Cloudformation for AMI with GPU fails with vCPU limit of 0?
0
Hi,
I'm using the following cloudformation template to create AMIs with GPU:
https://github.com/johnewart/ec2-batch-gpu-cloudformation/blob/master/gpu-ami-batch.yaml
I have made sure to use correct instance type by editing the following section:
** InstanceType:
Description: Instance type to launch EC2 instances.
Type: String
Default: p2.xlarge
AllowedValues: [p2.xlarge]**
and used
aws --region us-east-1 cloudformation create-stack .....
to use proper region on my submission. I have also made sure that I have a limit of 8 VCPUs on that region with no EC2s currently running (all terminated) under that region. I keep getting the following error:
"You have requested more vCPU capacity than your current vCPU limit of 0 allows for the instance bucket that the specified instance type belongs to. "
Where could be the problem as I do have the required vCPUs assigned in the region?
Are these answers helpful? Upvote the correct answer to help the community benefit from your knowledge.
0
Check your EC2 Service Quotas in the region - hopefully this link works.
When your account is first created p instances have 0 quota and this grows as you use your account. You can request a limit increase - more details here
Hi Peter, Thank you for your quick answer. I already have a limit of 8 vCPUs for my P instances in us-east-1.
Does using
aws --region us-east-1 cloudformation ...
mean that my resources get created in us-east-1 or if I don't explicitly specify a region in my .yml file they may get created in the default region us-east-2?
Hi Peter, Thank you for your quick answer. I already have a limit of 8 vCPUs for my P instances in us-east-1. Does using aws --region us-east-1 cloudformation ... mean that my resources get created in us-east-1 or if I don't explicitly specify a region in my .yml file they may get created in the default region us-east-2?