This guide covers the migration of an application that was developed for Heroku to the Datica Platform.
This guide assumes that you already have an application set up and working - for examples, this will use the application described in the Node+Mongo guide.
The other prerequisite is that you’ve already talked with our Sales team and worked with Datica to get an environment provisioned and ready for you, including all of the databases and caches that your application needs. To get that process started, send us a message at email@example.com.
Terminology, Similarities, and Differences
Many concepts translate between The Platform and Heroku. Some are a little different, and some have a different name.
Your code is built using buildpacks, the same way as on Heroku.
To deploy on The Platform, you push using git.
Heroku Application = The Platform Environment
On Heroku, you have one repository with one or more Procfile targets. On The Platform, you have one or more Code Services, each instance of which is equivalent to a Dyno on Heroku. What this means is that working with multiple codebases that use different buildpacks is easier, and only one Code Service needs to be updated at a time.
The Platform supports workers the same way Heroku does - they’re invoked out of a Procfile target of your choosing. Because there can be more than one code service, however, you choose which code service to launch the worker process out of.
- Heroku Addon = The Platform Service
On The Platform, your databases and caches are dedicated instances called database services and cache services, running in their own containers within your environment, in the configuration of your choice (either highly-available or single-instance).
- Heroku Config Variables = The Platform Environment Variables
Same concept, different name. Environment Variables can vary between code services.
- Heroku Organization Account = The Platform Organization
On The Platform, each environment belongs to an Organization. When a Datica environment is provisioned, an Organization is also created to which your new environment is assigned.
1. Gain Access to your Environment
When your first environment is provisioned, you’ll receive two emails from us. The first will be from a member of our support team letting you know that your environment is ready, along with some basic information about your new environment and a link to our Getting Started guides.
The second will be an invitation to join your newly-created organization. Follow the instructions in that email, and register a new account if you don’t already have one.
After you’ve joined your organization, navigate to The Platform dashboard. You should see your environment listed there.
Take note of its name - that’s “My Shiny New Environment” in this case. Also take note of the name of your code service - that’s “code-1” above.
2. Install the Datica CLI
The CLI can be found on Github: https://github.com/daticahealth/cli
Follow the instructions on the readme to install it. You’ll know it’s installed when you can open a terminal and run
3. Initialize Your Local Repo
To link up your local repo to your Platform environment, open a terminal to its root directory and run the
datica init command (it will sign you in).
$ cd my-repo $ datica init Username or Email: firstname.lastname@example.org Password: # Will prompt you to pick an environment and code service you have access to "datica" remote added.
The email and password here are the same credentials that you used to sign in to the Platform dashboard.
4. Set Variables
Checking variables on Heroku:
$ heroku config === my-heroku-app Config Vars MONGODB_URI: mongodb://username123:email@example.com:23624/username123
Checking variables on The Platform:
$ datica -E "<your_env_name>" vars list <service_name> DATABASE_URL=mongodb://mongodb-12345.internal:27017/catalyze MONGO01_URL=mongodb://mongodb-12345.internal:27017/catalyze
As you can see, the connection string for the database is provided in a similar way, but the variable names don’t match. You have two choices, here - either change your code to use a different name, or add a new variable to your code service to match its name from Heroku. To do the latter:
$ datica -E "<your_env_name>" vars set <service_name> -v MONGODB_URI=mongodb://mongodb-12345.internal:27017/catalyze Set. For these environment variables to take effect, you will need to redeploy your service with "datica redeploy" $ datica -E "<your_env_name>" vars list <service_name> DATABASE_URL=mongodb://mongodb-12345.internal:27017/catalyze MONGO01_URL=mongodb://mongodb-12345.internal:27017/catalyze MONGODB_URI=mongodb://mongodb-12345.internal:27017/catalyze
Note that warning. Environment variables are only picked up when a service is deployed or redeployed - any changes you make here require a fresh container to take effect.
5. Add an SSH Public Key
The Platform uses public key auth to authenticate git pushes. To add a public key to your Datica account (you only need to do this once):
$ datica -E "<your_env_name>" keys add my-ssh-key ~/.ssh/my-ssh-key.pub
For more details on SSH key usage and debugging, take a look at our SSH Keys Guide.
6. Add a Site (and Cert)
So that The Platform knows what hostnames to accept traffic on and what services to route that traffic to, we need to set up a Site.
If you’re migrating, you’d probably like to see your code working before cutting over fully to The Platform. You can use a different domain or subdomain for now (
staging.example.com, for example) - another site can always be added later.
Adding an SSL Cert
To add an SSL Cert you already have on hand, use the
certs create command in the CLI:
$ datica -E "<your_env_name>" certs create my-ssl-cert ./my-cert.crt ./my-cert.key
If you don’t have a signed cert for the subdomain you intend to use, you can use a self-signed cert for testing.
For more help on adding and debugging certs, take a look at our Self-Service SSL Guide.
Adding a Site
sites create command in the CLI:
$ datica -E "<your_env_name>" sites create staging.example.com code-1 my-ssl-cert Created 'staging.example.com'
Update Your Domain’s DNS
You will need to add a CNAME record from the hostname you used above to the Datica public hostname for your environment. To find that hostname, use the
sites list command:
$ datica -E "<your_env_name>" sites list NAME CERT UPSTREAM SERVICE pod0012345.catalyzeapps.com pod0012345.catalyzeapps.com staging.example.com my-ssl-cert code-1
The public hostname is the
*.catalyzeapps.com URL -
pod0012345.catalyzeapps.com in this case.
For more information on setting your DNS correctly, take a look at our Setting Up DNS Guide.
Restart Your Environment’s Service Proxy
Each environment has a special service called the Service Proxy. The Service Proxy is responsible for routing traffic from the outside world into your environment. In order for it to pick up the new site you just added, the service proxy needs to be reloaded, and redeploying it is the way to do that.
$ datica -E "<your_env_name>" redeploy service_proxy
7. Push Code
Just like on Heroku, push to datica master:
$ git push datica master Counting objects: 4, done. Delta compression using up to 4 threads. Compressing objects: 100% (4/4), done. Writing objects: 100% (4/4), 1012 bytes | 0 bytes/s, done. Total 4 (delta 0), reused 0 (delta 0) remote: remote: Prepping Build................. remote: [buildstep] Building application -----> Node.js app detected ... ... (truncating output for this guide) ... remote: [buildstep] Build finished remote: Finalizing Build............. remote: Redeploying Service.... remote: Complete. Built Successfully! To ssh://firstname.lastname@example.org:2222/pod0012345-code-123456789.git * [new branch] master -> master
8. Check out your running app
Navigate to the hostname you configured above, with https (https://staging.example.com/ for this guide).
Note that it may take a few minutes for your application to show up - several factors can cause delay in this (DNS propagation, AWS load balancers taking effect, or slow-starting applications are the most common causes).
If you think you’ve waited too long and your application still hasn’t started, check your logs. Your logs can be accessed via the
logs command in your CLI, or at
https://your-env's-public-hostname/logging/ - there is also a link to view logs in the Platform dashboard.
A topic this guide does not cover is database schema setup and data migration - for more on this, take a look at our Database Import Guide.
If your application is still having issues, or you have questions about any part of this process, please contact Datica Support.