kjuulh/terraform-hcloud-kube-hetzner
1
0
Fork 0
Code Issues 1 Pull Requests Projects Releases Wiki Activity

tweaked readme

Browse Source
This commit is contained in:
Karim Naufal 2022-02-11 13:28:27 +01:00
parent 4497a7fd7f
commit f1f6e8448c
3 changed files with 18 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
README.md
View File

@ -90,17 +90,15 @@ When the cluster is up and running, you can do whatever you wish with it! 🎉
### Scaling nodes ### Scaling nodes
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`. ⚠️ Once you start with Terraform, it's best not to change the state manually in Hetzner, otherwise when you try to scale up or down, Terraform will complain that things changed outside of it and will not be able to do it. _In the future, we will provide a tool to create bare nodes, either agents or control planes, to be joined manually._
**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/).** To scale the number of nodes up or down, just make sure to properly `kubectl drain` the nodes in question first if scaling down. Then just edit these variables in `terraform.tfvars` and re-apply terraform with `terraform apply -auto-approve`.
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: For instance:
```tfvars ```tfvars
servers_num = 3 servers_num = 4
agents_num = 2 agents_num = 3
``` ```
### Useful commands ### Useful commands
@ -124,12 +122,18 @@ hcloud network describe k3s
ssh root@xxx.xxx.xxx.xxx -i ~/.ssh/id_ed25519 -o StrictHostKeyChecking=no ssh root@xxx.xxx.xxx.xxx -i ~/.ssh/id_ed25519 -o StrictHostKeyChecking=no
``` ```
## By default HA
By default, we have 3 control planes configured and 2 agents, with automatic upgrades and reboots of the nodes.
**But if you want to remain 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, it's important to turn off automatic upgrades (see below) and reboots for the control-plane nodes (2 or less), and do the maintenance yourself.
## Automatic upgrade ## Automatic upgrade
By default, MicroOS and its embedded k3s instance get upgraded automatically on each node, and reboot safely via [Kured](https://github.com/weaveworks/kured) installed in the cluster. By default, MicroOS and its embedded k3s instance get upgraded automatically on each node, and reboot safely via [Kured](https://github.com/weaveworks/kured) installed in the cluster.
_You can also choose to automatically kustomize the Hetzner CCM and CSI to set their container images to "latest" with an imagePullPolicy of "Always". That means that when the nodes upgrade, these container images will be automatically upgraded too. For more info on this, see [terraform.tfvars.example](terraform.tfvars.example)._
_About [Kured](https://github.com/weaveworks/kured), it does not have a latest tag present for its image, but it's pretty compatible, so you can just manually update the tag from once every year for instance._ _About [Kured](https://github.com/weaveworks/kured), it does not have a latest tag present for its image, but it's pretty compatible, so you can just manually update the tag from once every year for instance._
_Last but not least, if you wish to turn off automatic upgrade on a specific node, you need to ssh into it and issue the following command:_ _Last but not least, if you wish to turn off automatic upgrade on a specific node, you need to ssh into it and issue the following command:_

9
terraform.tfvars.example
View File

@ -24,11 +24,10 @@ agents_num = 2
# hetzner_csi_version = "" # hetzner_csi_version = ""
# If you want to kustomize the Hetzner CCM and CSI containers with the "latest" tags and imagePullPolicy Always, # If you want to kustomize the Hetzner CCM and CSI containers with the "latest" tags and imagePullPolicy Always,
# to have them automatically update when the node themselve get updated via the rancher system upgrade controller, the default is "false". # to have them automatically update when the node themselve get updated via the rancher system upgrade controller, the default is "true".
# If you choose to keep the default of "false", you can always use ArgoCD to monitor the CSI and CCM manifest for new releases, # Instead, if you choose false, it will give you the versions that Hetzner uses at the time of install.
# that is probably the more "vanilla" option to keep these components always updated. # hetzner_ccm_containers_latest = false
# hetzner_ccm_containers_latest = true # hetzner_csi_containers_latest = false
# hetzner_csi_containers_latest = true
# If you want to use letsencrypt with tls Challenge, the email address is used to send you certificates expiration notices # If you want to use letsencrypt with tls Challenge, the email address is used to send you certificates expiration notices
# traefik_acme_tls = true # traefik_acme_tls = true

4
variables.tf
View File

@ -69,13 +69,13 @@ variable "hetzner_csi_version" {
variable "hetzner_ccm_containers_latest" { variable "hetzner_ccm_containers_latest" {
type = bool type = bool
default = false default = true
description = "Whether to kustomize the Hetzner CCM manifest with the latest or canary tags for containers and imagePullPolicy of Always" description = "Whether to kustomize the Hetzner CCM manifest with the latest or canary tags for containers and imagePullPolicy of Always"
} }
variable "hetzner_csi_containers_latest" { variable "hetzner_csi_containers_latest" {
type = bool type = bool
default = false default = true
description = "Whether to kustomize the Hetzner CSI manifest with the latest or canary tags for containers and imagePullPolicy of Always" description = "Whether to kustomize the Hetzner CSI manifest with the latest or canary tags for containers and imagePullPolicy of Always"
} }