Release v0.3.1

Updated porter run to spawn a container selection option when there are multiple containers (#739)

Release v0.3.0

Jobs and Cron Jobs Updates

About a month ago, we released native support for jobs and cron jobs. After a bunch of feedback and testing from our users, we added several features to improve the experience of running jobs on Porter. This includes toggling between jobs and cron jobs, setting concurrency limits for jobs, stopping currently running jobs, and handling graceful shutdown. To read more, please see the advanced configuration options for jobs and cron jobs.

RabbitMQ and LogDNA Add-Ons

RabbitMQ and LogDNA support -- our two most-requested add-ons -- have now been added to the Community Templates!

Errors for Failing Deployments

We've now added much better error handling to the launch screen, which will send more detailed errors if your deployment fails. We've also added event streams for failing containers, which will inform you of why a certain container is failing (which potentially prevents you from viewing the logs of that container). For example:

Normal     BackOff       Back-off pulling image "not-an-image:latest"
Warning    Failed        Error: ImagePullBackOff

Template Upgrades

All templates will now display a template version and template upgrades are supported from the dashboard for application templates only (add-on upgrades will be supported in a later release). Read more about template upgrades here.

Porter Charts Repository

We've created the repository porter-charts for contributors who wish to add new templates or modify existing templates that are displayed on Porter. If you'd like to contribute, please check out the contributing guide!

New Guides/Documentation

• Preserving Client IP Addresses for Custom Domains
• Setting Up Metrics Visualization
• Wildcard Domains on DigitalOcean
• Deploying Django on Porter
• Advanced Configuration of Jobs and Cron Jobs
• Template Versioning and Upgrades

Minor Improvements

• Split the launch flow into separate screens so there's less information to look at
• Handle problematic parsing of ANSI control codes from porter run commands, particularly in interactive shells
• Migrate public sidecar + helper images to ECR to avoid Dockerhub rate-limiting, particularly from EKS clusters where there is a single outbound IP address
• Add documentation to Github repo so that contributors can create PRs for docs
• Fix issue where service with the same name as a service in a different namespace will show pod statuses/logs from both services

Release v0.2.1

Fixes #616 -- certain interactive shells will no longer output ANSI control sequences.

Release v0.2.0

Secret Environment Variables

We've added support for creating secret environment variables from environment groups that will not be exposed after creation. After these secrets are created, they are stored in your own cluster and are mounted in to your applications from your cluster, so that they never get exposed to the Porter dashboard again. Read more about secret environment variables.

.env File Upload

We had a lot of users complain about inputting all of their environment variables manually, so we added the option to copy and paste a .env file to upload your environment variables. This should make it much easier to upload your environment variables in the future. When creating an environment group, you should click the "Copy from File" option, which will pull up a text area for you:

Locally Runnable Version

For users that just want to try out Porter without the hosted dashboard, we added docs for Running Porter Locally. After downloading the CLI, you can run porter server start to get the dashboard up and running.

Contributing Guide

We've had a number of users ask about contributing, so we added a contributing guide that should help you get started. All contributors welcome!

Cloud Provider Support

Google CloudSQL Proxy

We've had a number of users require a proxied connection to their Google CloudSQL instance, which we generally recommend to Google users for a fully-managed DB offering. Here are the docs.

AWS NLB

EKS clusters provisioned through Porter used to create a Classic Load Balancer to connect to an NGINX instance running on the cluster, which we have now migrated to an NLB to give users more flexibility.

Bonus Feature: ANSI Color Parsing

To make structured logs easier to view, we now parse ANSI escape codes and display their colors on the dashboard.

Minor Improvements

• Improve zero-downtime deployments by adding termination grace periods to deployments
• Update our start commands and added termination grace periods so that
• Updated Cache-Control headers so that you get the latest version of the Porter dashboard every time we update.
• Fix a bug where only completed jobs were listed after clicking on the "Jobs" tab.
• Hide the default *.porter.run domains in the template after creation, so that users do not try to edit them.
• Fix custom ingress annotations so that users can customize their NGINX configuration options.
• Update documentation around CI/CD, jobs, and metrics integrations

Release v0.1.0-beta.3.6

Jobs and Cron Jobs

We've added support for jobs and cron jobs (scheduled jobs) to the Porter dashboard. This makes it much easier to run scripts that are designed to run to completion, such as migration scripts, batch processing, scheduled notifications, etc. Read more about jobs.

Environment Groups

We've made it much easier to manage environment variables for multiple services through the use of environment groups. These allow you to create a shared configuration of environment variables that you can reuse across services. Read more about using environment groups.

Remote Execution

Porter now supports remote execution via porter run, which will execute in the same manner as heroku run. For example:

$ porter run web-test -- echo "HELLO WORLD!"
Running echo HELLO WORLD for release web-test
HELLO WORLD

View the documentation here.

NGINX Error Metrics

We now display error metrics for 5XX errors coming from NGINX. To enable this, Prometheus must be installed in the cluster (go to Launch > Community Add-ons > Prometheus), and then go to the ingress-nginx namespace and click on the nginx-ingress deployment. Reach out on Discord with any questions!

EKS Machine Type Selection

Porter now lets you choose different machine types for your EKS clusters.

Minor Improvements

• Ability to provision multiple clusters in the same project.
• Ability to specify start commands in deployments, and improved support for Procfiles and auto-build. If you have multiple processes specified in your Procfile, you can now choose the process by inputting the process name as the start command.
• Ability to delete of failed pods in releases.
• Ability to toggle deployment webhooks on and off
• Tighter integration between deployments and Github actions:
  • Inject --build-args during the docker build step in Github actions
  • Build-time environment variables are updated during deployment upgrade
  • Github actions files and environment variables are deleted during deployment deletion
• Custom Porter error page when awaiting deployments.

Release v0.1.0-beta.3.5

Merge pull request #334 from porter-dev/fix-cli-logout

CLI fixes

Release v0.1.0-beta.3.4

Merge branch 'beta.3.key-rotation'

mergin

Release v0.1.0-beta.3.3

Merge pull request #271 from porter-dev/beta.3.github-actions-ci

Beta.3.GitHub actions ci

Release v0.1.0-beta.3.2

Merge branch 'master' of https://github.com/porter-dev/porter into main

Release v0.1.0-beta.3.1

Fixes #199