Kubernetes - High Availabilty and Lifecycle management for Stateful Applications

Prabhat Sharma

@hiprabhat
@prabhatsharma

Basic principles

Redundancy is the key

If you have two, then you have one.

If you have one, then you have none.

How to achieve redundancy

Application specific mechanisms
- Replication
- e.g. MySQL, MongoDB, Postgres
- Implemented via helm charts and operators
Storage Layer
- RAID 1
- DRBD
- Portworx

Lifecycle management

HA is not enough. We need to manage day 1 & day 2

Day 1
1. Installation
2. High Availabilty
Day 2
1. Upgrades
2. Backup & restore
3. Auto recovery
4. Automatic/On-demand scalability
5. Configuration management
6. Deep insights

Helm

Helm helps you manage Kubernetes applications — Helm Charts help you define, install, and upgrade even the most complex Kubernetes application.

Operator

“A Kubernetes Operator is an abstraction for deploying non-trivial applications on Kubernetes. It wraps the logic for deploying and operating an application using Kubernetes constructs.” –Richard Laub, staff cloud engineer at Nebulaworks

Solutions for Application aware deployment

Solution	Pros	Cons
Manual Deployment	None	Time taking. Difficult
Helm charts	Easy	Day 2 ops missing
Operators	Day 1 & Day 2 covered	Still maturing

Elasticsearch operator

					
apiVersion: elasticsearch.k8s.elastic.co/v1alpha1
kind: Elasticsearch
metadata:
  name: quickstart
spec:
  version: 7.1.0
  nodes:
  - nodeCount: 1
  config:
    node.master: true
    node.data: true
    node.ingest: true

Elasticsearch Operator Lab