When you issue the instance start call, AWS OpsWorks engages the following process:

1. OpsWorks builds instance user-data based on the OS type.

2. OpsWorks issues an EC2 RunInstances call, along with EC2 user-data.

4. During the instance boot, OpsWorks runs user-data and installs the opsworks-agent, downloaded from S3.
Note:

For newer OS versions (Ubuntu 16.04 LTS, CentOS 7), Systemd manages the agent. Monit is used for monitoring and managing the agent on other supported OS versions.

5. After the instance is online and the agent is running, OpsWorks continuously checks for new commands and executes them.

6. The Setup lifecycle event is executed using Chef. When the Setup lifecycle event has completed, Chef logs are made available in the OpsWorks console and on the instance.

A failure at any of these points causes an instance to be in the start_failed state. To troubleshoot this, check the following:

1. Check the log file user-data.log at /var/log/aws/opsworks/ to find if the scripts in user-data of the instance have completed successfully.

2. Check installer.log to see if the OpsWorks agent installed successfully.
Note: The most common cause of failed agent installations is an incorrectly configured VPC. Make sure instances in your VPC can access the Internet; otherwise, the instance will attempt to download and install the agent, and the download will fail.

3. Make sure the agent is running.
For older OS versions: