We have load balancing, discovery, etc in the Spring cloud. But the same load balancing, discovery is available in Kubernetes as well. So just wanted to know when we should go with Spring cloud (load balancing or discovery) and when to use Kubernetes
It depends on your use-case. There can be situations where you need to directly use Eureka server registry and Eureka client discovery offered by Spring Cloud Netflix. Ribbon is the client side load balancer provided by Spring Cloud Netflix.
In my experience, it is not impossible to use Eureka in any environment. It can be your local data centre or cloud infrastructure. However when it comes to deployment environment, there are so many alternatives for us to achieve the server registry mechanism. Sometimes those alternatives are the best solutions. I will give you an example below...
If you host your application in your local server (Local data centre)
Now in this scenario you can use Eureka and continue your server registry and discovery mechanism. (That is not the only way. I mentioned Eureka for this scenario because it would be a good use case for it)
If you host your application in AWS infrastructure
The AWS environment gives you lots of benefits and services such that you can forget the burden of maintaining and implementing Eureka. You can achieve simply the same behaviour by AWS load balancers, AWS target groups and even more by adding AWS auto scaling groups. In AWS it self there are so many other ways to achieve this as well.
Long story in short that for your scenario, you can continue using the power of Kubernetes and get the privilege unless you have a specific reason to use Eureka and put a large effort to implement it. You should select what suits the best depending on time, effort, maintainability, performance etc.
Hope this helps for you to get an idea. Happy coding!
Related
For microservices architecture not in the cloud :
What is the difference between the load balancer of Kubernetes and the load balancer of spring cloud?
what are the advantages of implementing Eureka and spring boot load balancer when using Kubernetes for deployment rather than using Kubernetes load balancer and kubernetes service discovery?
Quite a complex question but here are my thoughts on the matter:
Before actually answering the questions, I must say that my experience setting up Kubernetes load balancer outside of public clouds is a hassle so I wouldn't recommend the approach but, assuming that it is not an issue for you, the main difference is how things are set ut. I would say that the biggest advantage of having it in Kubernetes is that you may also have service mesh solution like Istio that gives you other advantages that just load balancing. Furthermore, it would be easier to do canary (or other special type of) deployments with infrastructure load balancing than with Spring.
The only real advantage I see is if you have different teams with responsibility of the infrastructure, deployment and coding. Say if the Kubernetes team is responsible for creating services, deployments, etc and is overloaded, you might get code out faster if your dev team has capacity and competence but, again, there would be no point using Kubernetes then. Or in case you cannot actually create the LoadBalancer service in Kubernetes (as mentioned - not always straightforward in non cloud env)
On a side note, if you are on the way to deploying Kubernetes internally and want to get it to work, have a look at Metallb
We are currently setting up a Micro-service Architecture using spring boot and netflix components, for Deployment we are planning to go with aws kubernetes(EKS) setup. We are in a in a dilemma to choose whether to use netflix's Eureka & ribbon services on Kubernetes for service discovery or to use Kubernetes own service discovery Mechanism. The advantages I see in using k8s service discovery is that horizontal scaling becomes easy. Any thoughts on this will be really good for us to take it in the right direction.
Thanks.
If you don't have a strong case for client side load balancing I would not use Eureka and ribbon. I am getting the load balancing and service discovery from my platform (k8s) for free. If not for client side load balancing, there is no value that eureka and ribbon brings other than beefing up your resume :). On the other hand, it brings another stack that you need to maintain in the long term.
This is a kind of related question
I am learning building microservices using spring boot, Spring Cloud(netflix OSS Components). I have used netflix Eureka for service discovery, zuul for api gateway, ribbon, feign while running in my local machine.
Netflix eureka, zuul, ribbon, feign spring cloud config are not useful when we deploy to PCF?(if yes what are the alternatives available in pcf and how to configure them?)
As who are building microservices follows CI/CD approach, how developer verify working of their micro services before pushing code as we don't use eureka, zuul,ribbon,feign in production pcf. (how to simulate pcf environment in developer machine?).
I'd suggest to read below content before implementing if you have any doubt regarding usage of Eureka and Zuul, you will get all answers yourself.
https://github.com/Netflix/eureka/wiki/Eureka-at-a-glance
https://github.com/Netflix/zuul/wiki
As who are building microservices follows CI/CD approach, how developer verify working of their micro services before pushing code as we don't use eureka, zuul,ribbon,feign in production pcf.
Answer to this question is: You must be aware of JUnit test cases, so you can run you test cases using deployment pipelines to make sure all your functionalities are working as expected or you can use Test Automation for the same.
(how to simulate pcf environment in developer machine?).
