Troubleshooting
Important
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the updated Studio experience. For information about using the Studio Classic application, see Amazon SageMaker Studio Classic.
Important
Custom IAM policies that allow Amazon SageMaker Studio or Amazon SageMaker Studio Classic to create Amazon SageMaker resources must also grant permissions to add tags to those resources. The permission to add tags to resources is required because Studio and Studio Classic automatically tag any resources they create. If an IAM policy allows Studio and Studio Classic to create resources but does not allow tagging, "AccessDenied" errors can occur when trying to create resources. For more information, see Provide permissions for tagging SageMaker AI resources.
AWS managed policies for Amazon SageMaker AI that give permissions to create SageMaker resources already include permissions to add tags while creating those resources.
This section shows how to troubleshoot common problems in Amazon SageMaker Studio.
Cannot delete Code Editor, based on Code-OSS, Visual Studio Code - Open Source or JupyterLab application
This issue occurs when a user creates an application from Amazon SageMaker Studio that is only available in Studio, then reverts to the Studio Classic experience as their default. As a result, the user cannot delete an application for Code Editor, based on Code-OSS, Visual Studio Code - Open Source or JupyterLab because they can't access the Studio UI.
To resolve this issue, notify your administrator so that they can delete the application manually using the AWS Command Line Interface (AWS CLI).
EC2InsufficientCapacityError
This issue occurs when you try to run a space and AWS does not currently have enough available on-demand capacity to fulfill your request.
To resolve this issue, complete the following.
-
Wait a few minutes, then resubmit your request. Capacity can shift frequently.
-
Run the space with an alternate instance size or type.
Note
Capacity is available in different Availability Zones. To maximize capacity availability for users, we recommend setting up subnets in all Availability Zones. Studio retries all available Availability Zones for the domain.
Instance type availability differs between regions. For a list of supported instances
types per Region, see Amazon SageMaker AI
pricing
The following table lists instance families and their recommended alternatives.
| Instance family | CPU Type | vCPUs | Memory (GiB) | GPU type | GPUs | GPU Memory (GiB) | Recommended alternative |
|---|---|---|---|---|---|---|---|
| G4dn | 2nd Generation Intel Xeon Scalable Processors | 4 to 96 | 16 to 384 | NVIDIA T4 Tensor Core | 1 to 8 | 16 per GPU | G6 |
| G5 | 2nd generation AMD EPYC processors | 4 to 192 | 16 to 768 | NVIDIA A10G Tensor core | 1 to 8 | 24 per GPU | G6e |
| G6 | 3rd generation AMD EPYC processors | 4 to 192 | 16 to 768 | NVIDIA L4 Tensor Core | 1 to 8 | 24 per GPU | G4dn |
| G6e | 3rd generation AMD EPYC processors | 4 to 192 | 32 to 1536 | NVIDIA L40S Tensor Core | 1 to 8 | 48 per GPU | G5, P4 |
| P3 | Intel Xeon Scalable Processors | 8 to 96 | 61 to 768 | NVIDIA Tesla V100 | 1 to 8 | 16 per GPU (32 per GPU for P3dn) | G6e, P4 |
| P4 | 2nd Generation Intel Xeon Scalable processors | 96 | 1152 | NVIDIA A100 Tensor Core | 8 | 320 (640 for P4de) | G6e |
| P5 | 3rd Gen AMD EPYC processors | 192 | 2000 | NVIDIA H100 Tensor Core | 8 | 640 | P4de |
Insufficient limit (quota increase required)
This issue occurs when you get the following error when running a space. This error means that you have reached the limit on the number of instances of that type that you can launch in a Region. When you create your AWS account, we set default limits on the number of instances you can run in each Region.
Error when creating application for space: ... : The account-level service limit is X Apps, with current utilization Y Apps and a request delta of 1 Apps. Please use Service Quotas to request an increase for this quota.
To resolve this issue, request an instance limit increase for the Region that you are launching the space is. For more information, see Requesting a quota increase.
