Roman Iuvshyn

@brandond sure, I am using `aws eni cni`. my k3s config file: ``` egress-selector-mode: disabled node-name: $${LOCAL_HOSTNAME} flannel-backend: none disable-cloud-controller: true node-ip: $${LOCAL_IPV4} node-external-ip: $${LOCAL_IPV4} kube-apiserver-arg: - "service-account-key-file=/etc/rancher/service-account.pub" - "service-account-issuer=${service_account_issuer}"...

I've managed to reproduce this issue with 2.8.0 version

👋🏽 any update on this? Is it going to be fixed in 2.8 or in 2.9?

@okgolove I've discovered this during testing upgrade in test env and once detected decided to revert. So I don't know much about actual impact on a managed cluster.

@okgolove I was trying to upgrade from 2.7.5 to 2.8.0 but I think this regression was introduced in 2.7.6

@okgolove That is exactly what I did, and put that into a CronJob but still looking forward for proper fix :/

@jdbaudean yeah, that is why I put that script into a cron job which is running every 15minutes to make sure that machine with state`NodeNotFound` is removed as soon as...

any update on this? 🙏🏽 This is very important feature required for running stateful workloads like kafka clusters where graceful rotation of nodes is critical and also quite slow.

>Interesting, I hadn't expected nodes taking hours to drain. We can look into adding the heartbeat to cover the long draining case. Thanks for reporting! @bwagner5 This problem affects not...

Thank you for working on this! @jannfis is this supposed to be working with applicationSets?