Hello, there!
This topic describes floating IP.
1. Why floating IP?
Now there is a scenario where you deploy a web application on a Linux machine that runs on tomcat with IP 115.239.100.120 on the Linux adapter.
This is roughly the following deployment relationship:
Users can access the application by entering an IP address, but sometimes the service goes down for some reason. So the developers figured out a way to deploy the same application on another Linux, so that one service went down, the other one went up, and the architecture looked like this:
If a user cannot access 100.120, they can access 100.121, but this introduces the problem that users need to remember two IPs, which is inconvenient!
So we need floating IP.
2. What is the floating IP?
Above, we set up two identical services, but the user needs to remember two IP, which is too troublesome, so we can adopt another strategy.
We are adding an IP to one of the machines, allowing users to access this IP, a network card can add more than one
For example, we add a 100.80 IP to the 100.120 host, and the user accesses the 115.239.100.80 IP. If the service of this machine fails, we transfer this IP to the 100.121 machine, so that for the user, they always access only one IP.
In the above scenario: 115.239.100.80 this IP is floating IP, which will be transferred to another host that can provide the same service when the host service goes down.
As for the users, they always feel that the service is normal.
Translated from: https://blog.csdn.net/readiay/article/details/53538085.
