From e78ebe570ae38599210b94b2e45b94b920ea98c8 Mon Sep 17 00:00:00 2001 From: Karim Naufal Date: Thu, 10 Feb 2022 03:50:54 +0100 Subject: [PATCH] tweaked readme --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index d03e1e4..4fe4508 100644 --- a/README.md +++ b/README.md @@ -93,9 +93,9 @@ When the cluster is up and running, you can do whatever you wish with it! 🎉 You can scale the number of nodes up and down without any issues. If you are going to scale down, just make sure to properly `kubectl drain` the nodes in question first. Then just edit these variables in `terraform.tfvars` and re-apply terraform with `terraform apply -auto-approve`. -**If you want to remain HA, it's important to keep a number of control planes of at least 3, see [Rancher's doc on HA](https://rancher.com/docs/k3s/latest/en/installation/ha-embedded/).** +**If you want to be HA, it's important to keep a number of control planes nodes of at least 3 (2 to maintain quorum when 1 goes down for automated upgrades and reboot for instance), see [Rancher's doc on HA](https://rancher.com/docs/k3s/latest/en/installation/ha-embedded/).** -Otherwise, you may want to turn off automated updates and reboots of the control-plane nodes (2 or less), and do these maintenance manually. +Otherwise, it's important to turn off automatic updates and reboots for the control-plane nodes (2 or less), and do the maintenance yourself. For instance: