If a task that is executed as part of a build gets an error, like trying to delete a directory that doesn't exist, how do you capture that error so the build will also fail? Currently, any inline script commands that execute as part of task do not cause the build to fail even when the command gets an error. I am guessing that the script is currently returning a "zero" return code as long as it executes despite whether or not it fails. If this is the case, I could use an example of how to return a "non-zero" return code from an inline script task. Is there any documentation available about handling errors in build job tasks?