Connect your Warehouse

Which warehouse do you use?

• Redshift
• BigQuery
• Snowflake
• PostgreSQL (or Panoply)
• Azure / Synapse / Microsoft SQL Server
• Amazon Athena (Beta)
  
  

Narrator is committed to keeping your connections secure. Read more about our approach to Security.

📘

Don't see your warehouse on the list?

We're always looking to support new warehouses. Shoot us an email at [email protected] to see about partnering with us to set up a connection for your warehouse type.

How do you want to connect?

Narrator supports the following connection options

• Add Narrator to AWS's Allowlist
• Use an SSH tunnel

Adding Narrator to AWS's Allowlist (Redshift)

1. Log into your Redshift console

2. Navigate to clusters

If you don't see your cluster, check the region in the top right

4. Open Properties

7. Go back to the Narrator's warehouse page and click the 'Test and Save' button

Using an SSH Tunnel

Set up an SSH tunnel server that has access to your destination port. The tunnel server's SSH port needs to be accessible from Narrator's Static IPs.

Create an SSH User

Log in to your SSH tunnel host and run the following commands:

Create group narrator

sudo groupadd narrator

Create user narrator

sudo useradd -m -g narrator narrator

Switch to the narrator user

sudo su - narrator

Create the .ssh directory

mkdir ~/.ssh

Set permissions

chmod 700 ~/.ssh

Change to the .ssh directory

cd ~/.ssh

Create the authorized_keys file:

touch authorized_keys

Set permissions

chmod 600 authorized_keys

Using your favorite text editor, add the public SSH key from the destination setup page in your Narrator manage page to the authorized_keys file. The key must be all on one line. Make sure that you don’t introduce any line breaks when cutting and pasting.

ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIHQLZ/BQf5GDPRfXmHt9ecmChGjhOWZsZZOGsn82M1nJ

FAQ:

How does Narrator work on multiple warehouses?

We created a code based SQL abstraction that generates flavored SQL for the specific warehouse so all we actually do is just build a specific mapping and it is ready. We are planning to open source this soon but are waiting until we can go both ways flavored SQL to our abstraction then back. This would mean if you can make a Redshift Query work on Snowflake in a click!