As far as I understood, Istio Destination Rules can define load balancing policies to reach a subset of a service, e.g. subset based on different versions of the service. So the Destination Rules are the first level of load balancing.
The request will eventually reach a K8s service which is generally implemented by kube-proxy. Kube-proxy does a simple load-balancing with the pods in its back-end. Here is the second level of load balancing.
Is there a way to remove the second load-balancer? For example, could we create a lot of services instances that offer the same service and can be load-balanced by Destination Rules and then have only one pod per service instance, so that kube-proxy does not apply load-balancing?
According to istio documentation:
Istio’s traffic routing rules let you easily control the flow of traffic and API calls between services. Istio simplifies configuration of service-level properties like circuit breakers, timeouts, and retries, and makes it easy to set up important tasks like A/B testing, canary rollouts, and staged rollouts with percentage-based traffic splits. It also provides out-of-box failure recovery features that help make your application more robust against failures of dependent services or the network.
Istio’s traffic management model relies on the Envoy proxies that are deployed along with your services. All traffic that your mesh services send and receive (data plane traffic) is proxied through Envoy, making it easy to direct and control traffic around your mesh without making any changes to your services.
If you’re interested in the details of how the features described in this guide work, you can find out more about Istio’s traffic management implementation in the architecture overview. The rest of this guide introduces Istio’s traffic management features.
This means that the istio service mesh is communicating via envoy proxy which in turn relies on kubernetes networking.
We can have an example where a VirtualService that is using istio ingress gateway load-balances it's traffic to two different services based on labels. Then those services can have multiple pods.
Istio load-balancing in this case works only on (layer 7) which results with route to specific endpoint (one of the services) and relies on kubernetes to handle connections and the rest including service round-robin load-balancing (layer 4) in case of multiple pods.
The advantage of having single service with multiple pods is obviously easier configuration and management. In case of 1 pod per service, each service would need to be reconfigured separately and loses all of its ability to scale features.
There is a great video on Youtube which partially covers this topic:
Life of a packet through Istio by Matt Turner.
I highly recommend watching as it explains how istio works on a fundamental level.
Related
I see that for every knative service, 2 VirtualService objects are created namely ksvc-ingress which has knative-serving/knative-ingress-gateway & knative-serving/knative-local-gateway gateways configured and ksvc-mesh which has mesh as the gateway.
I can see the knative-serving/* gateways using kubectl but I am unable to find the mesh gateway object in any namespace. I would like to understand if mesh here denotes some special object or is it an istio keyword representing something else?
The mesh name is a keyword, as you guessed. That keyword represents the East-West traffic between Pods in the Kubernetes cluster, as managed by the Istio sidecar. You can think of those VirtualServices as being programmed onto each sidecar to do the routing and traffic splitting next to the request sender, rather than needing to route to a central service / gateway.
As you noticed, knative uses istio as a service mesh.
In the Istio context mesh is not an object (or resource) like, for example, a Service. Istio About page explain what Service Mesh is:
A service mesh is a dedicated infrastructure layer that you can add to your applications. It allows you to transparently add capabilities like observability, traffic management, and security, without adding them to your own code. The term “service mesh” describes both the type of software you use to implement this pattern, and the security or network domain that is created when you use that software.
So mesh is a term that encapsulate all Istio objects (istio-proxy containers, Virtual Services, Ingress Gateways etc.), that work together to allow for traffic management inside cluster.
A Gateway is a load balancer operating at the edge of the mesh receiving incoming or outgoing HTTP/TCP connections.
We have many internet services, what are the considerations whether to use alb per service or single alb for all using listener rule pointing to target 🎯 group.
The services has its own clusters/target group with different functionality and different url.
Can one service spike impact other services?
Is it going to be a single point of failure ?
Cost perspective ?
Observability, monitoring, logs ?
Ease of management ?
Personally I would normally use a single ALB and use different listeners for different services.
For example, I have service1.domain.com and service2.domain.com. I would have two hostname listeners in the same ALB which route to the different services.
In my experience ALB is highly available and scales very nicely without any issues. I've never had a service become unreachable due to scaling issues. ALB's scale based on "Load Balancer Capacity Units" (LBCU). As your load balancer requires more capacity, AWS automatically assigns more LBCU's which allows it to handle more traffic.
