googleapis VS microservices-demo

Rich Error Model: This model enables servers to return and clients to consume additional error details expressed as one or more protobuf messages. It further specifies a standard set of error message types to cover the most common error (QuotaFailure, PreconditionFailure, BadRequest, etc). When an error occurs, the server returns the appropriate status code along with an optional error message.

They list search in their public api

https://github.com/googleapis/googleapis/blob/288aa7fb71c9b6...

We structure the repo according to proto packages. It's quite similar to how the googleapis repository is structured.

It's unfortunate that the spec doesn't contain custom fields to a sub-object like other RPC specs, like proto Status [1]. They should have had the message go into a field named "message" and not "detail". And have a field like "details" where the opaque type is serialized, which should be named by the "type" field. The problem is that systems with existing error types may have field name conflicts with type, title, status, detail, or instance, so we'd just dump the actual error into a custom "extension member" which by definition, isn't standard.

[1] https://github.com/googleapis/googleapis/blob/1c8a25ab153eef...

Protobuf (https://github.com/googleapis/googleapis)

All the required changes can be viewed in our last demo, the grpc-rest-app implementation. First, we need to update our proto service interface to help the proxy service make our gRPC service methods available at the right URLs and for the correct HTTP operations. To do this, the Google API HTTP library provides annotations we can add to our proto to describe the correct mappings. The buf tool allows us to include the googleapis dependency as a plugin in our buf.yaml file).

I think this only makes sense if the 3rd party is also throwing custom exceptions.

If you want to reduce coupling you should avoid throwing custom exceptions at all. Semantic information can go in the error message and log. The error type should be used to indicate to your program whether an error is recoverable, retriable or some other action needs to be taken. For example google on has 16 canonical error codes for all APIs.

https://github.com/googleapis/googleapis/blob/master/google/...

OpenAPI and possibly developing reusable, versioned client libraries could help, but it's a major undertaking that gRPC makes redundant. I'd be tempted to use grpc-gateway even if I had to implement a REST API. Try looking into buf and monorepo structures for proto management, e.g. something like GoogleCloudPlatform/microservices-demo. For more thorough proto and grpc-gateway definition examples, see googleapis/googleapis.

Firestone does provide global consistency, so the following quote is incorrect:

> In Cloud Firestore, the data on the client are loaded from the database at different points in time. Even if you listen for realtime updates, results from separate queries will not remain in sync. This creates consistency anomalies and bugs in your app.

Here is a link to the protocol documentation that the clients use to support it: https://github.com/googleapis/googleapis/blob/d0b394f188e8c3...

I'd link to the client implementation but it's quite involved.

Well there is Timestamp defined as a well known type which is available to all implementations despite not being a primitive type [1]. Plus one is obviously able to define any other custom types if necessary- eg as seen in [2].

[1] https://developers.google.com/protocol-buffers/docs/referenc...

[2] https://github.com/googleapis/googleapis/blob/master/google/...

You can check https://github.com/GoogleCloudPlatform/microservices-demo for Kubernetes show-casing

Skyramp provides a sample project, sample-microservices, which serves as an excellent starting point for demonstrating testing and mocking with a full-featured distributed application. The application is based on Google's Online Boutique repo, which is an e-commerce store consisting of 11 different microservices. The docker-compose-demo branch referenced above showcases how Skyramp can be seamlessly integrated with Docker Compose for testing microservices with no local setup required. You can also clone the repository and explore the structure of the microservices setup for your own purposes.

Our starting point is the hot-code-reload-demo branch in Skyramp's letsramp/sample-microservices GitHub repo. You can use your browser to navigate to the correct branch in the repo here. The sample-microservices repo contains a demo project based on GCP's Online Boutique with added support for REST and Thrift APIs. This sample e-commerce application is perfect for demonstrating cloud-native development and testing, including debugging with Hot Code Reload with Skyramp.

This blog features the Skyramp fork of Google’s popular cloud-native microservices demo app, Online Boutique. Online Boutique is a web-based e-commerce app containing microservices that mimic real-world services, such as a product catalog, shopping cart, ad service, recommendation service, payment service, and others. The services use gRPC APIs by default, but Skyramp has also added support for REST and Thrift APIs.

I'm planning to host this application: Google Microservices Demo (It's an online shop)

Well, what is considered "real" data here? Why couldn't you simply set up a managed k8s cluster with some prometheus monitoring and run the microservices-demo on it. There is even a synthetic load generator. You could purposefully add in specific kinds of faults into the working system, ones that are supported in metasploit so you can automate intrusions. Consider some goals for gaining access like: exfiltration, denial of service, ransomware. Then consider how you might detect such attacks purely from what you can read out of the prometheus time series data (eg. high egress traffic plus high req/s to redis might mean an exfiltration).

The microservices source code repository for this project is from this link; google-microservices-demo, containing 11 services we will deploy with this demo. Also, from the same repo, it was illustrated and visualized how these services are connected to each other including a 3rd party service for database - redis. Among the services, Frontend serves as an entrypoint for all the services receiving external requests from the browser. Meanwhile, the load generator deployment is optional, so in this demo we wouldn't bother deploying it.

This demo repo might be a good starting point: https://github.com/GoogleCloudPlatform/microservices-demo

OpenAPI and possibly developing reusable, versioned client libraries could help, but it's a major undertaking that gRPC makes redundant. I'd be tempted to use grpc-gateway even if I had to implement a REST API. Try looking into buf and monorepo structures for proto management, e.g. something like GoogleCloudPlatform/microservices-demo. For more thorough proto and grpc-gateway definition examples, see googleapis/googleapis.

I also just discovered Google Cloud's microservices-demo repository, which has some samples of how to set up otel observability and GCP-specific Go profiling on GCP. I wish I'd found it before setting up otel myself.