Can we use NGINX as load balancer?

What is HAProxy vs NGINX cache?

If you're looking for a solution to improve your web applications performance, then you are at the right place.

I'm a big fan of HAProxy for its performance and capabilities. It's used on servers all around the world including my own personal website and the NGINX is a powerful option if you are looking for static content.

Here's a great read on what HAProxy is and it's different pros and cons. HAProxy vs NGINX cache: In this article, I'll compare HAProxy to NGINX for caching static content. As I mentioned in the introduction, both of these are popular options but they have different benefits and uses. But before we move forward, let's have a look at what we are going to cover in this post: Why you need to use NGINX in your web applications. Which caching option is best suited for your web applications. Why NGINX cache can be a game changer for your web applications. What are the pros and cons of using NGINX for caching static content? Read the full post to get an idea about HAProxy vs NGINX. What is NGINX? If you don't know what NGINX is, here's a quick intro. NGINX is a highly scalable web server which provides HTTP request caching support and more. Its popularity is attributed to its flexibility and it's very simple and easy to configure. It comes with lots of pre-built modules like proxying, load balancing, proxying, caching and so on. The fact that it's easy to configure makes it the go-to choice for most developers. Some of the biggest players in the web development community have adopted NGINX into their application's stack. NGINX vs HAProxy. One way to tell the difference between these two is by its capabilities. Both of them provide load balancing but NGINX can load balance requests across multiple servers whereas HAProxy is single server. For caching, HAProxy allows you to set cookies to remember the server IP address. I believe most of the time developers use HAProxy because it is easier to configure and install.

Can we use NGINX as load balancer?

How can we use NGINX as a load balancer?

I know that there is a new version of NGINX that supports this, but I would like to use the older version of NGINX. Is it possible? If yes, how can we do that? Thanks! I think what you want is "sticky sessions" in NGINX. You can set "sticky sessions" for both HTTP and HTTPS traffic by setting the "stickysession" directive in the http and https blocks of your server configuration.

For example, if you have a web application that needs to be served by more than one web server, you can use NGINX as a load balancer for the sessions.
It's also possible to use the HttpStickup module in the nginx-lua-module, which allows you to easily perform sticky sessions using Lua code.

Is HAProxy a reverse proxy?

My understanding is that a reverse proxy, like Squid, sits in between the client and the web server (and has knowledge of both to route traffic from the client, over an encrypted SSL connection, to the web server.) This seems to suggest that HAProxy, which is also a HTTP proxy sits in-between the client and the backend server which implies that HAProxy cannot (or does not claim to?) know the backend's IP address.

Then there's the concept of being a gateway and/or having access to load balancing features and content processing features but none of these seem to indicate that HAProxy is a reverse proxy. I'm fairly new to reverse proxies and would love to have someone explain to me what the terms "reverse proxy" and "gateway" mean with regards to the role HAProxy plays in the server's I/O hierarchy.

1 Answer.
You're spot on. Both HAProxy and Squid are very popular for traffic control, since they are highly configurable and easy to use, allowing them to be used for many tasks. You can use them as "gateways", but you can also setup other solutions as well, even if all you want to do is to perform simple caching. It really depends on your problem domain.

In your case, the main difference is that HAProxy gives you a lot more features. And I believe it has some load balancing abilities. With squid you usually just get the standard functionality. On the other hand, since it is web based and uses HTTP in most of its logic, it is not so easy to use in cases where you might have to connect via SSH or use SMTP directly.

Now, since you asked the question about your goal to understand how HAProxy works. The truth is that HAProxy is a reverse proxy. The definition above can be found here. Also, here's a wikipedia article that explains it better. I recommend you look at this link, since it was the one I first looked up for a quick answer.

If you would like to have a more thorough discussion on the meaning of proxies, or learn about how they work in general, you can go here. There's a nice diagram that explains proxies as a black box (ie where data comes from).