Source: Own experience working on an international system consisting of monoliths and microservices which have a large degree of scaling between timezones.
You don't have impact on service B if service A has a spike, but the identification of which service is having bad times could be a little pain.
For monitoring perspective it's is a bit hard because is not that easy to have a fast identification of which service/target is suffering.
For management, as soon as different teams need to create/management its targets it can create some conflicts.
I wouldn't encourage you using that monolith architecture.
From cost perspective you can use one load balancer with multi forward rules, but using a single central load balancer for an entire application ecosystem essentially duplicates the standard monolith architecture, but increases the number of instances to be served by one load balancer enormously. In addition to being a single point of failure for the entire system should it go down, this single load balancer can very quickly become a major bottleneck, since all traffic to every microservice has to pass through it.
Using a separate load balancer per microservice type may add additional overhead but it make single point of failure per microservice in this model, incoming traffic for each type of microservice is sent to a different load balancer.
Both seem to do the same things. From what I've gathered Istio does routing at Inngress level and ESP at container level. I'm still understanding Istio.
According to Google cloud documentation:
Extensible Service Proxy
The Extensible Service Proxy (ESP) is an Nginx-based high-performance, scalable proxy that runs in front of an OpenAPI or gRPC API backend and provides API management features such as authentication, monitoring, and logging. See About Endpoints and Endpoints: Architectural overview for more information.
Extensible Service Proxy V2 Beta
The Extensible Service Proxy V2 Beta (ESPv2 Beta) is an Envoy-based high-performance, scalable proxy that runs in front of an OpenAPI API backend and provides API management features such as authentication, monitoring, and logging. See About Endpoints and Endpoints: Architectural overview for more information.
ESPv2 Beta supports version 2 of the OpenAPI Specification. ESPv2 Beta does not currently support gRPC.
ESPv2 Beta is only supported for use for the Beta versions of Endpoints for Cloud Functions and for Cloud Run. ESPv2 Beta is not supported for Endpoints for App Engine, GKE, Compute Engine, or Kubernetes.
According to istio github documentation:
Introduction
Istio is an open platform for providing a uniform way to integrate microservices, manage traffic flow across microservices, enforce policies and aggregate telemetry data. Istio's control plane provides an abstraction layer over the underlying cluster management platform, such as Kubernetes.
Istio is composed of these components:
Envoy - Sidecar proxies per microservice to handle ingress/egress traffic between services in the cluster and from a service to external services. The proxies form a secure microservice mesh providing a rich set of functions like discovery, rich layer-7 routing, circuit breakers, policy enforcement and telemetry recording/reporting functions.
Note: The service mesh is not an overlay network. It simplifies and enhances how microservices in an application talk to each other over the network provided by the underlying platform.
Mixer - Central component that is leveraged by the proxies and microservices to enforce policies such as authorization, rate limits, quotas, authentication, request tracing and telemetry collection.
Pilot - A component responsible for configuring the proxies at runtime.
Citadel - A centralized component responsible for certificate issuance and rotation.
Citadel Agent - A per-node component responsible for certificate issuance and rotation.
Galley- Central component for validating, ingesting, aggregating, transforming and distributing config within Istio.
Operator- The component provides user friendly options to operate the Istio service mesh.
Istio currently supports Kubernetes and Consul-based environments. We plan support for additional platforms such as Cloud Foundry, and Mesos in the near future.
The ESP v2 Beta is also based on Envoy proxy just like Istio. However there are advanced features that Istio has that ESP v2 does not have yet as it is still in beta. As for ESP v1 it is more like nginx ingress. All of these tools are able to do the routing tasks, however each tool has different mechanisms under the hood and offer different amount of configuration flexibility and complexity.
Hope it helps.
I'm trying to work out whether I could use one of the (A/E/N)LBs to load balance gRPC traffic. A simple round robin would suffice in our case.
I've read that ALB doesn't fully support HTTP2 and therefore can't be used with gRPC. Specifically lack of support of sending HTTP2 traffic downstream and lack of support for trailer headers was mentioned. Is it still true?
