How to effectively manage a large group of droplets?

September 23, 2014 1.7k views


I own a company that is about to start hosting a number of servers [droplets] for bus companies... let's say 10 droplets for now. Each droplet has almost exactly the same configuration, the only difference being content in a MySQL database. We separate each company onto its own droplet to ensure that we can easily allocate the right droplet based on the size of the company.

My question is, let's say some crazy bug is found in Apache... an urgent patch is required. If there are 10 or more droplets all with Apache that need updating, it would mean I'd have to log into each droplet and update each one. Sure, in reality this is only going to take 10 minutes... but what if a config file needed editing on every single one?

Am I missing something here... is there an easier way to manage all the droplets so they have the same Apache build if you know what I mean?

I am relatively new to cloud hosting, Digital Ocean seems like a fantastic service but I just need to know whether I'm thinking along the right lines.


  • Urgent firefighting
    See clusterssh that replicates your commands to many ssh sessions.

    Serious approach
    Use a provisioning tool.

    • Ansible — very simple, no need to have anything at all on the managed machines, can leverage your existing scripts / playbooks.
    • SaltStack — more complex, but still rather lean and streamlined, and better documented than Ansible.

    I's say that Chef or Puppet would be overkill for 10 servers with a simple config.

  • Thanks very much! Ansible looks perfect.

1 Answer


take a look at provisioning tools like Chef, Puppet or Sprinkle. You can find some tutorials here in DO Community.

Have another answer? Share your knowledge.