From e83365bbe09adef5087ee518ce4194ae9ea5c236 Mon Sep 17 00:00:00 2001 From: kfilippopolitis Date: Wed, 15 Nov 2023 11:19:41 +0200 Subject: [PATCH 1/2] Update README.md to include microk8s recoverability feutures. --- kubernetes/README.md | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) diff --git a/kubernetes/README.md b/kubernetes/README.md index 7b8e4a6e..6dba565c 100644 --- a/kubernetes/README.md +++ b/kubernetes/README.md @@ -1,4 +1,4 @@ -# Kubernetes Deployment +**# Kubernetes Deployment ## Requirements ### Hardware @@ -174,3 +174,29 @@ For a more in-depth guide on deploying Exareme2, please refer to the documentati ``` microk8s helm3 install mip -f /opt/mip-deployment/kubernetes/ /opt/mip-deployment/kubernetes ``` + +# MicroK8s Automatic Recoverability + +## Overview +MicroK8s is designed for simplicity and resilience. One of its key features is the automatic recoverability of both federated clusters and individual local nodes. + +## Automatic Recoverability in Federation +In a federated cluster setup, MicroK8s ensures high availability and fault tolerance. If the master node in a federation faces downtime or operational issues, MicroK8s is designed to automatically recover its state and functionality. + +### Key Points: +- **Self-healing Mechanism**: MicroK8s employs a self-healing mechanism that triggers upon detecting issues with the master node. +- **State Restoration**: Automatically restores the master node to its last known healthy state without manual intervention. + +## Local Node Recoverability +For individual local nodes, MicroK8s offers a robust recovery process. This process is vital in scenarios where local nodes experience disruptions. + +### Key Points: +- **Node Health Monitoring**: Continuous monitoring of node health to quickly identify any disruptions. +- **Automatic Restoration**: On reboot or reconnection, the local node automatically synchronizes and restores its state to align with the federation's current status. + +## Recovery Time Frame +The recovery process in MicroK8s, whether for a federation or a local node, typically completes within a brief period. + +### Expected Timeline: +- **Minimum Recovery Time**: Approximately 1 minute. +- **Maximum Recovery Time**: Up to 5 minutes, depending on the complexity and scale of the cluster. From c08ef08c29696dda61c78778cbe0e68cfb1b9649 Mon Sep 17 00:00:00 2001 From: kfilippopolitis Date: Wed, 15 Nov 2023 11:27:53 +0200 Subject: [PATCH 2/2] Added the recovery on the main README.md too. --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index ff295546..d2725177 100644 --- a/README.md +++ b/README.md @@ -6,6 +6,7 @@ 1. [Deployment Components](kubernetes/README.md#components) 1. [Setting up the data](kubernetes/README.md#taking-care-of-the-medical-data) 1. [Deploying the stack](kubernetes/README.md#configuration) +1. [Stack recovery](kubernetes/README.md#microk8s-automatic-recoverability) 1. [Configuring new users](doc/keycloak/UsersConfiguration.md) 1. [Authorization for new user](doc/keycloak/UserAuthorizations.md) 1. [Backup and Recovery](doc/BackupAndRecovery.md)