Fully fledged (HA) Kubernetes Cluster using official kubeadm, ansible and helm. Tested on RHEL/CentOS/Ubuntu with support of http_proxy, dashboard installed, ingress controller, heapster - using official helm charts
On slower machines (eg development one node ones), delay of 3 seconds is not enough, causing Error: UPGRADE FAILED: another operation (install/upgrade/rollback) is in progress because it's retried too early.
this allows to override CHARTS_DELAY in ansible vars, with default being the original 3 seconds.
On slower machines (eg development one node ones), delay of 3 seconds is not enough, causing
Error: UPGRADE FAILED: another operation (install/upgrade/rollback) is in progress
because it's retried too early.this allows to override CHARTS_DELAY in ansible vars, with default being the original 3 seconds.