Run a Command

Run one shell command on the servers you choose, follow each server’s output as it arrives, and review every exit code afterwards.

Before you start: At least one connected server, and a command that does not wait for input.

  1. 1

    Open Command Runner

    Command Runner is under Automation. It is not a terminal: the command runs to completion on each server exactly as typed, and a command that waits for input receives none. During the free trial, the page shows how many of its 25 runs are left.

  2. 2

    Enter the command and choose the account

    Choose Root to run it as root, directly or through sudo, or Signed-in account to run it as the account the platform connects with. Do not include passwords, tokens or keys: commands and their output are kept in the run history.

  3. 3

    Set a time limit

    Choose between one minute and one hour per server. A command still running at the limit is stopped on the server itself.

  4. 4

    Choose servers and run

    Select servers the same way as for a playbook: by server, group, tag, operating system or all. Commands have no dry run, so keeping Try one server first on matters most here.

  5. 5

    Run it again later

    Open a finished command run and choose Run again. The command, account and time limit are filled in, and you choose the servers again.