Couldn't find any definitive answers with regards to NLBs or "classic" ELBs. Any hints?
As of October 29, 2020, Application Load Balancers now support HTTP/2 and gRPC load balancing. From the announcement:
To use the feature on your ALB, choose HTTPS as your listener protocol, gRPC as the protocol version for your target group and register instance or IP as targets for the configured target group. ALB provides rich content based routing features that will let you inspect gRPC calls and route them to the appropriate target group based on the service and method requested. Within a target group, ALB will use gRPC specific health checks to determine availability of targets and provide gRPC specific access logs to monitor your traffic.
The support for gRPC and end-to-end HTTP/2 is available for existing and new Application Load Balancers at no extra charge in all AWS Regions. To learn more, please refer to the blog post, demo, and the ALB documentation.
Using gRPC on AWS had some major challenges. Without full HTTP/2 support on AWS Application Load Balancer, you have to spin up and manage your own load balancers. Neither NLB and ELB are viable alternatives on AWS due to issues with traffic to and from the same host, dynamic port mappings, SSL termination complications, and sub-optimal client and server-side round-robining of TCP connections.
gRPC demonstrated performance improvements, however, it would take considerable infrastructure efforts to adopt, whether it be using LBs such as Nginx or Envoy; or setting up a service mesh with something of the likes of Istio. Another possibility would be to make use of thick client load balancing, though this would also require additional service discovery infrastructures such as Consul or ZooKeeper.
AWS recently announced a new service called AWS App Mesh. AWS App Mesh supports HTTP2 and gRPC services
gRPC can now model and manage their inter-service communications using AWS App Mesh.
Reference:
https://aws.amazon.com/about-aws/whats-new/2019/11/aws-app-mesh-now-supports-http2-and-grpc-services/
https://aws.amazon.com/app-mesh/
https://docs.aws.amazon.com/app-mesh/latest/userguide/what-is-app-mesh.html
I have been working with spring and now would like to learn spring boot and microservices. I understand what microservice is all about and how it works. While going through docs i came across many things used to develop microservices along with spring boot which i am very much confused.
I have listed the systems below.and the questions:
Netflix Eureka - I understand this is service discovery platform.
All services will be registered to eureka server and all
microservices are eureka clients. Now my doubt is , without having
an API gateway is there any use with this service registry ? This is
to understand the actual use of service registry.
ZUULApi gateway- I understand ZUUL can be used as API gateway which is basically a load balancer , that calls appropriate
microservice corresponding to request URL. iS that assumption
correct? will the api gateway interact with Eureka for getting the
appropriate microservice?
NGINX - I have read NGINX can also be used as API gateway? Is that possible? Also i read some where else like NGINX can be used as a service registry , that is as an alternate for Eureka ! Thus which is right? Api gateway or service registry or both? I know nginx is a webserver and reverse proxies can be powerfully configured.
AWS api gateway - Is this can also be used as an alternate for ZUUL?
RIBBON - for what ribbon is used? I didn't understand !
AWS ALB- This can also be used for load balancing. Thus do we need ZUUL if we have AWS ALB?
Please help
without having an API gateway is there any use with this service registry ?
Yes. For example you can use it to locate (IP and port) of all your microservices. This comes in handy for devops type work. For example, at one project I worked on, we used Eureka to find all instances of our microservices and ping them for their status (/health, /info).
I understand ZUUL can be used as API gateway which is basically a load balancer , that calls appropriate microservice corresponding to request URL. iS that assumption correct?
Yes but it can do a lot more. Essentially because Zuul is more of a framework/library that you turn into a microservice, you can code it to implement any kind of routing logic you can come up with. It is very powerful in that sense. For example, lets say you want to change how you route based on time of day or any other external factors, with Zuul you can do it.
will the api gateway interact with Eureka for getting the appropriate microservice?
Yes. You configure Zuul to point to Eureka. It becomes a client to Eureka and even subscribes to Eureka for realtime updates (which instances have joined or left).
I have read NGINX can also be used as API gateway? Also i read some where else like NGINX can be used as a service registry , that is as an alternate for Eureka ! Thus which is right? Api gateway or service registry or both?
