias a Nesquena.
I'm writing some code which takes a file, passes that file to one of several binaries for processing, and monitors the conversion process for errors. I've written and tested the following routine on OSX but linux fails for reasons about which I'm not clear.
#run the command, capture the output so it doesn't display
PTY.spawn(command) {|r,w,pid|
until r.eof? do
##mark
puts r.readline
end
}
The command that runs varies quite a lot and the code at the ##mark has been simplified into a local echo in an attempt to debug the problem. The command executes and the script prints the expected output in the terminal and then throws an exception.
The error it produces on Debian systems is: Errno::EIO (Input/output error - /dev/pts/0):
All of the command strings I can come up with produce that error, and when I run the code without the local echo block it runs just fine:
PTY.spawn(command) {|r,w,pid|}
In either case the command itself executes fine, but it seems like debian linux isn't sending eof up the pty. The doc pages for PTY, and IO on ruby-doc don't seem to lend any aid here.
Any suggestions? Thanks.
-vox-
So I had to go as far as reading the C source for the PTY library to get really satisfied with what is going on here.
The Ruby PTY doc doesn't really say what the comments in the source code say.
My solution was to put together a wrapper method and to call that from my script where needed. I've also boxed into the method waiting on the process to for sure exit and the accessing of the exit status from $?:
# file: lib/safe_pty.rb
require 'pty'
module SafePty
def self.spawn command, &block
PTY.spawn(command) do |r,w,p|
begin
yield r,w,p
rescue Errno::EIO
ensure
Process.wait p
end
end
$?.exitstatus
end
end
This is used basically the same as PTY.spawn:
require 'safe_pty'
exit_status = SafePty.spawn(command) do |r,w,pid|
until r.eof? do
logger.debug r.readline
end
end
#test exit_status for zeroness
I was more than a little frustrated to find out that this is a valid response, as it was completely undocumented on ruby-doc.
It seems valid for Errno::EIO to be raised here (it simply means the child process has finished and closed the stream), so you should expect that and catch it.
For example, see the selected answer in Continuously read from STDOUT of external process in Ruby and http://www.shanison.com/2010/09/11/ptychildexited-exception-and-ptys-exit-status/
BTW, I did some testing. On Ruby 1.8.7 on Ubuntu 10.04, I don't get a error. With Ruby 1.9.3, I do. With JRuby 1.6.4 on Ubuntu in both 1.8 and 1.9 modes, I don't get an error. On OS X, with 1.8.7, 1.9.2 and 1.9.3, I don't get an error. The behavior is obviously dependent on your Ruby version and platform.
As answered here and here, EIO can be avoided by keeping a file descriptor to the pty slave device open in the parent process.
Since PTY.spawn closes the slave file descriptor passed to the child process, a simple workaround is to open a new one. For example:
PTY.spawn("ls") do |r, w, pid|
r2 = File.open(r.path)
while IO.select([r], [], [], 1)
puts r.gets
end
r2.close
end
ruby-doc.org says this since ruby 1.9:
# The result of read operation when pty slave is closed is platform
# dependent.
ret = begin
m.gets # FreeBSD returns nil.
rescue Errno::EIO # GNU/Linux raises EIO.
nil
end
Ok, so now I get this behavior is "normal" on Linux, but that means it's a little tricky to get the output of a PTY. If you do m.read it reads everything and then throws it away and raises Errno::EIO. You really need to read the content chunk by chunk with m.readline. And even then you risk losing the last line if it doesn't end with "\n" for whatever reason. To be extra safe you need to read the content byte by byte with m.read(1)
Additional note about the effect of tty and pty on buffering: it's not the same as STDOUT.sync = true (unbuffered output) in the child process, but rather it triggers line buffering, where output is flushed on "\n"
I'm trying to connect, using Net::SSH, to a server that immediately after
login executes a script that requires input from user. The user has to enter "1" or "2" and will receive some data via in the terminal afterwards.
My problem is that, although I am able to connect, I can not figure out a way to send "1\n" to the server and to receive the output.
The following code stops at "INFO -- net.ssh.connection.session[80906b74]: channel_open_confirmation: 0 0 0 32768".
Using channel.exec( "1\n" ) instead of channel.send_data unsurprisingly does not work either.
Net::SSH.start('host', 'user', :password => "pass", :auth_methods => ["password"], :verbose => :debug) do |session|
session.open_channel do |channel|
channel.on_data do |ch, data|
STDOUT.print data
end
channel.send_data( "1\n")
end
session.loop
end
Any ideas, anyone?
Thanks in advance
Can you verify that your send_data call is happening after you get the prompt from the remote server? Try constructing a channel.on_data block around your send_data call so that you can verify that you get the expected prompt from the server before you send a response.
You might not want to be using exec here. From the docs for Net::SSH::Connection::Channel:
Sends a channel request asking that
the given command be invoked.
You are wanting to send a text string to reply to a prompt, not invoke a command. The docs show exec being used to send full CLI commands like "ls -l /home".
Instead, send_data is probably what you want. The docs show it used to send arbitrary text such as channel.send_data("the password\n"). Note, however, this sentence in the docs:
Note that it does not immediately send
the data across the channel, but
instead merely appends the given data
to the channel‘s output buffer,
preparatory to being packaged up and
sent out the next time the connection
is accepting data.
You might want to take a look at channel.request_pty. It appears to be designed for interaction with a console-based application.
If you are trying to (in essence) script an SSH session that you would normally do manually, you may find it easier to use an expect-like interface (for example, a gem like sshExpect might be worth a try).
Thank you all for the pointers. I have been able to put my finger on the problem – besides using channel.request_pty it was also necessary to request a shell. The following finally works as expected:
Net::SSH.start('host', 'user', :password => "pass", :auth_methods => ["password"]) do |session|
session.open_channel do |channel|
channel.request_pty do |ch, success|
raise "Error requesting pty" unless success
ch.send_channel_request("shell") do |ch, success|
raise "Error opening shell" unless success
end
end
channel.on_data do |ch, data|
STDOUT.print data
end
channel.on_extended_data do |ch, type, data|
STDOUT.print "Error: #{data}\n"
end
channel.send_data( "1\n" )
session.loop
end
end
I'm not terribly familiar with the Net::SSH libs so I can't help with that per-se but it sounds like you could achieve what you want using Capistrano.
For example I have a capistrano task which connects to a remote server, runs a command which expects input and then continues. Capistrano takes care of the remote i/o. Maybe that could be a solution for you?
Hope it helps!
If I execute "1\n" in a shell the reply I get is: bash: 1: command not found
If I execute echo "1" I get: 1
Are you sure you want to try to execute the text you are sending? Maybe you were looking for something like:
output = ""
Net::SSH.start('host', 'user', :password => "pass") do |ssh|
output = ssh.exec! "echo 1"
end
puts output
I'm not proficient with that lib, but on SSH you can open multiple channels. Maybe the server only responds to the first default channel and if you open another one you get a fresh shell.