CoreOS-GettingStarted

Sample Application and Module demos for Pluralsight course on "Getting Started with CoreOS"

The base configuration of each module leverages the great work at https://github.com/coreos/coreos-vagrant for running CoreOS locally via Vagrant.

Common Issues

Cluster doesn't start correct / is reported as unhealthy

Depending on how you have configured your local environment, you may run into an issue where the cluster doesn't start correctly.

If etcdctl cluster-health returns the following error:

cluster may be unhealthy: failed to list members
Error:  client: etcd cluster is unavailable or misconfigured
error #0: dial tcp 127.0.0.1:2379: getsockopt: connection refused
error #1: dial tcp 127.0.0.1:4001: getsockopt: connection refused

You can resolve this problem by running vagrant reload --provision to restart the VMs and force loading the latest configuration changes in your user-data file.

systemctl appears hung when restarting a unit

When running sudo systemctl restart hello.service, it can appear that the command has hung. However, this command can take several minutes to complete on some virtual machines (depending on host specs, CoreOS version, etc.).