Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

This is sort of a follow-up question to an unanswered question I have regarding administration of Cloudera cluster, but I figure generalizing the question to all of Ubuntu may help me get an answer.

I want to be able to start/stop the same service on all my cluster's nodes. The only way I see how to do this is to create keys for the root user on my master node to each of the other nodes. Since I follow the Ubuntu recommendation to not use the root account, and instead use a sudo'ing user, I am hesitant to touch the root account (i.e. adding the keys).

Perhaps an alternative would be to add keys for my user to each of the boxes, then use some command to run sudo remotely?

I know there has to be other scenarios where people do something similar. What is the standard practice for this type of cluster management?

share|improve this question
Is there any particular reason you don't use something like Puppet instead of manually connecting to each system? – Alex Holst Sep 16 '11 at 19:57
Puppet's new to me, but it looks like a promising solution. If you got any favorite how-to's (outside of their documentation) or ways you've used, please let me know. Thanks! – Dolan Antenucci Sep 17 '11 at 4:32
up vote 2 down vote accepted

I agree that you should explore puppet. But to be more specific, puppet agents "pull" instructions from a central puppet master. To "push" commands from a central server out to N nodes, one might use mcollective (also a puppetlabs project).

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.