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

UnsupportedAvailabilityZoneException: lack of capacity to support cluster #118

Closed
niksrc opened this issue Jul 13, 2018 · 17 comments · Fixed by #131
Closed

UnsupportedAvailabilityZoneException: lack of capacity to support cluster #118

niksrc opened this issue Jul 13, 2018 · 17 comments · Fixed by #131

Comments

@niksrc
Copy link

niksrc commented Jul 13, 2018

unable to create cluster control plane: UnsupportedAvailabilityZoneException: Cannot create cluster '****-**' because us-east-1c, the targeted availability zone, does not currently have sufficient capacity to support the cluster. Retry and choose from these availability zones: us-east-1a, us-east-1b, us-east-1d

@errordeveloper
Copy link
Contributor

This appears to be a well known-issue with EKS currently (see #75 (comment)), I am not sure how eksctl could possibly address this – wdyt?

@niksrc
Copy link
Author

niksrc commented Jul 13, 2018

Thanks for the quick reply. I'm wondering what eksctl does differently as i am able to create a cluster from the console every time in the same region.

@niksrc
Copy link
Author

niksrc commented Jul 13, 2018

See this issue coreos/coreos-kubernetes#442 where @mumosho suggests letting cloudformation decide which availabilty zone to choose for a user.
Here is the corresponding gist https://gist.github.com/mumoshu/d375a37a87d09ecdc0cc0d1be322e628

@errordeveloper
Copy link
Contributor

@niksrc thanks for your suggestion, I will definitely look into this. Have you been able to create a cluster with eksctl after all or still no dice?

@errordeveloper
Copy link
Contributor

@niksrc
Copy link
Author

niksrc commented Jul 17, 2018

@errordeveloper i was able to create the cluster by hard coding the index of the subnet in the template files and rebuilding eksctl. I don't think we can do anything to fix this unless aws provides an api for availability detection for a region.

Thanks for looking into this.

@niksrc niksrc closed this as completed Jul 17, 2018
@errordeveloper
Copy link
Contributor

@niksrc did you do something similar to what @jorihardman did in https://github.com/swipesense/eksctl/commit/ca185e446da624f0f65b27b57dd121d54adbe7c9? I think we should be able to incorporate something similar to that, but we might want to take it out of CloudFormation and also fix #101.

@errordeveloper
Copy link
Contributor

I'd rather keep this issue open for now.

@niksrc
Copy link
Author

niksrc commented Jul 17, 2018

@errordeveloper yes, i did the same thing, and please let me know if you need any help from me.

@errordeveloper
Copy link
Contributor

@niksrc we are working on a fix in #124, testing would be very helpful, we will let you know once it's ready to test (it's not yet).

@errordeveloper
Copy link
Contributor

@niksrc we have merged a fix for this in #124. I've tagged 0.1.0-beta.2 and binaries should be available shortly – please let us know if it works for you!

@errordeveloper
Copy link
Contributor

errordeveloper commented Jul 20, 2018

@niksrc turns out there is a bug in 0.1.0-beta.2 (#129) and we have just merged a fix into master (#130) – would you mind to help us testing it before we figure if we need to do more to really solve the problem?

@errordeveloper
Copy link
Contributor

After thinking about this for a while, it looks like we are better of exposing a flag for the time being, as it turns this is somehow different in every AWS account (at least that what seems to be the case from the docs.

You may receive an error that one of the Availability Zones in your request does not have sufficient capacity to create an Amazon EKS cluster. If this happens, the error output contains the Availability Zones that can support a new cluster. Retry creating your cluster with at least two subnets that are located in the supported Availability Zones for your account.

@errordeveloper
Copy link
Contributor

In the future we may try parsing the error message and retrying automatically, but it doesn't seem feasible right now as we would need to improve some of the internals and provide stack update functionality.

@niksrc
Copy link
Author

niksrc commented Jul 23, 2018

@errordeveloper This worked smoothly, Thanks.

@errordeveloper
Copy link
Contributor

errordeveloper commented Jul 23, 2018 via email

@niksrc
Copy link
Author

niksrc commented Jul 23, 2018

I used the latest master.
Tag 0.1.0-beta.3 c332060

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants