I'm using the Review Apps feature integrated with Github on Heroku. In one of my apps, I set an environment variable called HOST_NAME . For example, if the site is http://www.purplebinder.com, then HOST_NAME would be set to www.purplebinder.com. It's used in a couple of places where we work with cookies and in our transactional emails.
When I open up a new pull request and spin up a review app, HOST_NAME should be something like purplebinder-pr-27.herokuapp.com.
Is there a way to set this value automatically? The Heroku documentation on review apps says an env var can inherit a value from the parent app or be hardcoded in app.json. Neither of those approaches work here, because the value needs to be different each time, and also different from the parent app.
Heroku also says an env var can be set "through a generator", but doesn't go into detail about what that is.
This question might be a duplicate of Setting ROOT_URL for Review Apps, but nobody answered that one. It's also similar to How to get Heroku app name from inside the app, but the answers there involved running a script after the app was created - here I'd like to set this value as part of the initial build.
From https://devcenter.heroku.com/articles/github-integration-review-apps#heroku_app_name-and-heroku_parent_app_name:
To help with scripting, two special config vars are available to
review apps. If you specify HEROKU_APP_NAME or HEROKU_PARENT_APP_NAME
as required or optional config vars in your app.json file, Heroku will
set those config vars to the new application name and the parent
application name respectively. They will then be available for use in
the postdeploy script so that you can do more advanced bootstrapping
and configuration.
Here is an example app.json file that uses
HEROKU_APP_NAME and HEROKU_PARENT_APP_NAME:
{
"name":"Advanced App",
"scripts": {
"postdeploy": "rake db:setup && bin/bootstrap"
},
"env": {
"HEROKU_APP_NAME": {
"required": true
},
"HEROKU_PARENT_APP_NAME": {
"required": true
}
}
}
If you add the heroku-buildpack-cli to your parent app, then it enables you to set environment variables from your post-deploy script. The command should look something like the following:
heroku config:set HOST_NAME=${HEROKU_APP_NAME}.herokuapp.com --app ${HEROKU_APP_NAME}
Here's an approach ignoring app.json for Rails installations:
in the relative config/<environment>.rb. I personally use production.rb and staging just references it.
if ENV.fetch("HEROKU_APP_NAME", "").include?("staging-pr-")
ENV["APPLICATION_HOST"] = ENV["HEROKU_APP_NAME"] + ".herokuapp.com"
ENV["ASSET_HOST"] = "https://" + ENV["APPLICATION_HOST"]
config.action_mailer.default_url_options = { host: ENV.fetch("APPLICATION_HOST") }
end
...
It's a bit misleading as the heroku environment variables will still have the old variables, but it works.
You can also create review environment for you application copying staging.rb or production.rb from config/environments. This would be useful.
After adding HEROKU_APP_NAME and HEROKU_PARENT_APP_NAME to your app.json, you can easily set;
config.action_mailer.default_url_options = { host: "#{ENV['HEROKU_APP_NAME']}.herokuapp.com" }
config.action_mailer.asset_host = "http://#{ENV['HEROKU_APP_NAME']}.herokuapp.com"
config.action_controller.asset_host = "#{ENV['HEROKU_APP_NAME']}.herokuapp.com"
config.action_cable.url = "wss://#{ENV['HEROKU_APP_NAME']}.herokuapp.com/cable"
Related
In my Heroku environment where I deploy my NestJS application, I set all my Config Vars through the Heroku UI. I have set my NODE_ENV config var to staging using the Heroku UI. I've even ran this command heroku config:set NODE_ENV=staging -a <my-staging-environment. When I run heroku config -a <my-staging-environment>, I see that NODE_ENV is set to staging, but for whatever reason, when I console.log this variable from my code, it outputs development. Below is some example code where it is logs development as the value for NODE_ENV in my TypeOrm configuration. This is causing the ssl property to not get set to the correct value, and I cannot connect to my Heroku Postgres database because of it. I only intend to set it to development for local development purposes.
require('dotenv').config();
console.log(process.env.NODE_ENV); // outputs "development" - idk where this value is coming from
console.log(process.env.DATABASE_URL); // outputs the correct value that I set in Heroku Config Vars
const typeOrmConfig: TypeOrmModuleOptions = {
type: 'postgres',
url: process.env.DATABASE_URL,
ssl: process.env.NODE_ENV !== 'development' ? { rejectUnauthorized: false } : false, // ternary evaluates to the wrong value
// ... other config options
};
I use dotenv, and I made sure to .gitignore my .env file. I don't recall ever setting this variable to development. The only place I see it set to NODE_ENV=development in my code is in my .env-example file. I do commit this file to source control, but it's just an example file and not a real .env file and it shouldn't actually be being used.
Does anyone have any idea why this is happening?
I have found that it is not a Heroku issue. Rather it is a NestJS/Nx issue.
See:
process.env.NODE_ENV always 'development' when building nestjs app with nrwl nx
My Problem
I am having trouble loading my environment variables on Heroku production.
When pushing to Heroku I get following error message during the build script:
Failed to load ./.env.
Current Setup
I am using a .env file in the root of my app locally. I can succesfully load my environment variables using the dotenv-webpack plugin as follows:
//webpack.config
const Dotenv = require('dotenv-webpack')
module.exports = {
// other settings...
plugins: [
new Dotenv(),
]
};
Loading the environment variables:
//server.js
require('dotenv').config();
console.log(process.env.MY_VARIABLE);
This works like a charm locally, but fails on Heroku.
Note: My config vars have been set on Heroku, so that's not the problem.
What I tried
I have already tried to force load the .env file from the root of my app like this:
new Dotenv({ path: path.resolve(__dirname, './.env') });
Someone also pointed out that the Heroku environment might be system wide environment variables so I tried to load them using:
new Dotenv({ systemVars: true });
Neither of these attempts worked for me.
My guess
I have noticed that Heroku saves their .env file under ./tmp/build_someRandomBuildId/.env. My guess is that the .env file is not on the root of the directory, hence why dotenv can't find it. There is also no way to hardcode the location of this file in my Webpack configuration as the build ID is randomized with every build. Is there a way to tell Webpack to look for the file in a dynamic location?
Today i stumbled upon this problem, i tried several solutions but none worked. My App was working locally but in production mode (heroku) it was not loading process.env correctly.
then i found this https://www.npmjs.com/package/dotenv-webpack
//webpack.config.js
plugins: [
new Dotenv({ systemvars: true }),
],
Just setting systemvars to true does the trick..
For now I have tested this using different keys for the .env file and the heroku dashboard; They are not connected, and they replace themself correctly in production or dev mode.
Use the package "dotenv-webpack" instead of "dotenv".
I hope this saves some time to anyone facing the same problem
I finally found the solution, leaving this here for others who have the same problem as I did.
I used dotenv-webpack to set my environment variables locally, which worked like a charm. Heroku on the other hand sets their environment variables automatically, so there is no need to set them yourself. There is no need to look for a .env file. All I had to do was split up my webpack.config in 2 separate files.
//webpack.dev
require('dotenv').config();
plugins: [
new Dotenv()
],
Load .env file locally.
//webpack.prod
require('dotenv').config();
plugins: [
new webpack.DefinePlugin({
'process.env': {
'YOUR_VARIABLE': JSON.stringify(process.env.YOUR_VARIABLE),
}
});
]
Get your environment variables from Heroku and write them to your own process.env
If you are not using Webpacks, the idea's solution is similar to the accepted answer.
Heroku works on "production" mode by default, so if the problem in Heroku is with Dotenv (which should not be used anyways in Heroku), disable the use of Dotenv in production time like this:
if (process.env.NODE_ENV !== 'production') {
require('dotenv').config() }
}
...and then, access env variables just by doing:
var someVar = process.env.SOME_VARIABLE;
Don't forget to set the environment variables on Heroku first by using Console Commandline in your app's dashboard, or with an app.json file.
