tencent cloud

Tencent Smart Advisor-Chaotic Fault Generator

Product Introduction
Overview
Strengths
Scenarios
Purchase Guide
Purchase Instructions
Getting Started
Quick Start with the Console
Quick Start with API
Operation Guide
Template Library
Experiments
Fault Action
Guardrail Monitoring
Tag
Agent Management
Fault Action Library
Compute
Database
Network
Container
Big Data
Cloud Load Balancer
Message Queue
Direct Connect
Custom Actions
Cloud Streaming Services (CSS)
Permission Management Guide
Overview
Authorization Policy Syntax
Authorizable Resource Types
Service Authorization and Role Permissions
Sub-users and Authorization
API Documentation
History
Introduction
API Category
Making API Requests
Task APIs
Template Library APIs
Data Types
Error Codes
FAQs
Product Feature Issues
Action Execution Issues
Agent FAQ
Related Protocol
PRIVACY POLICY MODULE CHAOTIC FAULT GENERATOR
DATA PRIVACY AND SECURITY AGREEMENT MODULE CHAOTIC FAULT GENERATOR
Contact Us

Simulating Primary-secondary Switch in Redis

PDF
Focus Mode
Font Size
Last updated: 2024-09-26 15:47:38

Background

Redis cluster is an important component for storing hot data of businesses. To ensure business availability, Gossip protocol between nodes in the cluster is used to determine node status. Heartbeat timeout (cluster-node-timeout) is 15s by default. If the fault node is the primary node, Tencent Cloud Redis will apply a failover mechanism and select a new primary node from the secondary node.
Based on the above features, Tencent Smart Advisor-Chaotic Fault Generator provides a manual method to skip node fault stage and directly simulate fault actions for HA policy. You can simulate the impact on businesses in the short period when failover occurs in Redis cluster by manual fault method.

Experiment Implementation

Experiment Preparation

Prepare a multi-node cross-availability zone Redis instance.

Experiment Steps

Step 1: Create an Experiment

1. Log in to the Tencent Smart Advisor > Chaotic Fault Generator, go to the Experiment Management page, and click Create a New Experiment.
2. Click Skip and create a blank experiment .
3. Fill in basic information. For Experiment Resource Object, select TencentDB for Redis memory edition under Cloud Resource Type, and add an instance.

Step 2: Add Actions

1. Click Add Now to add a fault action. Select Redis primary-secondary switch in Redis for fault action.
2. In action parameters setting, flexibly select primary/replica switch mode based on a simulated disaster recovery scene:
Prefer switching within the same availability zone
Simulate the real HA policy scene of Tencent Cloud Redis when the primary node is faulty: the latest data node is uplifted to the primary node; when data is identical, priority is given to other nodes in the same availability zone.
Prefer switching within the same availability zone
When an entire availability zone is faulty, nodes in another availability zone will be raised for the primary scene.

Step 3: Execute Experiment Actions

Go to Experiment Details. In the Experiment Action Group, click Execute to start executing an experiment.

Check Results

Take cross-availability zone mode as an example, check whether the availability zone state has changed before and after fault injection.


Help and Support

Was this page helpful?

Help us improve! Rate your documentation experience in 5 mins.

Feedback