Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

'Max spot instance requests per region' limit check is out of date #502

Closed
jwu2 opened this issue Nov 5, 2020 · 5 comments
Closed

'Max spot instance requests per region' limit check is out of date #502

jwu2 opened this issue Nov 5, 2020 · 5 comments

Comments

@jwu2
Copy link

jwu2 commented Nov 5, 2020

Bug Report

Version

9.0.0

Installation Method

Docker

Supporting Software Versions

3.8.6

Actual Output

Limit of this check always returns 20.

Expected Output

Refer to https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-spot-limits.html, this is replaced by following six limits:

  • All F Spot Instance Requests
  • All G Spot Instance Requests
  • All Inf Spot Instance Requests
  • All P Spot Instance Requests
  • All Standard (A, C, D, H, I, M, R, T, Z) Spot Instance Requests
  • All X Spot Instance Requests

Testing Assistance

Yes

@nakkurt
Copy link

nakkurt commented Nov 23, 2020

I also ran into this and after speaking with AWS Support I can confirm that the 'Max spot instance requests per region' is no longer used by AWS.
Instead, AWS now manages Spot Instance limits in terms of the number of virtual central processing units (vCPUs) and the above 'per region' request limit is not a thing anymore.

Implication:
If your code uses awslimitchecker to check for 'Max spot instance requests per region', then you'd probably want to change that until awslimitchecker is updated to better match AWS.

Background:
latest AWS docs on this.

@jantman
Copy link
Owner

jantman commented Dec 2, 2020

Apologies for the delay in dealing with this.

So, it looks like the spot instance limits now match up to the on-demand EC2 limits, in that they're vCPU-based. Handling this for EC2 was a pretty big task (for reference, see #432 and https://awslimitchecker.readthedocs.io/en/latest/changes.html#changelog-8-0-0-vcpu-limits ). It's a bit of a pain, because China and GovCloud don't use the vCPU units yet, so two alternate code paths need to be maintained.

I'll do my best to work on this within the next few days or week.

@jantman
Copy link
Owner

jantman commented Dec 7, 2020

I've promised a release today that includes a lot of staged changes, and this isn't going to make it. I'm sorry. I'll try to get another out within the next week that includes this.

@skraga
Copy link

skraga commented Feb 18, 2021

We step into the same - adding of this will help us a lot with limits monitoring. We can help with China testing as we use both RoW and China partitions. Looking at AWS WebUI - China switched also to vCPU for Spots.

TagadaPoe added a commit to quortex/awslimitchecker that referenced this issue Jun 16, 2021
… limits

BREAKING CHANGE

As mentioned in issue jantman#502, the limit "Max spot instance requests per region" was removed from AWS and replaced by the following limits, based on the number of vCPU:
- All F Spot Instance Requests
- All G Spot Instance Requests
- All Inf Spot Instance Requests
- All P Spot Instance Requests
- All Standard (A, C, D, H, I, M, R, T, Z) Spot Instance Requests
- All X Spot Instance Requests

As a result, the usage reported by awslimitchecker is not relevant anymore.

In order to fix this, this commit replaces the call to the EC2 API describe_spot_instance_requests() with a call to the CloudWatch API, since the usage of Spot requests vCPUs is now available as a metric in CloudWatch.

In addition, the applied limits are now retrieved from the Service Quotas service.
antonincms pushed a commit to quortex/awslimitchecker that referenced this issue Jul 22, 2021
… limits

BREAKING CHANGE

As mentioned in issue jantman#502, the limit "Max spot instance requests per region" was removed from AWS and replaced by the following limits, based on the number of vCPU:
- All F Spot Instance Requests
- All G Spot Instance Requests
- All Inf Spot Instance Requests
- All P Spot Instance Requests
- All Standard (A, C, D, H, I, M, R, T, Z) Spot Instance Requests
- All X Spot Instance Requests

As a result, the usage reported by awslimitchecker is not relevant anymore.

In order to fix this, this commit replaces the call to the EC2 API describe_spot_instance_requests() with a call to the CloudWatch API, since the usage of Spot requests vCPUs is now available as a metric in CloudWatch.

In addition, the applied limits are now retrieved from the Service Quotas service.
@jantman
Copy link
Owner

jantman commented Aug 4, 2021

A fix for this has been released in 12.0.0, which is now live on PyPI and Docker Hub. Thank you so much!

@jantman jantman closed this as completed Aug 4, 2021
derrix060 pushed a commit to derrix060/awslimitchecker that referenced this issue Jun 28, 2024
… limits

BREAKING CHANGE

As mentioned in issue jantman#502, the limit "Max spot instance requests per region" was removed from AWS and replaced by the following limits, based on the number of vCPU:
- All F Spot Instance Requests
- All G Spot Instance Requests
- All Inf Spot Instance Requests
- All P Spot Instance Requests
- All Standard (A, C, D, H, I, M, R, T, Z) Spot Instance Requests
- All X Spot Instance Requests

As a result, the usage reported by awslimitchecker is not relevant anymore.

In order to fix this, this commit replaces the call to the EC2 API describe_spot_instance_requests() with a call to the CloudWatch API, since the usage of Spot requests vCPUs is now available as a metric in CloudWatch.

In addition, the applied limits are now retrieved from the Service Quotas service.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants