Closed drcannoli closed 2 years ago
Nvm got it working. Seems the first node I needed to switch to type: init
under machine
. Must've missed a step in the docs, was under the impression the type init
no longer needed to be used.
@drcannoli When it was controlplane
instead of init
did you run talosctl bootstrap
?
No @andrewrynhard same commands. Just changed the following
version: v1alpha1 # Indicates the schema used to decode the contents.
debug: true # Enable verbose logging to the console.
persist: true # Indicates whether to pull the machine config upon every boot.
# Provides machine specific configuration options.
machine:
type: controlplane # Defines the role of the machine within the cluster.
TO
version: v1alpha1 # Indicates the schema used to decode the contents.
debug: true # Enable verbose logging to the console.
persist: true # Indicates whether to pull the machine config upon every boot.
# Provides machine specific configuration options.
machine:
type: init # Defines the role of the machine within the cluster. <---- this line
Exact same commands as before were issued
talosctl gen config talos-cluster https://192.168.0.107:6443 --output-dir _out
talosctl apply-config --insecure --nodes 192.168.0.107 --file _out/controlplane.yaml
Afterwards I joined 2 more nodes successfully changing back to type: controlplane
Ah I misread your comment. No I never ran talosctl bootstrap
Bootstrapping is required when not using init type
Yup, I found where I missed the steps in the docs. Thanks!
Bug Report
Description
Fresh install on new Proxmox instance seems to fail to start due to the k8s controller failing to start. Used instructions from here https://www.talos.dev/docs/v0.12/virtualized-platforms/proxmox/. Tried with version
0.11.5
as well as changing the kubernetes image versions.Commands to get cluster going were:
Only change made to
controlplane.yaml
wasdebug: true
Logs
Full Logs: https://pastebin.com/kzn6G67P The below is the end snippet that keeps repeating as the controller keeps restarting
Environment