Many companies have adopted Kubernetes or have a migration plan in place, making it clear that the platform is here to stay. While it provides many benefits to users, to take advantage of them, you need to really learn Kubernetes and how it works in production. Typically, the most difficult aspect of Kubernetes learned through experience real-world problems.
If you have to start somewhere, learn the concepts of cluster Kubernetes. Large cloud providers (Google, AWS, Azure, etc.) all provide Kubernetes cluster as a service. This is a good place to start. With just a few clicks on their cloud console, you can create a cluster. You can use file-based backup for Kubernetes for backup and restore the property.
In addition, cloud providers manage the cluster; perform routine tasks such as rotation of certificates and patching version. However, if you must arrange your own Kubernetes cluster, you will need to manage everything themselves, making the situation a bit more complex.
Described below are several scenarios that may come in time to manage the cluster on your own, as well as some resources that can help you become more adept at this process.
Image Source: Google
Bootstrap Kubernetes from Scratch
Although you can easily use tools such as Kubernetes bootstrap kubeadm to establish a cluster, it is important to know how the clusters bootstrapped from the beginning. Thus, you will be able to overcome the production problems that may arise in the future. The following two sources explained this well:
- Kubernetes Hard Way on Bare Metal, Oahcran
- Kubernetes Hard Way, Kelsey Hightower
Backup and Restore
Each system must have a backup and restore plan in place, and Kubernetes no exception. The resources listed below explain how to imagine and set up an effective plan. Note that the persistence layer in Kubernetes key-value store called etcd.
- Kubernetes proposal – how and why (with examples for TCD), Elastisys
- TCD cluster operations for Kubernetes, Kubernetes Documentation
- Backup TCD, CoreOS