Kubereboot/Kured: Kubernetes Reboot Daemon

edude03

Insert the "No god no" meme here - you really shouldn't be updating nodes in place and thus shouldn't be restarting nodes. I'm aware bare metal exists and it's not always practical to just provision more servers, yet I think for most workloads you're not getting the benefit of Kubernetes if you have say 3 servers and lose 1/3 of your capacity to do software updates.

AntiUSAbah

I like it. K8s should be more opiniated about this. Whats also missing is rebalancing of pods. Rescheduler

adamtulinius

Almost 3000 lines of code for automating draining nodes and rebooting it. And it requires that another component has already queued up an update that requires a reboot. Looking at the issues, people try to shoehorn a thousand unique behaviours into a general purpose tool, just to avoid a bit of old school sysadmin-ing. There's a guy wanting to change TZ of the running cluster, and want "Kured" to support that use case so it's only updated during night - in an ever changing TZ.

captn3m0

What’s the usecase where you are okay cordoning a node but not okay with just terminating it and starting a new one? Physical nodes where you have to reclaim them and don’t run any virtualisation ?