Chapter 60: Kubernetes Workload Controllers in Depth: ReplicaSet, Deployment, StatefulSet, DaemonSet, Jobs and CronJobs
Pod swata kahi karat nahi. Controller pod banavto, mojto aani punha banavto. Pratyek controller kind cluster var chalvun bagha aani delete karun self-healing anubhava.
What you will learn in this chapter
- Run the legacy ReplicationController and see why ReplicaSet replaced it
- Use ReplicaSet set-based selectors, ownership and adoption
- Read Deployment revisions, strategy, maxSurge and maxUnavailable
- Run MySQL as a StatefulSet with stable names and volumes
- Put one agent on every node with a DaemonSet
- Run batch work with Jobs and CronJobs
- Choose the right controller for any app
Concepts in this chapter
- 60.1ReplicationController: the first controller, and why it is legacy
- 60.2ReplicaSet: set-based selectors, ownership and adoption
- 60.3Deployment internals: revisions, strategy, maxSurge and maxUnavailable
- 60.4StatefulSet: stable names, headless Service, volumeClaimTemplates and a MySQL lab
- 60.5DaemonSet: one pod on every node, tolerations and node selectors
- 60.6Jobs: completions, parallelism, backoffLimit, indexed jobs and TTL
- 60.7CronJobs: schedule, concurrencyPolicy, history limits and time zones
- 60.8Choosing the right controller: one app, every controller, side by side
The chapter recap is at the end of the last concept page.