Error with the example

[Brownie2002]

Hello,

I try to use your module. But for now, I am not able. I tried the example (with adding my credentials) and I get the same error.

Error: Error creating Auto Scaling Group: InvalidQueryParameter: Incompatible launch template: Each network interface requires a unique device index.
 status code: 400, request id: ea9c7cfb-fe66-4896-9e67-cd35d45e20d2

with module.ssh-bastion-service.aws_autoscaling_group.bastion-service,
on .terraform\modules\ssh-bastion-service\main.tf line 74, in resource "aws_autoscaling_group" "bastion-service":
74: resource "aws_autoscaling_group" "bastion-service" {

I am very noob to terraform, so I try to reproduce the example without success. So maybe I done something wrong.

Regards,

Brownie

[joshuamkite]

Hi

Sorry to hear of this- are you able to provide the code you are using to call this module?

Joshua

[Brownie2002]

Hello, easy, it is the example : https://github.com/joshuamkite/terraform-aws-ssh-bastion-service/tree/master/examples/full-with-public-ip

So no modification ... and thank you for your attention.

[joshuamkite]

I am afraid that I don't see this error when running the example as provided with no changes- deploys without issue for me. Are you adding any optional values such as additional security groups?

[Brownie2002]

Hello, unfortunately no ... I will try again later. So, we could close the issue since you cannot reproduce it :(

[joshuamkite]

I don't know what to say- I don't really want to dismiss your difficulty as 'works on my machine' but without additional detail or being able to reproduce it I don't know how I could proceed

[karmajunkie]

I'm also seeing this issue in my own use. It appears to be related to hashicorp/terraform-provider-aws#23766. I've installed v4.5 of the hashicorp/aws provider (per the linked issue) but to fix the issue you have to tear down the launch template created by v4.6 and let it recreate.

[joshuamkite]

Thanks for this @karmajunkie ! IIRC this is a manifestation of an upstream bug with version 4.60 specifically of the AWS provider. Although it is fixed with 4.70 and above these will still break if you have a pre-existing launch template created with 4.6?

Assuming above is correct, the best I could do here is advise a workaround procedure for 4.6 but this would only be relevant for people using or who had used the 4.6 provider, which was only 'current' from March 18, 2022 to March 24, 2022, i.e. 5 days. Whilst I am happy to be challenged on this, I don't think that this is appropriate or worthwhile to do so within this project and propose to close as 'upstream bug now fixed'

[joshuamkite]

Closing as 'upstream bug wontfix (now fixed upstream)'