Immediate Fix
The fastest way to resolve the VcpuLimitExceeded error is to request a service quota increase via the AWS Management Console. This error occurs because your account has hit the regional “soft limit” for concurrent vCPUs.
Follow these steps to increase your limit:
- Log in to the AWS Service Quotas console.
- In the navigation pane, choose AWS services and search for “Amazon Elastic Compute Cloud (Amazon EC2)”.
- Search for the specific quota name, typically: “Running On-Demand Standard (A, C, D, H, I, M, R, T, Z) instances”.
- Select the quota and click Request increase at account-level.
- Enter the desired total vCPU value and click Request.
AWS Support usually approves these requests within 15 minutes to a few hours, though complex requests may take longer.
Technical Explanation
AWS implements quotas (formerly known as limits) to protect users from unexpected billing spikes and to ensure infrastructure stability. These limits are applied per AWS Region and per Instance Family.
The “Standard” quota covers the most common general-purpose, compute-optimized, and memory-optimized instances. If you are launching G or P instances (GPU), those have separate vCPU quotas that must be managed independently.
| Quota Name | Instance Families Included | Applied Per |
|---|---|---|
| Standard instances | A, C, D, H, I, M, R, T, Z | Region |
| F instances | F | Region |
| G instances | G | Region |
| P instances | P | Region |
Alternative Methods
If you cannot wait for a quota increase or want to verify your current usage programmatically, you can use the AWS CLI. This is helpful for DevOps engineers who need to automate environment checks.
Checking Quotas via AWS CLI
Run the following command to see your current vCPU quota for standard instances in your current region:
aws service-quotas get-service-quota \
--service-code ec2 \
--quota-code L-1216C47AOptimize Existing Resources
Before requesting an increase, check if you have idle or “zombie” instances running in your account. Terminating unused instances immediately frees up vCPU capacity within your current limit.
Switch Regions
Since vCPU limits are regional, you may find that you have available capacity in a different AWS Region (e.g., switching from us-east-1 to us-west-2). This is a viable temporary workaround for non-latency-sensitive workloads.
Finally, consider using Spot Instances. While they have their own separate quotas, they can sometimes be used to bypass On-Demand limits if your application is fault-tolerant.