DSH (Distributed Shell) – Run Commands on Multiple Linux Servers

Systems Administrators are well aware of the importance of being able to monitor and administer numerous machines in a short amount of time, preferably with minimal physical movement. Whether it’s a small cloud environment or a large server cluster, the ability to centrally manage computers is essential.

To partly accomplish this, I am going to show you how to use a nifty little tool called DSH (dancer’s shell/distributed shell) that allows a user to run commands across multiple machines.

What is DSH?

DSH is short for ‘Distributed Shell‘ or ‘Dancer’s Shell‘, which is a freely available tool on most major distributions of Linux but can be easily built from a source if your distribution does not include it in its package repository.

Install DSH (Distributed Shell) in Linux

We are going to assume a Debian / Ubuntu environment for the scope of this tutorial. If you are using another distribution, please substitute the appropriate commands for your package manager.

On Debian-based distributions, you can install DSH using the following apt command in the terminal.

sudo apt install dsh

On RHEL-based distributions, you need to compile it from source tar balls, but before doing so, make sure to compile and install the ‘libdshconfig‘ library.

wget http://www.netfort.gr.jp/~dancer/software/downloads/libdshconfig-0.20.10.cvs.1.tar.gz
tar xfz libdshconfig*.tar.gz 
cd libdshconfig-*
./configure ; make
sudo make install

Then compile dsh and install.

wget https://www.netfort.gr.jp/~dancer/software/downloads/dsh-0.22.0.tar.gz
tar xfz dsh-0.22.0.tar.gz
cd dsh-*
./configure ; make 
sudo make install

How to Use DSH in Linux

The main configuration file “/etc/dsh/dsh.conf” (For Debian) and “/usr/local/etc/dsh.conf” (for Red Hat) is pretty straightforward, but since rsh is an unencrypted protocol, we are going to use SSH as the remote shell.

Using the text editor of your choice, find this line:

remoteshell =rsh

and change it to:

remoteshell =ssh

There are other options you can pass in here if you choose, and there are plenty of them to find on the dsh man page. For now, we are going to accept the defaults and have a look at the next file, /etc/dsh/machines.list (for Debian).

For Red Hat-based systems you need to create a file called “machines.list” in the “/usr/local/etc/” directory.

The syntax here is pretty easy. All you have to do is enter in a machine’s credentials (Hostname, IP Address, or FQDN) one per line.

Note: When accessing more than one machine simultaneously, it would behoove you to set up key-based password-less SSH on all of your machines. Not only does this provide ease of access, but security-wise, it hardens your machine as well.

My “/etc/dsh/machines.list” or “/usr/local/etc/machines.list” file says:

Once you have entered the credentials of the machines you wish to access, let’s run a simple command like uptime to all of the machines.

dsh -aM -c uptime

Sample Output: 05:11:58 up 40 days, 51 min, 0 users, load average: 0.00, 0.01, 0.05 05:11:47 up 13 days, 38 min, 0 users, load average: 0.00, 0.01, 0.05

So, What Does the ‘dsh’ Command Do?

Pretty simple. First, we ran dsh and passed the “-a” option to it, which says to send the “uptime” command to “ALL” of the machines listed in “/etc/dsh/machines.list“.

Next, we specified the “-M” option, which says to return the “machine name” (specified in “/etc/dsh/machines.list“) along with the output of the uptime command. (Very useful for sorting when running a command on a number of machines).

The “-c” option stands for “command to be executed” in this case, “uptime“.

DSH can also be configured with groups of machines in the “/etc/dsh/groups/” file, where is a file with a list of machines in the same format as the “/etc/dsh/machines.list” file. When running dsh on a group, specify the groupname after the “-g” option.

For Red Hat-based systems you need to create a folder called “groups” in the “/usr/local/etc/” directory. In that “groups” directory you create a file called “cluster“.

For example, run the “w” command on all machines listed in the “cluster” group file “/etc/dsh/groups/cluster” or “/usr/local/etc/groups/cluster“.

dsh -M -g cluster -c w

DSH provides much more flexibility, and this tutorial only scratches the surface. Aside from executing commands, DSH can be used to transfer files, install software, add routes, and much more.

To a Systems Administrator tasked with the responsibility of a large network, it is invaluable.

Hey TecMint readers,

Exciting news! Every month, our top blog commenters will have the chance to win fantastic rewards, like free Linux eBooks such as RHCE, RHCSA, LFCS, Learn Linux, and Awk, each worth $20!

Learn more about the contest and stand a chance to win by sharing your thoughts below!

Rob Krul
Rob is an avid user of Linux and Open Source Software, with over 15 years experience in the tech geek universe. Aside from experimenting with the many flavors of Linux, he enjoys working with BSDs, Solaris, and OS X. He currently works as an Independent IT Contractor.

Each tutorial at TecMint is created by a team of experienced Linux system administrators so that it meets our high-quality standards.

Join the TecMint Weekly Newsletter (More Than 156,129 Linux Enthusiasts Have Subscribed)
Was this article helpful? Please add a comment or buy me a coffee to show your appreciation.


Leave a Reply
  1. If you have 1000 servers, It will ask 1000 times to put yes/no to known hosts and for first time to add ssh-key again it will ask 1000 times for password, then what is the use of this???

    • @Aurelien,

      Thanks for informing us about clustershell, we will definitely test run right away and write a detailed article on this, till then stay tuned..

  2. For those of you who are asking about ports, I can suggest setting up config file in ~/.ssh/config. The file should contain something like:

    Host server1.net
    user root
    Port 22
    IdentityFile ~/.ssh/your_id
    Host server2.net
    User root
    Port 12345
    IdentityFile ~/.ssh/your_id
    Host server3.net
    User root
    Port 54321
    IdentityFile ~/.ssh/your_id

    • @Norm,

      Thanks for informing us about that pdsh took, let me give a try and see how it is useful than dsh, will test it and write a detailed article on this, till then stay tuned..

  3. How can use this with other SSH ports than 22. this only works with ssh port 22. How can use this with other ports

  4. Hi! Do you know if they make any plugins to protect against hackers?
    I’m kinda paranoid about losing everything I’ve worked hard on. Any recommendations?

  5. the -c flag does not mean the “command to be executed” , it means to run the command concurrently on all servers.

    • @niraj

      The easiest way to set up ssh connections to other hosts is to use a ~/.ssh/config file. The syntax is pretty simple:

      Host (host)
      HostName (host name)
      IdentityFile ~/.ssh/your_id_file_rsa
      Port (port number)
      User (the username you use with SSH to login)

  6. On CentOS/RHEL it’s just as easy as in Debian, no need to compile anything. Just:

    yum –enablerepo=epel install pdsh

    On Fedora, you don’t even need EPEL, it’s included in the standard repository:

    yum install pdsh

  7. Good article, But im getting this error

    Pseudo-terminal will not be allocated because stdin is not a terminal.

    System info
    Distributor ID: elementary OS
    Description: elementary OS Luna
    Release: 0.2
    Codename: luna


Got Something to Say? Join the Discussion...

Thank you for taking the time to share your thoughts with us. We appreciate your decision to leave a comment and value your contribution to the discussion. It's important to note that we moderate all comments in accordance with our comment policy to ensure a respectful and constructive conversation.

Rest assured that your email address will remain private and will not be published or shared with anyone. We prioritize the privacy and security of our users.