Answer to this one:
You can use eclipse plugin you are using eclipse/STS IDE. Or you can connect all PCF services from you local machine using CloudFactory
#Bean
public Cloud cloud() {
return new CloudFactory().getCloud();
}
https://docs.pivotal.io/pivotalcf/2-1/buildpacks/java/sts.html
Here are some thoughts:
Eureka Service discovery: in my opinion this is not strictly necessary when running on PCF. When you push an app on PCF usually a route is assigned to your app, and you can use this Route as a poor man's service discovery. Eureka would allow you to use client-side load balancing in the case of container-to-container networking, but usually you wouldn't need this.
Zuul: Can be very useful also on CloudFoundry in case you are doing things like writing frontend-for-backend services, providing frontends for different devices (mobiles, desktops, i-pads) that use the same backend services. Might also be useful for an authentication/authorization layer or rate-limiting. One native CloudFoundry alternative would be to use route-services for tasks such as rate limiting, authentication/authorization.
spring-cloud-config: makes sense if you want your configuration to be under version control for different environments. This is useful no matter if you are running on CloudFoundry or not. I don't know of any alternatives on plain CloudFoundry.
spring-cloud-feign: makes sense if you want use annotations such as #RequestMapping with your Feign client interfaces. This is independent on if you are running on CloudFoundry or not. AFAIK there are no alternatives for this in case you want to use Spring MVC annotations with Feign.
ribbon: makes sense if you want to use client side load balancing as opposed to let the CloudFoundry router to do the load balancing for you.
How developers can check locally if this works for them:
In general, I don't believe developers should need to check locally if their app is working fine together with zuul, cloud-config-service, and eureka.
They could check this in a dev or test space or environment though.
If they really want to check this on their local machine, they could download PCFDev and run these infrastructure components there.
Hope this helps.
Is Spring cloud config server an example of sidecar application for microservices?
Do you mean if the Spring Cloud Config Server itself is what the Spring Cloud documentation labels as Sidecar? Then no, as far as I know it is just a plain, regular Spring Boot app.
A Sidecar as referred to in Polyglot support with Sidecar is a Spring Boot application that acts as a bridge between your service infrastructure and a service that is not written in a JVM language. Apps written in Python, Go, Ruby, C#, NodeJS, Erlang or really any other language that can bind something to a port come to mind.
The benefits of the Sidecar are, that your Non-JVM apps
service discovery become automatically discoverable through Eureka, which means that JVM services can resolve the host:port/<service-id> of the Non-JVM apps as well as the other way around,
monitoring are monitorable through the same health-endpoints-infrastructure that is available in Spring Boot (Actuator), i.e. by manually providing the health endpoint in the Non-JVM app Eureka knows when the Non-JVM service is down
routing/proxying query the services by either manually looking up their hosts/ports or proxying these requests through Zuul, which in turn resolves their current addresses through Eureka
balancing be load balanced by Ribbon and
configuration may consume configuration properties provided via Spring Cloud Config.
I hope this answer addresses your question, if not (or someone finds it to be inaccurate or misleading) just let me know and I delete it to make room for something more suitable. ;-)
If I have multiple Spring Boot embedded tomcat containers and each can have service endpoints like
http://localhost:8080/employeeSelfService/getDetails
http://localhost:8081/employeeSelfService/getDetails
How can do load balancing using 2 micro services such that clients can hit any of the URL's mentioned based on some load balancing startegy
One option thats come to my mind is to use NetFlix Curator (or) have a apache webserver acting as reverse proxy but with apache, when you create new instances of your services, you will have have an entry of that service as a member in httpd.conf
Does Spring Boot provides any service discovery and load balancing mechanism ?
Spring Boot does not provide this feature, as it is already usually provided by a reverse proxy such as apache/nginx running in front of the Spring Boot server.
See here for an example here how the commercial version of nginx provides the functionality of dynamically scaling and reducing the upstream nodes.
So in this case it's for the dynamic instance, in this case the Spring Boot process to signal it's presence/unregister itself to the upstream server at initialization/shutdown.
See here how to do so in the case of nginx, this procedure will be different from server to server.
Arguably it's not really an application's role to manage its own load-balancing, and Spring Boot focuses on the implementation of an application (or service, equivalently). We have been thinking about whether we could provide features in Spring (Boot or otherwise) to make it easy to write your own load-balancer, or service registry app, but even then I don't think that was what the question was really about (or was it?).
If I interpret the question, and the example use case, literally, I would say that the most natural answer is an out-of-the-box reverse proxy solution (as the other answers pointed out). I also note that such a reverse proxy is an essential and natural part of a PaaS solution, so if you need it to "just work" and don't want to know about the details, PaaS would be a natural path (e.g. see cloudfoundry as an example of such a solution that I happen to have worked on).
Indeed Spring Boot has not inherit support for load-balancing. Just to add to the list of available solutions for load-balancing, here are the instructions to configure an Apache for load-balancing.