Recorded a few pits encountered during the deployment of the eureka cluster, especially the last pit. After searching on the Internet for a long time, no results were found, and I finally found myself negligent. So make a special note and provide experience to other friends who have such problems.

In development, a single eureka registration center was used, and the configuration file is as follows:

server:
     port: 5761 # Specify the port of this Eureka instance

eureka:
  server:
         enable-self-preservation: false #Turn off self-protection and delete the node that has been shut down
  instance:
         hostname: discovery1 # Specify the hostname of this Eureka instance
  client:
         registerWithEureka: false #Under the default setting, the service registration center will also try to register itself as a client, and disable its client registration behavior here
         fetchRegistry: false #Since the responsibility of the registry is to maintain service instances, it does not need to retrieve services, so it is also set to false
    serviceUrl:
      defaultZone: http://discovery1:5761/eureka/

security:
  basic:
    enabled: true
  user:
     name: root
     password: root

spring:
  rabbitmq:
      host: localhost
      username: guest
      password: guest
  application:
    name: eureka-server

There is no problem in local development and operation. When you want to go to the test environment, you need to deploy two registration centers to ensure the high availability of the service. Since you have not deployed the eureka cluster before, you searched the configuration file on the Internet. You only need to change the defaultZone. To allow the two registration centers to register with each other. The configuration is as follows:

The first:

server:
     port: 5761 # Specify the port of this Eureka instance

eureka:
  server:
         enable-self-preservation: false #Turn off self-protection and delete the node that has been shut down
  instance:
         hostname: discovery1 # Specify the hostname of this Eureka instance
  client:
         registerWithEureka: false #Under the default setting, the service registration center will also try to register itself as a client, and disable its client registration behavior here
         fetchRegistry: false #Since the responsibility of the registry is to maintain service instances, it does not need to retrieve services, so it is also set to false
    serviceUrl:
      defaultZone: http://discovery2:5761/eureka/

The second station:

server:
  port: 5761 # Specify the port of this Eureka instance

eureka:
  server:
         enable-self-preservation: false #Turn off self-protection and delete the node that has been shut down
  instance:
         hostname: discovery2 # Specify the hostname of this Eureka instance
  client:
         registerWithEureka: false #Under the default setting, the service registration center will also try to register itself as a client, and disable its client registration behavior here
         fetchRegistry: false #Since the responsibility of the registry is to maintain service instances, it does not need to retrieve services, so it is also set to false
    serviceUrl:
      defaultZone: http://discovery1:5761/eureka/

After the operation, I went to the test and entered the http://discover1:5761 page as follows:

The smile on the face gradually disappeared, and the second machine appearedunavailable-replicas (unavailable copy), then went online to check the problem and concluded:

These two items need to be set registerWithEureka = true, fetchRegistry = true

After setting up, deploy again, visit http://discover1:5761 The problem still exists, emmmm......

After trying the various solutions provided by the students on the Internet, it still didn't work, and I fell into contemplation...

Suddenly found that my configuration file seems to be different from theirs. I suddenly realized that I set eureka security in the configuration file, and configured the user name and password.

Therefore, after directly changing the defaultZone, it will definitely not work, and it will certainly not be accessible. Modify the configuration again as follows:

server:
     port: 5761 # Specify the port of this Eureka instance

eureka:
  server:
         enable-self-preservation: false #Turn off self-protection and delete the node that has been shut down
  instance:
         hostname: discovery1 # Specify the hostname of this Eureka instance
  client:
    registerWithEureka: true 
    fetchRegistry: true 
    serviceUrl:
      defaultZone: http://root:root@discovery2:5761/eureka/

security:
  basic:
    enabled: true
  user:
     name: root
     password: root

spring:
  rabbitmq:
      host: localhost
      username: guest
      password: guest
  application:
    name: eureka-server

I was careless, just modified it according to the method on the Internet, did not consider the configuration of my project, emmmm... wasted a lot of time, so record it to remind classmates like me to solve the problem quickly.

Deploy after modifying again, visithttp://discover1:5761

Perfect, you're done