I am deploying a React.js front-end (built with create-react-app) and a back-end with a CRUD API that connects to Cloud SQL.
Is this a good way?
React.js front-end is a default service.
Back-end API is backend service
I'm familiar with deploying to Heroku, which had front-end and back-end on different ports. Would that work for App Engine?
This is all in Node.js.
I don't see any issues with the described design.
To get you on track you can take a look in Stack Overflow thread How to deploy create-react-app to Google Cloud or the following tutorial.
You can run applications on different ports with setting port forwarding in your app.yaml file.
The design pattern is good.
You only need to create a dispatch.yaml file which is only one per project.
Your dispatch.yaml file would look something like this:
dispatch:
- url: "*/backend/*"
service: backend
Then your frontend at project-id.appspot.com will simply make requests to https:/project-id.appspot.com/backend/ *, and these requests will be redirected to the API service.
Related
ho guys, i want to deploy my next.js application to vercel. well there's no problem about that. But for the reason of a scalable application i have my used backend written separately in Node.js. I also want this to be used with the Next.js Frontend preferably by hosting it on vercel.
So is this possible by default? I don't understand how to achieve this and what will be the necessary steps after having the backend, say, deployed on vercel.
Can you elaborate?
unfortunately you cannot host a backend on Vercel. If you want to host a backend simply you should look at Heroku or Railway. You should then be able to connect your Vercel frontend to your backend hosted by Heroku or Railway
I'm absolutely new to Azure, CI, and CD.
I have an app consisting of a frontend developed under React + Typescript which sends CRUD requests to a typescript backend. Both front and end are NPM projects.
I've no idea on how to step from development environment to production, deploying to Azure.
Should I create two separate AppServices, one for each project?
If that's the case, how do I then connect them?
Do I just need to change the URL the server listens to?
As you can see I've no idea of what I'm doing but I'm eager to learn so any feedback is much appreciated
The most common approach is
Deploy frontend app ( react ) in one app service.
Backend nodejs app on a different app service.
Then, call the api of the nodejs app from the frontend.
Or,
Set up and run the client and server on the same server using a gulp file and deploy that project on an app service. In that case, there is no cross-site communication. This is the most preferred one.
Using the same app service for both front end and back end application. Refer nodeexpress-backend-with-angular-front-end-in-a-single-azure-web-app.
Using Cors to connect front end and backend Check here
Refer SO Link 1 & Link 2
And You can use deploy staging slots in app service to use different environment like (test/dev, Production).
I have a react native app that I'm building using expo. For the backend, I am using node.js and Express.
My app hierarchy is such that I have a frontend folder and a backend (server) folder. For example,
Root
|
|Frontend
|Various screens & associated files
|Server
|Routes & other files
For my project, is it possible to just host the backend and not the rest of the app? Such that when I fetch data in the frontend using HTTP requests, instead of routing through localhost (http: //RandomIP:PORT/route) I would use the heroku address as the routing address. I would also host the SQL database along with it.
I tried to follow along with the Heroku documentation, but it seemed like that was for hosting the entirety of the app / web apps instead of mobile, and I ran into constant errors.
I would like to point out that, unlike web pages, mobile apps cannot be hosted on the server and fetched on-demand. In other words, do not try to upload your react-native code to Heroku instead just upload your backend only and then make HTTP requests through the URL provided by Heroku after you have deployed your code.
Therefore go into your backend codebase, initialize a git repository and just deploy that Heroku. Also, you will need to host your SQL database on another service such as Google's Cloud SQL or Amazons AWS Database Services.
I have frontend and backend both are using nodejs. Frontend is exposed and every request for api goes through the proxy module in the frontend, and both frontend and backend communicate on different modes.
Is there a way to deploy the above module, using code pipeline onto the Elastic Beanstalk platform ?
Why not deploy the front end on Amplify AWS?
AWS Amplify is a set of tools and services that can be used together or on their own, to help front-end web and mobile developers build scalable full stack applications, powered by AWS.
You just need to link your Repo and select your branch and everything will be deployed and generate also a random free domain or you can set your domain from the domain management section
Using the App Engine Flexible Environment, I'm preparing to deploy an Angular 4 client and am looking into Cloud Endpoints to handle my node.js/express microservices as it seems to simplify securing and authenticating endpoint requests, and I wanted to clarify a few things:
Do I use cloud-endpoints as an API Gateway which routes requests to the individual microservice backends or are the microservices supposed to be built as individual endpoints-apps themselves?
Do I host the Angular 4 app statically (server agnostic), and make endpoint requests directly to the Gateway/microservice from the ng client, or is the app hosted through a server framework (e.g. node.js/express) which then passes on the request along to the Gateway/microservice
Endpoints is an API gateway, but it currently only routes to a single backend. On Flex, it's whatever app you deploy. The Endpoints proxy sits in front of your backend, transparently to the client, and the client requests will pass through the gateway. See the docs for how to set up your Flexible environment.
In conjunction with a dispatch.yaml directive to handle routing, you can use GCE as a gateway to multiple microservices running as services in a given project.
You can call the services directly or proxy through a server.