ForgeRock/forgeops

Pulumi for EKS failing with extIngresstg80arn and extIngresstg443arn does not exist

Closed this issue · 4 comments

After successfully running the AWS infra pulumi stack, the eks stack is failing with 2 missing outputs from the infra stack:

Error: Required output 'extIngresstg443arn' does not exist on stack 'aws-infra'.

and
Error: Required output 'extIngresstg80arn' does not exist on stack 'aws-infra'.

Those are not mentioned as outputs though, so, I am wondering if I should create them manually or what else should it be done to overcome these expected parameters for Ingress:

Outputs:
bastionEnable : false
bastionPublicIp : "undefined"
bastionSgId : "undefined"
clusterAdministratorRoleID: "clusterAdministratorRole-d9858ca"
vpc : {
id : "vpc-0186e1b6db722b778"
urn: "urn:pulumi:aws-infra::aws-infra::awsx:x:ec2:Vpc::eks-cdm"
vpc: {
arn : "arn:aws:ec2:ap-southeast-2::vpc/vpc-0186e1b6db722b778"
assignGeneratedIpv6CidrBlock: false
cidrBlock : "192.168.0.0/16"
defaultNetworkAclId : "acl-03d24ca5313d40ab7"
defaultRouteTableId : "rtb-03bae1150aef9e9e0"
defaultSecurityGroupId : "sg-0ad3be7ea6331af96"
dhcpOptionsId : "dopt-6449ae01"
enableClassiclink : false
enableClassiclinkDnsSupport : false
enableDnsHostnames : true
enableDnsSupport : true
id : "vpc-0186e1b6db722b778"
instanceTenancy : "default"
mainRouteTableId : "rtb-03bae1150aef9e9e0"
ownerId : "
"
tags : {
CreatedBy: "process.env.USER"
Name : "eks-cdm"
}
urn : "urn:pulumi:aws-infra::aws-infra::awsx:x:ec2:Vpc$aws:ec2/vpc:Vpc::eks-cdm"
}
}
vpcCIDR : "192.168.0.0/16"
vpcIsolatedSubnetsIds : [
[0]: "subnet-0631f6ac56f9e27e2"
[1]: "subnet-0deeaa49e2601d4df"
[2]: "subnet-03949218cdc5f69ac"
]
vpcPrivateSubnetsIds : [
[0]: "subnet-0c4f58139b6c8c745"
[1]: "subnet-09ca8cd7c3a89b87e"
[2]: "subnet-0237899b3c6d31d4f"
]
vpcPublicSubnetsIds : [
[0]: "subnet-0c7eafc521f68c8b2"
[1]: "subnet-0fc05d7d9a0b24992"
[2]: "subnet-05da433c7c8066feb"
]
vpcid : "vpc-0186e1b6db722b778"

Resources:
+ 57 created

Duration: 2m9s

Permalink: file:///~/.pulumi/stacks/aws-infra.json

Thanks.

Jose

Found HighAvailability config and piece of code that tells the assumption of loadbalancer service type will be used. Maybe document that default behaviour in the pulumi repo will help ?

index.ts:if (highAvailability) //If highAvailability is False, it is assumed that the user will use type loadbalancer.

pulumi config set aws-infra:highAvailability true

Thanks.

Jose

Hi, those 2 values are exported in the code and which results in them being available in the EKS stacks.

export let extIngresstg80arn : any = undefined
export let extIngresstg443arn : any = undefined

What changes did you make to the eks stack?

Hi,

I've pulled the code again and ran a pulumi refresh to make sure that extIngresstg80arn and extIngresstg443arn were being created. Then ran pulumi update and noted that the load-balancer was created.

Thanks.

Jose

Ok great. Glad its sorted