I've popped into console and ran the "heroku addons:add mongohq:free" which added an environment variable: ENV['MONGOHQ_URL']. It is a url that looks like this:
mongodb://heroku:xxxxxxxxxxxxxxxxxxxxxxxxx#staff.mongohq.com:10059/appxxxxxx
I want to be able to use (and share) TWO mongodb instances. Running addons:add mongohq:free on the same app fails saying that the addon is already present.
Do I have to create a separate, blank app and just use the ENV variable from it?
Finally, is it possible to share these urls between apps?
To have multiple Mongo instances, you can either:
* Create another account on MongoHQ or another provider
* Create your own server and install mongo on it with authentication and multiple DBs
You then would need to add environmental variables for as many mongo DB's as you want.
Then, your application would need to know which MongoDB to connect to, under whichever circumstances are applicable in your app.
Note: you can replace Mongo with Postgres or MySQL above and its the exact same procedure.
And yes, you can share these between apps -- simple set the environment variable to match and you're good to go.
Related
I am working on an app where I now need to add test mode or sandbox mode features like Stripe provides in the dashboard. In this feature, when the user turned the test mode on, the user can test the functionality and can create dummy data in the same login session/token.
I have tried to use the Rails 6 feature to use multiple databases but have a few questions:
is it good to switch the connection in production for test mode even we will not have many test requests?
will it be good to have a separate instance for test mode with the test subdomain? in this case how we should manage the login sessions? should we copy data to the test the database? will it be good and common practice?
I only can manage to implement this if I have user data in the test database so that when I switch database connection system will not send an unauthenticated response.
Note:
for login and user update, it will use the primary database all time and for other actions, it will use the test_mode database. I am doing this by skip_around_action in a specific controller.
We are doing this so that when the user turned off test mode, it will update the primary database and the next request will use the primary database as per around_action logic
Here is my current code in application_controller.rb:
around_action :setup_db_connection
def setup_db_connection
database_key = (user.test_mode?) ? :test_mode : :primary
ActiveRecord::Base.connected_to(database: database_key) do
yield
end
end
test_mode database key has test database configuration in database.yml and similarly for primary database key. Both are completely two different databases.
Can anyone please tell me if I am going in to correct direction? Any help or thought will be appriciated. Thank you in advance!!!!
You should just create an additional environment as this is the exact problem that they are designed to solve.
Rails just ships preconfigured with development, test and production as thats the bare minimum you can get by with. You can actually have an unlimeted number of environments.
So lets say you want to create an enviroment named staging which closely matches your production environment. You can start by copying the productions settings:
cp app/environments/production.rb app/environments/staging.rb
And then just setup an additional hash in your database.yml and credentials files.
staging:
<<: *default
database: my_app_staging
You can then set the environment by using the RAILS_ENV env var or the command line arguments when starting the Rails server. Your staging environment can be run for example in a separate Heroku application made available on a subdomain. If you're using Heroku or some other SAAS platform a hobby tier will often be sufficient for staging.
When it comes to data you can either work on a set of dummy data generated by wiping and seeding or regularily wipe the staging database with a backup from your production database.
I would not consider using a "test mode switch" a good idea as it makes far to easy to inadvertantly mess things up. Using a separate environment lets you use a completely different set of credentials as well so you won't accentially do something like create a real credit card charge or destroy data on third party services. Sandboxes should be isolated.
I have a multi-domain Rails 4 app where the request.domain of the http request determines which functionality I expose a given visitor to.
Each domain in my app should be served by its own MongoDB database. E.g. domain1.com is served by db_for_domain_1, etc.
I can read in the MongoDB docs on runtime persistence that
Mongoid.override_database("db_for_#{request.domain}")
enables me to switch database on the fly.
But how do I keep the persistence when I bypass Mongoid and use the mongo Shell method db.collection.insert()? I will still do it from within my application though.
The answer might be in the MongoDB docs on collection access, but I don't understand it. So how do I switch database before/during this operation?:
MyModel.collection.insert({field_1: "Value 1", field_2: "Value 2"})
If I understand your question correctly: you have an app that connects to different mongodbs on different servers, but want to use mongo shell to connect to the database outside of your application? If true, you would connect to the desired database through the shell with
mongo db_for_domain_1:<port>/<dbName>
and then
db.<collectionName>.insert({doc})
see mongo --help for username and password options.
I recently had a short term project served off of Heroku that approached 10,000 Postgres records and needed to be upgraded from hobby-dev to hobby-basic. Being new to Heroku, I did my best in finding out how to do so and followed heroku's upgrade using PG Backup guide. With a bit of trial and error, I thought I got it working. I erred on the side of caution and didn't do the last deletion of the old DB step. However, I soon got a message from Heroku telling me I've passed the 10,000 record limit. I went to my Heroku dashboard and checked out the DB size. It doesn't look like the Hobby-Basic DB had anything stored in it. I waited a couple weeks and checked again, thinking it may have been a realtime updating issue. Still no tables in the new Hobby-Basic DB.
So my question is, is there a way for me to find out which DB my Heroku app is set up to use currently? How do I force it to switch to a different DB in my app?
This is the guide I followed to upgrade from Hobby-Dev to Basic: https://devcenter.heroku.com/articles/upgrade-heroku-postgres-with-pgbackups
Presumably your app is reading from the environment variable DATABASE_URL. If you have multiple database then you should also have other environment variables like HEROKU_POSTGRESQL_(color). You can access those environment variables from within your app to inspect the configured database information, and you can change them in the Heroku panel. If you change DATABASE_URL to one of the values of HEROKU_POSTGRESQL_(color), then your app should pick up on that and switch to the different database. You can also edit your configuration and access another environment variable while bypassing DATABASE_URL entirely.
In my rails 4 application directory, I typed "heroku pg:credentials DATABASE" into terminal to get all the information about the database for my application which is deployed on heroku. Since I'd like to view the data inside my postgresql database, I tried inputting the information into Induction, but it ended up not responding and I was forced to enter the activity monitor and force quit. I followed the same procedure several more times all with the same result. Is my version of Induction faulty? Should I be using a different program to view my database? Or am I doing something wrong?
I'm new to rails, so thanks for your help!
Yes, you can. Heroku Postgres allows connections from outside Heroku's network. The credentials displayed from heroku pg:credentials will be all you need and should work. Maybe try using the official PGAdmin tool instead?
I have on Heroku 2 apps - the first is in written in RoR and contains also the database. The second app is written in Sinatra framework and in this app I would need to work with data, which are saved in the database of the RoR app.
I've tried to look over some examples of how to load database data in Sinatra from the database of other app, but I wasn't very successful...
Could you give me please any advices, how to do that?
My best piece of advice would be to create an API on your rails app that your sinatra app consumes (you can use something like fernet for signed auth tokens). Don't have more than one application connect to the same database.
If all you want to do is share the database, you can do so by copying the DATABASE_URL config variable from one of the applications to the other.
You can get the value like this
heroku config --app <app_name_of_app1>
Once to have the value of DATABASE_URL, you can set it for the other app like this.
heroku config:add DATABASE_URL=<long_url_string_from_step_one> --app <app_name_of_app2>