EC2 Server Instance Lifecycle

It is importnt to understand the different states of an EC2 server instance which can initially be a little confusing. The server can be in 3 main states or a number of transition states:
  • Pending (transition): The server has been instantiated and is about to run for the first time or a stopped server has been started and is about to move into the running state
  • Running (main): The server is active and is being billed at the usual rate depending on the actual offering
  • Rebooting (transition): The server is cycling i.e. restarting to return to the running state
  • Stopping (transition): The server is moving from the running state to the stopped state
  • Stopped (main): The server is not running and can be restarted to resume the running state.. In this state the server does not attract any charge. However, other attached resources may attract charges e.g. any volumes will continue to be charged according to the space allocated per GB/month, static IPs actually attract a higher charge when the server they are assigned to is shutdown or terminated. If you are shutting down a server for a long time and do not need the IP to remain the same, it is good practice to release any static IPs.
  • Shutting down (transition): The server is moving from the running state to the terminated state
  • Terminated (main): The server is gone to be seen no more. It cannot be recovered or restarted. To start up a server you must instantiate another server.
EC2 server instance lifecycle