Related Reference Issue: Redeploying OpenAPI spec into App Engine standard environment
When using Cloud Endpoints via App Engine, what would be the best way to approach continuous deployments? If the OpenApi spec changes, the backend service tied to the endpoint needs to be redeployed meaning that modifications to a "v1" of an endpoint would require downtime in the service while the service is in deployment.
Does this mean that any time a new spec is generated a "breaking change" occurred and I should increment the version, redeploy the backend service, and update the url on any services that use that endpoint to allow more seamless transitioning in modifications to the endpoint? Because as of now working with endpoints I'm having trouble seeing in what instances redeploying the spec doesn't break the backend service and vice-versa.
I realized that when a new spec is generated, there is an ENDPOINTS_SERVICE_VERSION that is generated which seems to be persistent. So the act of deploying a new spec does not cause a breaking change in the API endpoint AFAIK as the backend service will still be pointing to a valid previous deployment.
I don't know how many ENDPOINTS_SERVICE_VERSIONs are persistent as I could not find this in the docs but if the backend service is updated with the latest endpoint service version soon afterwards, updates to the api can be seamless without disruption of the service as far as deployments go.
Related
I have four services running within the same app on App Engine. I have a frontend SvelteKit application, and three backend services. If possible, I'd like to set up security in such a way that the backend services will only accept HTTP requests from the frontend application (which sends all API requests via its Node server).
Is there a way of doing this without spending a load of money on a Serverless VPC Access connector?
Ideally I want to keep these all within the same GCP project as well. So far the only solution I can come up with is to ship the services with a secret that they check against when receiving a request, but there must be a better way to do it.
Take a look at Identity Aware Proxy
Pay attention to the part of the above documentation that says
In order to make a resource publicly-accessible (while sibling resources are restricted), grant the IAP-secured Web App User role to allUsers or allAuthenticatedUsers.
Per your use case, your front-end application will be available to the public while your 3 backend services will only be available to the front-end application
Since your backend services are now secured (via IAP), you have to programmatically invoke them in your front end. See documentation on how to do that.
My Django app is working on GAE with Cloud SQL（MySQL）.
By using New Relic Monitoring, I tracked requests done by my app to www.googleapis.com.
I don't know those requests. Can somebody explain this behavior?
(app connecting with cloud proxy)
As merely described here, there are various processes running in the instance along side (with?) your application... scaling, billing, logging, etc. Communication with other Google Cloud Services (like the ones I previously mentioned) is mainly done through Google Cloud APIs. Since App Engine is a managed Platform as a Service, this management needs to be done somehow... a myriad of REST API requests and responses do this. Partly through the www.googleapis.com endpoint.
You don't have to worry, though... your application's performance is not affected by this, nor your billing account.
We need to migrate from one app engine project to another (due to the constraints put in place for changing region).
The ideal solution would just be to proxy all requests through to the new server however we are using Google Cloud Endpoints which are intercepted by the server and delivered as POST requests.
We can't redirect as we have mobile apps relying on the API.
Does anyone have a solution (rather than proxying every API method we have) to proxy to a new server?
I would write a ServletFilter on the old app that intercepts /_ah/spi/* and forwards it to the new app, also on /_ah/spi/*. Keep in mind that you'll have to keep the existing Endpoints code in place, or the proxy will delete your configuration and not forward anything.
I 'm building an AppEngine application that stores data in Google Cloud Storage. I use the Google Cloud Storage Client (GCS) library as suggested.
My app is working when deployed on AppEngine (reading/writing/listing objects) but I cannot make it work on the development server. The development server keeps returning error 404 and GCS raises NotFoundError. The dev-appserver is supposed to emulate the cloud storage functionality without any specific configurations etc. I see in the log files that the dev server is accepting requests at "/_ah/gcs" yet it seems that there is no handler for that url. I 've tried with version 1.8.5 and 1.8.6. Apart from my app, not even the demo app provided by Google works.
Is there something that I 'm missing here, e.g. a special configuration for the dev-appserver?
Sorry the following change was pushed out too early by mistake. It only works with 1.8.8 SDK. We are streamlining the release process of gcs client to align with SDK. Sorry
https://code.google.com/p/appengine-gcs-client/source/detail?r=125
Without this change, it works on 1.8.7 SDK.
I'm developing a Java app in GAE, which offers an API through Google Cloud Endpoints.
Basically it receives requests in the endpoints and uses a number of web services from different providers, stores some data and returns some data through the endpoints...
I understand that my app is conceptually a backend, because it doesn't provide any web page, but only the endpoints, don't you think so?
But there's no way to create only a backend, without being associated to any frontend app, is there? At least Google Plugin for Eclipse only allow you to "Generate App Engine Backend", from an existing app, and moreover this app must be an Android project...
I'm using it as a frontend and there's no problem, but apart from the conceptual issue, I've read that backends are kind of optimized to be backends, with more memory and CPU...
I think you're just confused because the Cloud Endpoints documentation uses the word 'backend' to refer to the entire cloud-hosted server implementation. It doesn't specifically refer to the use of GAE backend instances. Endpoint requests can be served by frontend or backend instances, based on how you set them up and the url being accessed.
From the App Ending docs:
"When an application is called to serve a web request, it must issue a response within 60 seconds"
"App Engine Backends are instances of your application that are exempt from request deadlines and have access to more memory (up to 1GB) and CPU (up to 4.8GHz) than normal instances."
So unless you're requests are doing something crazy, you don't need to use a backend. In the google-plugin-for-eclipse, "generate appengine backend" is talking about creating a backend for your android app... a server for your android app to contact (in this case your android app is the frontend and you're appengine app is the backend). In the example app you can remove the web side (index.html) to the appengine application and you'll have no web frontend. Index.html is using the gapi javascript library to make endpoints calls to your appengine service.