Menu
Amazon AppStream 2.0
Developer Guide

Troubleshooting

The following are possible problems you may have while trying to use Amazon AppStream 2.0.

I cannot connect to the Internet from my image builder.

Image builders cannot communicate to the Internet by default. To resolve this issue, launch your image builder in a VPC subnet that has Internet access. You can enable Internet access from your VPC subnet using a NAT gateway. Alternatively, you can configure an Internet gateway in your VPC, and attach an Elastic IP address to your image builder. For more information, see Network Settings for Fleet and Image Builder Instances.

When I tried installing my application, I see an error that the operating system version is not supported.

Only applications that can be installed on Windows Server 2012 R2 can be added to an AppStream 2.0 image. Check if your application is supported on Microsoft Windows Server 2012 R2.

When I connect to my image builder, I see a login screen asking me to enter Ctrl + Alt + Delete to log in. However, my local machine intercepts the key strokes.

Your client may intercept certain key combinations locally instead of sending them to the image builder session. To reliably send the Ctrl + Alt + Delete key combination to the image builder, choose Admin Commands, Send Ctrl + Alt + Delete. The Admin Commands menu is available on the top right corner of the image builder session toolbar.

When I switched between admin and test modes, I saw a request for a password. I don’t know how to get a password.

AppStream 2.0 usually logs you into the user mode that you choose automatically. On some occasions, the switch may not happen automatically. If a password is requested, choose Admin Commands, Log me in. This sends a one-time password, securely, to your image builder and pastes it into the Password field.

I get an error when I add my installed application.

Check if your application type is supported. You can add applications of the types .exe, .lnk, and .bat.

Check if your application is installed under the C:\Users folder hierarchy. Any application installed under the C:\Users is not supported. Select a different installation folder under C:\ when installing the application.

I accidentally quit a background service on the image builder and got disconnected. I am now unable to connect to that image builder.

Try stopping the image builder, restarting it and connecting to it again. If the problem persists, you need to launch (create) a new image builder. Do not stop any background services running on the image builder instance. Doing so may interrupt your image builder session or interfere with the image creation.

The application fails to launch in test mode.

Check if your application requires elevated user privileges or any special permissions that are usually available only to an administrator. The Image Builder Test mode has the same limited permissions on the image builder instance as your end users have on an AppStream 2.0 test fleet. If your applications require elevated permissions, they do not launch in the Image Builder Test mode.

The application could not connect to a network resource in my VPC.

Check if the image builder was launched in the correct VPC subnet. You may also need to verify that the route tables in your VPC are configured to allow a connection.

I customized my image builder desktop, but my changes are not available when connecting to a session after launching a fleet from the image I created.

Changes that are saved as part of a local user session, such as wallpaper, are not persisted when creating an image. To persist any local user session changes, add them to the local group policy on the image builder instance.

My application is missing a command line parameter when launching.

You can provide a command line parameter when using image builder to add an application to an image. If the launch parameters for the application do not change for each user, you can enter them while adding an application to the image in the image builder instance.

If the launch parameters are different for every launch, you can pass them programmatically when using the CreateStreamingURL API. Set the sessionContext and applicationID parameters in the API fields. The sessionContext is included as a command line option when launching the application.

If the launch parameters need to be computed on the fly, you can launch your application using a script. You can parse the sessionContext parameter within your script before launching your application with a computed parameter.

I am unable to use my image with a fleet after installing an antivirus application.

You can install any tools, including antivirus programs, on your AppStream 2.0 stack by using the image builder before creating an image. However, these programs should not block any network ports or stop any processes that are used by the AppStream 2.0 service. We recommend testing your application in Image Builder Test mode before creating an image and attempting to use it with a fleet.

My image creation failed.

Verify that you did not make any changes to AppStream 2.0 services before starting the image creation. Try creating your image again; if it fails, contact AWS Support.

SAML federation is not working. The user is not authorized to view AppStream 2.0 applications.

This may happen because the policy of the IAM role, which is assumed by the federated user who is accessing an AppStream 2.0 stack, does not include permissions to the stack ARN. Edit the role permissions to include the stack ARN. For more information, see Enabling Single Sign-on Access to AppStream 2.0 Using SAML 2.0 and Troubleshooting SAML 2.0 Federation with AWS in the IAM User Guide.

After federating from an ADFS portal, my streaming session doesn’t start. I am getting the error “Sorry connection went down”.

Set the claim rule's Incoming Claim Type for the NameID SAML attribute to UPN and try the connection again.

I get an invalid redirect URI error.

This error happens due to a malformed or invalid AppStream 2.0 stack relay state URL. Make sure that the relay state configured in your federation setup is the same as the stack relay state available in stack details through the AppStream 2.0 console. If they are the same and the problem still persists, contact AWS Support. For more information, see Enabling Single Sign-on Access to AppStream 2.0 Using SAML 2.0.

My stack's Home Folders aren't working correctly.

Home Folder backup to the S3 bucket can have problems under the following scenarios:

  • There is no Internet connectivity from the streaming instance, or there is no access to the private Amazon S3 VPC endpoint, if applicable.

  • Network bandwidth consumption is too high. For example, multiple large files being downloaded or streamed by the user while the service is trying to back up the Home Folder containing large files to Amazon S3.

  • The administrator deleted the bucket created by the service.

  • The administrator incorrectly edited the Amazon S3 permissions of the AmazonAppStreamServiceAccess service role.

For more information, see the Amazon Simple Storage Service Console User Guide and Amazon Simple Storage Service Developer Guide.