I am doing some automatic testing on Circleci, with different enviromental variables: I need one port for my local testing and a different one for Circleci.
How can I make Cypress do that? I tried making cypress.env.circle, but that does not seem to work
The cypress docs explain 5 ways to set variables.
To use one port locally and one on CircleCI I would:
Add a default port to cypress.json under the env section for local use so you don't have to think about it, and anyone else contributing will have a working version.
Set an environment variable in CircleCI named cypress_VAR_NAME which will override default in cypress.json
cypress.json example
{
"env": {
"the_port": 5000
}
}
CircleCI variable would then be cypress_the_port and you would read it in your specs as parseInt(Cypress.env('the_port')) (assuming your spec needs an integer for port)
I was following this guide on deploying to Heroku and this one for sending email.
Everything works fine in development. My variables are set in Heroku:
heroku config
...
MAILGUN_DOMAIN: https://api.mailgun.net/v3/xxxxxx.mailgun.org
MAILGUN_KEY: key-3-xxxxxx
...
And loaded from the config files like so:
config :take_two, Mailer,
domain: System.get_env("MAILGUN_DOMAIN"),
key: System.get_env("MAILGUN_KEY")
However when I try to send email on Heroku when the Mailgun config is set from environment variables I get this error:
** (FunctionClauseError) no function clause matching in IO.chardata_to_string/1
(elixir) lib/io.ex:346: IO.chardata_to_string(nil)
(elixir) lib/path.ex:467: Path.join/2
(elixir) lib/path.ex:449: Path.join/1
lib/client.ex:44: Mailgun.Client.send_without_attachments/2
This happens when the domain is not set for the Mailgun Client. But it is supposed to be set from the environment variable. I made a simple module to test:
defmodule TakeTwo.Mailer do
require Logger
use Mailgun.Client,
Application.get_env(:take_two, Mailer)
def blank_shot do
Logger.info Application.get_env(:take_two, Mailer)[:domain]
Logger.info Application.get_env(:take_two, Mailer)[:key]
send_email from: "steve#xxx.com", to: "speggy#xxx.com", subject: "Hello", text: "This is a blank shot"
end
When I run TakeTwo.Mailer.blank_shot I see the correct domain/key variables logged followed by the error. I am not sure how to debug the Mailgun client remotely.
Finally, if I recreate the above module in the shell (after running heroku run iex -S mix) it works just fine!?
I feel like when the original module is being loaded perhaps the environment variables have yet to be loaded??
The answer was a little buried in a comment so I wanted to make it easier to find. As the other answer mentions, the environment variables aren't available, but the buildpack lets you configure them to be:
I created a elixir_buildpack.config file and added the following:
config_vars_to_export=(DATABASE_URL MAILGUN_DOMAIN MAILGUN_KEY SECRET_KEY_BASE)
The environment variables aren't available at build time. I had the same issue and decided to get rid of the macro carrying the configuration. You can use this patch to move on.
When an integrated service, like CodeShip CI, runs tests, an instance is spun-up on Heroku to run the CI suite.
How can I get the name of that branch/build-specific Heroku app, in a programmatic manner?
My use case: I want to give a developer heroku-cli access to the staging instance that was spun-up for their branch. Also, I want the instance URL so that QA can check it for accuracy.
I don't know if this is principally a CodeShip question, or a Heroku question. I can solve the rest of the integration, if I simply can get the name/info for this new instance.
Codeship defines certain environment variables with each run.
Thus I can build a URL based on all of the known pieces, per env:
app_name = 'foo'
env = 'staging'
name_parts = [
app_name,
env,
'pr'
ENV['CI_BUILD_NUMBER']
]
testing_url = 'http://' + name_parts.join('-') + '.herokuapp.com'
#=> foo-staging-pr-2729.herokuapp.com
send_to_chatroom(app_name + '_dev_notification')