Google Cloud Platform
Deploying a Python serverless cloud function in minutes with GCP, Google Cloud Community
It is important to note that the always free tier is not set up as “1 instance” it is set up as a number of instance hours (or instance months, or ip-address-months). So, you could run 3 instances concurrently for a third of a month as well (as long as they each only had no more than a 30GB disk attached each — the limit of total disk usage is 30GB-months). [source]
Google Cloud Tasks, asynchronous task execution Followup item
Getting the Best out of Google Cloud Shell
Google Cloud Repository docs | Google Developers Codelabs |
Google Cloud Functions without authentication enabled can be run from any IP address. Securing Google Cloud Functions
Project quota info | Creating and Managing Projects | Google Cloud resource hierarchy in detail: At the lowest level, resources are the fundamental components that make up all Google Cloud services. Examples of resources include Compute Engine Virtual Machines (VMs), Pub/Sub topics, Cloud Storage buckets, App Engine instances. All these lower level resources can only be parented by projects, which represent the first grouping mechanism of the Google Cloud resource hierarchy. [Google Cloud Resource Hierarchy]
Cloud Platform docs overview | Cloud Key Management Service docs | Php encryption and Google cloud key management service |
Cloud Scheduler pricing: “Each Google billing account gets 3 jobs per month free. Note that the free tier is measured at the account level not the project level.” Cloud Pub/Sub and Cloud Functions are included in the Always Free tier. So there should be no cost to automatically start and stop Compute Engine instances on a regular schedule. So take advantage of this to save money by stopping and starting Compute Engine instances on schedule. reddit search on cloud scheduling
Build robust serverless architecture on Google Cloud Platform Cloud Security Scanner | Serving websites | SSH from the browser, gce | For SSH from the browser, firewall rules must allow connections from GCP Console services (74.125.0.0/16). See Disable external SSH on GCP while retaining the SSH web console | Also this |
Cloud Source Repositories, 50 gb on free tier
Hosting The Cloud Platform offers two hosting options: the App Engine, which is their Platform-as-a-Service and Compute Engine as an Infrastructure-as-a-Service. In the standard App Engine hosting environment, Google manages all of the components outside of your application code.
app.yaml config | default php.ini |
For (python only?) apps in appengine, this is necessary for a preview test
dev_appserver.py --php_executable_path=/usr/bin/php-cgi $PWD (not sure now, got an error last time I tried this after an app update; to deploy it I only needed to do: gcloud.app.deploy
Then launch Web Preview.
Php7.2 app errors:
+ “(gcloud.app.deploy) The [threadsafe] field is specified in file [/home/userid/search-arxiv/app.yaml]. This field is not supported with runtime [php72] and can safely be removed.”
+ “INVALID_ARGUMENT: The following fields are not allowed in app.yaml: api_version.”
+ “INVALID_ARGUMENT: script field for handler '/.*' must be set to 'auto' for runtime php72.”
When you deploy an app, you are prompted for a region.
File upload done.
Updating service [default]...done.
Setting traffic split for service [default]...done.
Deployed service [default] to [https://appname.appspot.com]
You can stream logs from the command line by running: $ gcloud app logs tail -s default
To view your application in the web browser run: $ gcloud app browse
Updating a deployed app and managing earlier versions | Git source repositories on GCP |
The Cloud Platform also offers managed VM environments that blend the auto-management of App Engine, with the flexibility of Compute Engine VMs.The managed VM environment also gives users the ability to add third-party frameworks and libraries to their applications. [source]
Cloud SQL is MariaDB 5.7; not part of Free Tier
Cloud Storage can host static websites but requires a custom domain. [source] | Host a Website for FREE on Google Cloud Storage – Cloudbooklet | Use Buckets to Serve Static Website | Cloud Storage docs | How to Set Up MySQL on Google Compute Engine |
Compute Engine: FAQ | Reserving static external IP |
AWS vs Azure vs Google Cloud Building a Web Site on Google Compute Engine
Set up 'always free' micro VPS on GCP | Best Use | Setup LAMP stack | GCP Essentials free tier | App Engine standard environment | Getting Started with App Engine (Python) | Connecting to instances | starting and stopping an instance |
GCP free tier | GAE python | GAE php |
Using the Compute Engine VM instance, from the shell, could access a web page served by a python http.server at the external IP of the instance.
From the shell icon, you can upload/download files.
What's next:
– Work through a tutorial to Set up a basic web server on your new instance.
– Work through a guided Getting Started tutorial in the Google Cloud Platform Console.
– Work through Managing GCP Projects with Terraform tutorial.
– Learn how to create and manage virtual machine instances using the gcloud command-line tool and API.
– Learn more about the Linux images available on Compute Engine.
– Learn how to Transfer your files to your instance.
Google Cloud Shell provides you command-line access to your cloud resources directly from your browser. You can conveniently manage your projects and resources without the need to install the Google Cloud SDK or other tools for that matter. With Cloud Shell, multiple utilities are always available, up to date and fully authenticated when you need them. You can now enjoy access to a wide range of pre-configured developer tools like Go, PHP, Java, Ruby, etc. Google Cloud Shell also allows you to run your web applications inside the Cloud Shell instance and preview them from the browser. You can commit it back to your repo with the preconfigured Git and Mercurial clients. Google Cloud Shell provides you the ability to do more than just “Google Cloud.” It also offers 5GB of persistent disk storage that is mounted as your $HOME directory on the Cloud Shell instance, giving you a free playground to experiment in. [source]
Compute Engine: sudo apt-get update You are running apt-get inside of Cloud Shell. Note that your Cloud Shell machine is ephemeral and no system-wide change will persist beyond session end. You can customize your environment to permanently include this package by updating your environment at https://cloud.google.com/console/cloudshell/environment/view. To suppress this warning, create an empty ~/.cloudshell/no-apt-get-warning file. The command will automatically proceed in 5 seconds or on any key. Visit https://cloud.google.com/shell/help for more information.
GCP: The Google Cloud Platform Compute Stack Explained. | What Is Cloud Computing? How Does Cloud Technology Work? | Google Cloud Platform: A cheat sheet |
The following ports need to be allowed for inbound traffic 443 – This will be used to receive the SSL protected HTTPS requests 80 – This will be used temporarily to recieve your SSL certificate 22 – This should be open already however if it isn’t, allow 22 traffic for SSH connections.