Nginx is pretty powerful and can do API gateway type work. But there are some major differences. AFAIK, microservices cannot dynamically register with Nginx, please correct me if I am wrong... as they can with Eureka. Second, while I know Nginx is highly (very highly) configurable, I suspect its configuration abilities do not come close to Zuul's routing capabilities (due to having the whole Java language at your disposal within Zuul to code your routing logic). It could be the case that there are service discovery solutions that work with Nginx. So Nginx will take care of the routing and such, but service discovery will still require a solution.
Is this can also be used as an alternate for ZUUL?
Yes AWS API Gateway can be used as a Zuul replacement of sorts. The issue here, just like Nginx, is service discovery. AWS API Gateway lets you apply logic to your routing... though not as open ended as Zuul.
for what ribbon is used?
While you can use the Ribbon library directly, for the most part consider it as an internal dependency of Zuul. It helps Zuul do the simple load balancing that it does. Please note that this project is in maintenance mode and not recommended any more.
This can also be used for load balancing. Thus do we need ZUUL if we have AWS ALB?
You can use ALB with ECS (elastic container service) to replace Eureka/Zuul. ECS will take care of the service discover for you and will map all instances of a particular service to a Target Group. Your ALB routing table can then route to Target Groups based on simple routing rules. The routing rules in ALB are very simple though, but improving over time.
Different systems which can be used for the working of microservices, that comes along with spring boot:
Eureka:
Probably the first microservice to be UP. Eureka is a service registry, means , it knows which ever microservices are running and in which port. Eureka is deploying as a sperate application and we can use #EnableEurekaServer annotation along with #SpringBootAPplication to make that app a eureka server. So our eureka service registery is UP and running. From now on all microservices will be registered in this eureka server by using #EnableDiscoveryClient annotation along with #SpringBootAPplication in all deployed microservices.
Zuul: ZUUL is a load balancer , routing application and reverse proxy server as well. That is before we were using apache for reverse proxy things , now , for microservices we can use ZUUL. Advantage is, in ZUUL we can programatically set configurations, like if /customer/* comes go to this microservice like that. Also ZUUL can act as a load balancer as well , which will pick the appropriate microservice in a round robin fashion. SO how does the ZUUL knows the details of microservices, the answer is eureka. It will work along with eureka to get microservice details. And in fact this ZUUL is also a Eureka client where we should mark using #EnableDiscoveryClient, thats how these 2 apps(Eureka and zuul) linked.
Ribbbon:
Ribbon use for load balancing. This is already available inside ZUUL, in which zuul is using Ribbon for load balancing stuff. Microservices are identified by service-name in properties file. IF we run 2 instances of one microservices in different port, this will be identified by Eureka and along with Ribbon(Inside zuul), requests will be redirected in a balanced way.
Aws ALB , NGINX , AWS Api gateway etc: There are alternatives for all the above mentioned things. Aws is having own load balancer, service discovery , api gateway etc . Not only AWS all cloud platofrms ,like Azure, have these. Its depends which one to use.
Adding a general question as well , How these microservices communicate each other: Using Resttemplate or Feignclient actual rest API can be called or Message queues like Rabbit MQ etc can be used .
Eureka can be used in conjunction with NGINX, which leads to very powerful combination.
I am using it on AWS EC2 environment. Previously instead of NGINX I was using Spring Cloud Gateway and before that Zuul. Depending of the load Spring Cloud Gateway was running on AWS t3.medium or t3.large instances. After moving to NGINX I am using t3.micro (8 times less memory) instance. I am almost sure that I can do the trick and with t3.nano (16 times less memory) instance, but I wanted to be sure that there will be no surprises.
Below are the high level steps what you have to do in order to plug NGINX in the Eureka ecosystem. More details you can find in NGINX With Eureka Instead of Spring Cloud Gateway or Zuul article.
Create a service which can read the configuration of all applications from Eureka and to 'translate' it to NGINX configuration.
Create a cronjob entry which at certain period will read the configuration from the above service and will call the NGINX hot reload
NGINX which will consume the configuration produced from the service and the cronjob and will work as API Gateway