You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I wanted to create a cluster with no nodegroups and the aws-efs-csi-driver addon.
I want to create the cluster without nodes because I need to create ENIConfigs for custom networking after the cluster is created, but before any nodes join the cluster.
What happened?
The cluster creates; however, eksctl exits with an error because the aws-efs-csi-driver addon is degraded since there are no nodes and it can't run the minimum number of replicas. Once nodes are added, the addon will become active, but I need eksctl to not finish with an error code because of the build pipeline will fail.
How to reproduce it?
Build a cluster with a cluster-config.yaml with no nodegroups, but with efs-csi-driver addon enabled.
cluster-config.yaml:
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig
metadata:
name: test-cluster
region: us-east-1
version: "1.24"
tags:
TAG1: tag 1 value
TAG2: tag 2 value
vpc:
id: vpc-XXXXXXXXXXX
securityGroup: sg-XXXXXXXXXXX
clusterEndpoints:
publicAccess: false
privateAccess: true
subnets:
private:
us-east-1c:
id: subnet-XXXXXXXXXXXXXXX
us-east-1d:
id: subnet-XXXXXXXXXXXXXXX
us-east-1a:
id: subnet-XXXXXXXXXXXXXXX
addons:
- name: vpc-cni
configurationValues: "{\"env\":{\"AWS_VPC_K8S_CNI_CUSTOM_NETWORK_CFG\":\"true\", \"ENI_CONFIG_LABEL_DEF\":\"topology.kubernetes.io/zone\"}}"
version: latest
attachPolicyARNs:
- arn:aws:iam::aws:policy/AmazonEKS_CNI_Policy
- name: coredns
version: latest # auto discovers the latest available
- name: kube-proxy
version: latest
- name: aws-ebs-csi-driver
wellKnownPolicies: # add IAM and service account
ebsCSIController: true
- name: aws-efs-csi-driver
wellKnownPolicies: # add IAM and service account
efsCSIController: true
iam:
withOIDC: true
Logs
End of the output from eksctl create cluster -f cluster-config.yaml:
2023-08-11 18:08:49 [ℹ] waiting for CloudFormation stack "eksctl-test-cluster-addon-aws-efs-csi-driver"
2023-08-11 18:09:19 [ℹ] waiting for CloudFormation stack "eksctl-test-cluster-addon-aws-efs-csi-driver"
2023-08-11 18:10:11 [ℹ] waiting for CloudFormation stack "eksctl-test-cluster-addon-aws-efs-csi-driver"
2023-08-11 18:10:11 [ℹ] creating addon
2023-08-11 18:10:22 [!] 1 error(s) occurred while creating addons
2023-08-11 18:10:22 [✖] addon status transitioned to "DEGRADED"
Error: failed to create addons
Anything else we need to know?
This looks to be exactly the same issue as #5891 but for the EFS csi driver addon instead of the EBS csi driver addon. 5891 was fixed in PR #5929
Versions
$ eksctl info
$ eksctl info
eksctl version: 0.150.0-dev
kubectl version: v1.24.13-eks-0a21954
OS: linux
The text was updated successfully, but these errors were encountered:
Hello kryczech 👋 Thank you for opening an issue in eksctl project. The team will review the issue and aim to respond within 1-5 business days. Meanwhile, please read about the Contribution and Code of Conduct guidelines here. You can find out more information about eksctl on our website
kryczech
changed the title
[Bug]
[Bug] cluster build fails if no nodegroups and EFS csi addon required
Aug 14, 2023
My apologies for the formatting of the cluster config.
TiberiuGC
changed the title
[Bug] cluster build fails if no nodegroups and EFS csi addon required
[Bug] Cluster creation fails if no nodegroups and EFS csi addon required
Aug 16, 2023
What were you trying to accomplish?
I wanted to create a cluster with no nodegroups and the aws-efs-csi-driver addon.
I want to create the cluster without nodes because I need to create ENIConfigs for custom networking after the cluster is created, but before any nodes join the cluster.
What happened?
The cluster creates; however, eksctl exits with an error because the aws-efs-csi-driver addon is degraded since there are no nodes and it can't run the minimum number of replicas. Once nodes are added, the addon will become active, but I need eksctl to not finish with an error code because of the build pipeline will fail.
How to reproduce it?
Build a cluster with a cluster-config.yaml with no nodegroups, but with efs-csi-driver addon enabled.
Logs
Anything else we need to know?
This looks to be exactly the same issue as #5891 but for the EFS csi driver addon instead of the EBS csi driver addon. 5891 was fixed in PR #5929
Versions
$ eksctl info
eksctl version: 0.150.0-dev
kubectl version: v1.24.13-eks-0a21954
OS: linux
The text was updated successfully, but these errors were encountered: