k8s AWS downtime during minor version upgrades

[pkoutsovasilis]

Hi, according to this link when Thingsboard is deployed with the micro-services architecture, which is what I am currently exploring with EKS on AWS and everything else AWS managed (RDS, MSK, Elasticache), there should be no downtime at least for minor version updates. However, based on the upgrade instructions found here there are some cases that tb-nodes need to be stopped and as a result experience some downtime. I looked through the upgrade instructions here but there is no mention when the tb-nodes should be stopped. Is the information of when to stop tb-nodes before an upgrade somewhere visible? Thx