Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

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

I have a server hosted on EC2. The server is built on the Linux AMI, based on CentOS. It is dedicated to a client project, and I connect via SSH. Sometimes, I forget to shut down the server when unused, therefore generating unnecessary costs.

How do I make the server shut down automatically on inactivity, for example after 30 min without an SSH connection?

Naturally, I am looking for the simplest solution. A check every hour or so would also be OK, just something that works reliably.

share|improve this question
up vote 4 down vote accepted

As of January 2013 Amazon CloudWatch provides an option to Use Amazon CloudWatch to Detect and Shut Down Unused Amazon EC2 Instances, see the introductory blog post Amazon CloudWatch - Alarm Actions for details on this functionality:

Today we are giving you the ability to stop or terminate your EC2 instances when a CloudWatch alarm is triggered. You can use this as a failsafe (detect an abnormal condition and then act) or as part of your application's processing logic (await an expected condition and then act). [emphasis mine]

Your use case is listed in section Failsafe Ideas specifically:

If you (or your developers) are forgetful, you can detect unused EC2 instances and shut them down. You could do this by detecting a very low load average for an extended period of time. This type of failsafe could be used to reduce your AWS bill by making sure that you are not paying for resources you're not actually using.

As outlined, this depends on being able to heuristically detect the appropriate condition triggering the alarm and stopping the instance - you might take it to the next level by means of Publishing Custom Metrics to CloudWatch based on the logged in SSH user count, idle time or else and gain more control/precision for the desired detection and shutdown process in turn.

share|improve this answer
As a first experiment, I set up an alarm which shuts down the instance when NetworkIn (averaged over 5 minutes) is below 1000 bytes (per minute) for more than 30 minutes. – feklee Mar 25 '13 at 23:26

If all you're looking for is to keep the server running while you're logged-in, and your session idle for less than n units of time, you can do so via a cron script which you mentioned, by monitoring the output of w(1) which reports IDLE time for logged-in user, which you can use.

share|improve this answer
w does indeed look interesting. To get the idle time in seconds for user ec2-user: w -hs ec2-user | sed -e 's/.* \([0-9]\+\.[0-9]\{2\}\)s .*/\1/' – feklee Mar 24 '13 at 13:02
I think he wanted to shut down automatically if nobody was logged in for a certain length of time? – Michael Hampton Mar 24 '13 at 15:28
@MichaelHampton I'm the only user. – feklee Mar 24 '13 at 15:39
@feklee If you aren't logged in, then "nobody is logged in" :) – Michael Hampton Mar 24 '13 at 15:40
@abbe If nobody logs on, then it's time to format and reinstall. – Michael Hampton Mar 24 '13 at 17:14

You could create a cron job script right on the instance itself that uses a command like

netstat | grep ssh | grep ESTABLISHED

and if no result is returned write that to file, then the cron tries again and if it returns no results again the script runs this.

/sbin/shutdown -h now
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.