SSH keys

From ArchWiki
Revision as of 17:11, 13 April 2009 by Thelucster (talk | contribs) (Cleanup first section (full cleanup in progress))
Jump to: navigation, search

Template:I18n links start Template:I18n entry Template:I18n entry Template:I18n entry Template:I18n links end

What are SSH Keys?

By using SSH Keys (a public and private key to be precise), you can easily connect to a server, or multiple servers, without having to enter your password for each system.

Note that it is possible to setup your keys without a passphrase, however that is unwise as if anyone gets hold of your key they can use it. This guide describes how to setup your system so that passphrases are remembered securely.

Generating SSH Keys

If you don't already have OpenSSH installed, install it now as it is not installed by default on Arch.

# pacman -S openssh

The keys can then be generated by running the ssh-keygen command as a user:

$ ssh-keygen -b 1024 -t dsa
Generating public/private dsa key pair.
Enter file in which to save the key (/home/mith/.ssh/id_dsa):
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/mith/.ssh/id_dsa.
Your public key has been saved in /home/mith/.ssh/
The key fingerprint is:
x6:68:xx:93:98:8x:87:95:7x:2x:4x:x9:81:xx:56:94 mith@middleearth

It will prompt you for a location (which you should leave as the default), however the passphrase is the important bit! I hopefully need not tell you the rules of a good passphrase?

So what did we just do? We generated a 1024 bit long (-b 1024) public/private dsa (-t dsa) key pair with the ssh-keygen command. You can also create a rsa key (-t rsa). You can leave out the bit length parameter (default bit length is 1024).

Now there are two paths which you can follow or "Choose your destiny" (as the guy from Mortal Kombat would say):
a) the short, easy but insecure way: use no passphrase, have easy access to your remote-server, feel paranoid about someone stealing your private key
b) the a bit longer, slightly less comfortable but secure way: use a passphrase, use ssh-agent and feel safe

Step 2: copy your public key to the remote server {SAME for path A and B}

mith@middleearth||[[~]]:~ > scp .ssh/

Copy the public key ( to your remote server via scp (note the : at the end of the server address. That way the file actually ends up in our server home directory but you can specify another path if you like.

Step 2+3 - The easy way

mith@middleearth||[[~]]:~ > ssh-copy-id -i .ssh/


NOTE: that as the username on both machines is the same you can omit it in the command (i.e. instead of

Step 3: login to the remote computer and put your key into the right place {SAME for path A and B}

mith@middleearth||[[~]]:~ > ssh's password:
-bash-2.05b$ mkdir .ssh/
-bash-2.05b$ cat >> .ssh/authorized_keys
-bash-2.05b$ rm
-bash-2.05b$ chmod 700 .ssh
-bash-2.05b$ chmod 600 .ssh/authorized_keys

We connect to our remote server and use cat to add the content of to authorized_keys which is in the hidden directory .ssh. NOTE: In case you get an error because the .ssh directory doesn't exist you should simply create it (mkdir .ssh).
Now we remove the public key (rm and set the correct permissions for .ssh and authorized_keys.
Log out and re-login. Depending on the path you chose, you will/won't be asked for a passphrase. Path a ends here...

...path b continues...

Now what makes this path almost as easy as a no-passphrase key? The magic word is ssh-agent.. What it does is basically asking you once every session for the passphrase of your private key and every time you would have to type it in, ssh-agent does it for you. ssh-agent is included in the openssh package so no trouble there...

mith@middleearth||[[~]]:~ > ssh-agent
SSH_AUTH_SOCK=/tmp/ssh-vEGjCM2147/agent.2147; export SSH_AUTH_SOCK;
echo Agent pid 2148;

When you run ssh-agent it will print out what environment variables it would use... Well to make ssh-agent use these variables run

mith@middleearth||[[~]]:~ > eval `ssh-agent`
Agent pid 2157

The process id will vary for you of course. Adding eval `ssh-agent` to your .bashrc is an option so it's started every time you create a new shell.
Now that the ssh-agent is running, we need to tell it that we have a private key and where that is.

mith@middleearth||[[~]]:~ > ssh-add .ssh/id_dsa
Enter passphrase for .ssh/id_dsa:
Identity added: .ssh/id_dsa (.ssh/id_dsa)

We were asked for our passphrase, entered it, that's all. Now you can login to your remote server without having to enter your password while your private key is password-protected. Sweet isn't it? The only downside is that a new instance of ssh-agent needs to be created for every new console (shell) you open, that means you have to run ssh-add every time again on each console. There is a workaround to that with a program or rather a script called keychain which is covered in the next section.

Using keychain

Keychain manages one or more specified private keys. When initialized it will ask for the passphrase for the private key(s) and store it. That way your private key is password protected but you won't have to enter your password over and over again.

Install keychain from the extra repo:

# pacman -S keychain

Edit your ~/.bashrc and add the following lines:

/usr/bin/keychain -Q -q ~/.ssh/id_dsa
[[ -f $HOME/.keychain/$HOSTNAME-sh ]] && source $HOME/.keychain/$HOSTNAME-sh

If necessary, replace ~/.ssh/id_dsa with ~/.ssh/id_rsa. For those using a non-Bash shell, see keychain --help or man keychain for details on other shells.

Close your shell and open it again. Keychain should come up and if it's your first run it will ask your for the passphrase of the specified private key.

Using ssh-agent and x11-ssh-askpass

You need to start the ssh-agent everytime you start a new Xsession. The ssh-agent will be closed when the X session ends.
Install x11-ssh-askpass which will ask your passphrase everytime you open a new Xsession:
sudo pacman -S x11-ssh-askpass

Prepend this into your ~/.xsession :

eval `/usr/bin/ssh-agent`
SSH_ASKPASS=/usr/lib/openssh/x11-ssh-askpass ssh-add < /dev/null
# then the end of the file with for example "exec dwm"

SSH connection control

In ~/.ssh/config, add the following lines:

host *
   controlmaster auto
   controlpath /tmp/ssh-%r@%h:%p

What this does is set a "master control" socket when you make an SSH connection. The socket is named based on the controlpath setting (%r = username, %h = hostname, %p = port).

This master socket is used for each successive connection after the first, as long as one connection still exists. That is, if you connect via ssh, a socket named /tmp/ is created. If you then ssh again to the same host, the socket is found and the remote ssh session is told to spawn a new shell. This shell does not require a login, and spawns immediately, as you're already logged in.


The above procedure is a bit complicated when using PuTTY on Windows since PuTTY can't directly use keys generated by ssh-keygen. The private key needs to be converted using PuTTYgen which you can find here. The procedure is then as follows:

  1. Generate the key pair with ssh-keygen on you Linux computer (you can log in with your usual username/password using PuTTY)
  2. Add the public key to the ~/.ssh/authorized_keys file
  3. Move the private key to the Windows machine
  4. Load the private key with PuTTYgen and click Save private key. This will convert the key so that PuTTY can use it.
  5. Start PuTTY, go to SSH->Auth and find the private key. Then simply connect to your Linux machine. You will be prompted for your username and passphrase (if you chose to enter one when you generated the keys).

Note that reversing the procedure, that is, generating the key pair with PuTTYgen and converting the public key with ssh-keygen, will NOT work.

Useful links / information