Load Balancing with NGINX

preview_player
Показать описание
Load balancing is one of the most widely deployed use case for NGINX Instance Manager – it is easy to configure and provides multiple load balancing algorithms to choose from to ensure smooth flow of network traffic between clients and upstream servers.

In this basic how-to video, we cover:
◆ High level understanding of load balancing works
◆ Upstream directive
◆ Available load balancing methods

NGINX Load Balancing GitHub

How to Get Started With NGINX

Introducing NGINX Instance Manager

Chapters:
0:00 - Load Balancing with NGINX Session Agenda
0:43 - Multitude of Use Cases
6:18 - Load Balancing Methods
12:01 - Enabling Session Persistence with NGINX
13:53 - Things to Note When Load Balancing with NGINX
14:50 - Live Demo
Рекомендации по теме
Комментарии
Автор

Im using Nginx for loadbalancing my 4 server setup for a custom application. The application has a media share option that if disabled the forward option does not allow people from other networks to see it. It uses the IP of the user for this. The problem is if i use load balancing all the hits to my web server are coming from the load balancers IP. So even if a user from external network tries to access the media after disabling that option, he is able to access the media because server is getting the IP of Nginx setup. Is there any feature to overcome this issue?

rijosimon
Автор

thanks for this training, it was very good

guillemurto
Автор

Workload I need load-balanced is VOIP streamed via UDP. Can NGINX Open Source forward UDP or it needs to be Plus?

NewHOD
Автор

Is there any way to make the load balancer not be the single point of failure ?? If it goes down, all the healthy machines and services become useless right ??

blissweb
Автор

please note that sticky session is not free, it come with nginx plus

redosem
Автор

Thank you for the video. In studying the Nginx Cook-Book and trying to further understand the "hash directive" from this video, it's not 100% clear how it works. Do I create a hash of the specific data I want to match and if it's "anywhere" in the request, then it goes to a specific server? Or do I have to create a hash of the entire "request" I'm looking to match? Any help in this area is greatly appreciated. 🙏

jsaenzMusic
Автор

all ports 9001, 9002 and 9003 pointing to different project path, my question, can we point to same project folder path? does pointing to same folder path cause performance degradation?

sabinshrestha
Автор

This LB isn't DSR (Direct server return) or is it? 🤨
And can nginx do such LB?

Hellmiauz
Автор

These three web servers still share one physical server right?

kobebyrant
Автор

is there a way to install a web GUI to administer the load balancer?

Eleazar.Chavira
Автор

I cant see, increase font size on no.20

prezlamen
Автор

Love the explanation, hate the indentations.

jamestrowbridge
Автор

Epic explanation.. made it very easy.
Thanks!
Could you add the links you mentioned in the video?

flexairz
Автор

Oww so the correct way to call it is engine X. I was mistaken my whole life.

zerocoll