Solved (was: Wrong/outdated grains on kubic instances added via kubicctl?)
Hi Thorsten, On 08.06.21 at 21:31 Thorsten Kukuk wrote:
This hasn't changed. My guess: kubicctl/kubicd is not designed to have nodes, which are workers and masters at the same time. But it looks like you did setup a cluster of master nodes and you did enable scheduling of pods on them with kubectl afterwards. So for kubicctl, there are no worker nodes in your cluster, only masters. That's a not supported scenario. Best is to use directly kubeadm in this case.
I found my notes from when I deployed the cluster, and I just initialized the cluster with three master nodes and untainted them. Kind Regards, Johannes -- Johannes Kastl Linux Consultant & Trainer Tel.: +49 (0) 151 2372 5802 Mail: kastl@b1-systems.de B1 Systems GmbH Osterfeldstraße 7 / 85088 Vohburg http://www.b1-systems.de GF: Ralph Dehner Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537
participants (1)
-
Johannes Kastl