AWS RoboMaker
Developer Guide

Troubleshooting Deployments

This section provides more information about common issues encountered when deploying a robot application to a fleet.

My Deployment Failed

See the following topics for common solutions.

Is your robot part of a fleet?

A robot must be part of a fleet to receive a deployment. To check the status of your robots in the AWS RoboMaker console, expand Fleet management and then choose Robots. Robots that are registered to a fleet will include the Fleet name.

Is AWS IoT Greengrass running on your robot?

To configure and run the AWS IoT Greengrass core software, follow the steps in Module 1: Environment Setup for Greengrass, then follow the steps in Start AWS Greengrass on the Core Device.

Is a resource missing?

In the deployment details page, review the Failure reason. It will list the missing resource. Verify that the resource exists. For example, if the robot application is missing, it might have been deleted from the Amazon S3 location or the Amazon S3 etag information might be incorrect.

Did the AWS IoT Greengrass deployment encounter a problem?

In the deployment details page, review the Failure reason. It will contain more details. To troubleshoot, see Troubleshooting AWS IoT Greengrass.

Was the failure threshold breached?

Deployment will halt if the failure threshold is breached. You can raise the threshold to attempt to deploy to more of your fleet.

Is the deployment taking longer than expected?

Deployment time depends on the size of the robot application package. It also depends on the robot network conditions. If you have many robots and are deploying few concurrently, deployment might take longer.

Once started, a single robot deployment timeout is 5 hours. Use the deployment detail page to identify robots that have an active deployment. Use SSH to connect. Use the command ps aux | grep 'greengrass' to verify AWS IoT Greengrass is running. To troubleshoot, see Troubleshooting AWS IoT Greengrass.

Did your robot receive the deployment request?

Your robot might not receive the deployment request if AWS IoT Greengrass is not properly configured and running.

First, verify that your robot received the deployment request. SSH to the robot. Once connected, use s aux | grep 'greengrass' to see if AWS IoT Greengrass is running. check for errors in the log located at /greengrass/ggc/var/log/user/region/account/aws-robomaker-deployment-function_DO_NOT_DELETE.log.

If there are no errors in the log, make sure you have AWS IoT Greengrass version 1.7.0 or above installed. See Module 1: Environment Setup for Greengrass for more information.

Next, make sure AWS IoT Greengrass is running with the following command:

ps aux | grep 'greengrass'

If it is running, look at the AWS IoT Greengrass system logs located on the robot at /greengrass/ggc/var/log/system/runtime.log. See Troubleshooting AWS IoT Greengrass for additional troubleshooting information.

On this page: