Dogfood Cluster Stability Spike
-
New ELB was created for the dogfood cluster kube controlplane and new nodes are not able to automatically join as a result. I believe if we run terragrunt applyfrom the customers/bigbang terraform it will regenerate the updated user-data for the nodes with the correct ELB. -
Scale down ci-optimized nodes (ASG) for the dogfood cluster to be desired+minimum of 2 instead of 3. -
Update the IAM credentials that our release stages in pipelines utilize, currently the credentials are set as variables within the project CI/CD settings, but maybe an IAM role we define/create via IaC would be better. -
ECK cluster in dogfood RKE2 cluster is hitting close to 85% storage usage. May need to expand the PV(C)s. If EBS backed, can use this guide
Edited by Jason Krause