Skip to content
Try free
Learn › Connectors: open the docs menu

Network requirements & IP whitelisting

Database and file server connectors require Mammoth's servers to reach your infrastructure over the internet. If your database, warehouse, or SFTP server sits behind a firewall, you need to add Mammoth's IP address to your allowlist before the connection can succeed.

This guide covers where to find the IP address, the default ports, and how to configure your firewall for each connector type.

Which connectors need a firewall rule?

The following connectors connect directly to your infrastructure. Their connection forms show a message with the IP address Mammoth connects from:

ConnectorDefault portProtocol
PostgreSQL5432TCP
MySQL3306TCP
SQL Server1433TCP
MongoDB27017TCP
Snowflake443HTTPS
Apache Hadoop10000TCP
Databricks443HTTPS
SAP HANA443 (form default); confirm with your adminTCP
SFTP22SSH
SendGrid443HTTPS

Google BigQuery and the Parquet (Azure) connector connect over standard HTTPS and don't require a firewall rule for typical setups—see the notes for each below if you've restricted access on your end. OAuth-based connectors (Google Drive, Salesforce, HubSpot, Shopify, etc.) connect to vendor-managed APIs and do not require firewall rules on your side.

Mammoth's IP address

When you open New Connection for one of the connectors above, the form shows the IP address Mammoth connects from. The message reads "Please make sure IP [address] is whitelisted in your database settings." The SFTP form shows "Mammoth connects from IP [address]. Add it to your server's allowlist if needed."

Allow that IP address in your firewall before you click Connect. In the examples below, <MAMMOTH_IP> stands for the address shown on the form.

Configure your firewall for each connector

PostgreSQL

  1. Open your pg_hba.conf file (typically at /etc/postgresql/{version}/main/pg_hba.conf)

  2. Add a line for Mammoth's IP:

    host    your_database    your_user    <MAMMOTH_IP>/32    md5

  3. Reload PostgreSQL: sudo systemctl reload postgresql

  4. If your server is behind a cloud security group (AWS, GCP, Azure), also open port 5432 inbound from Mammoth's IP in the security group settings.

MySQL

  1. Sign in to your MySQL instance as an admin user

  2. Create a user that can connect from Mammoth's IP and grant it read access:

    CREATE USER IF NOT EXISTS 'your_user'@'<MAMMOTH_IP>' IDENTIFIED BY 'password';
GRANT SELECT ON your_database.* TO 'your_user'@'<MAMMOTH_IP>';

  3. Open port 3306 inbound in your firewall or cloud security group.

SQL Server

  1. Open SQL Server Configuration Manager
  2. Under SQL Server Network Configuration → Protocols, ensure TCP/IP is enabled
  3. In your firewall or Windows Firewall, add an inbound rule allowing port 1433 from Mammoth's IP
  4. For cloud-hosted instances (Azure SQL, AWS RDS), update the inbound security group rules accordingly.

MongoDB

  1. Open your mongod.conf file
  2. Under net:, ensure bindIp allows connections from outside the host (or use a firewall to restrict at the network level)
  3. Add an inbound rule allowing port 27017 from Mammoth's IP in your firewall or cloud security group.
  4. Ensure your MongoDB user has at least read access on the relevant database.

Snowflake

If your Snowflake account has a network policy restricting IP access, add Mammoth's IP address to the allowed IP list of that policy in Snowflake.

Apache Hadoop (Hive)

Open port 10000 inbound from Mammoth's IP in your firewall or cloud security group.

Databricks

If your workspace restricts access by IP, add Mammoth's IP address to your workspace's IP access list.

See the Databricks connector guide for the rest of the connection setup.

SAP HANA

Open your HANA instance's listening port inbound from Mammoth's IP in your firewall or cloud security group. The port varies by instance—confirm the correct value with your HANA administrator (see the SAP HANA connector guide for details).

SFTP

Open the port your SFTP server listens on (default 22) inbound from Mammoth's IP. For password authentication, ensure the SFTP user exists and has read access to the relevant directory. For key-based authentication, see the SFTP connector guide.

Google BigQuery

BigQuery authenticates with a Google Cloud service account JSON key over HTTPS. Access is controlled through IAM permissions on the service account.

Parquet (Azure)

The Parquet connector reads from Azure Data Lake Storage over HTTPS. If you restricted access to your storage account with Azure storage firewalls or virtual networks, allow Mammoth's outbound IP address there. Contact Mammoth support to confirm the address for your workspace.

SendGrid

If your SendGrid account restricts API access by IP, add Mammoth's IP address, shown in the message on the SendGrid connection form, to the allowed list. Otherwise Mammoth's requests to your account may be blocked.

Test your connection after allowing the IP address

After allowing the IP address, fill in the connection form and click Connect. Mammoth validates the credentials and saves the connection. If validation fails, the error message appears at the bottom of the form.

If it fails, check:

  • Mammoth's IP address is in your allowlist
  • The correct port is open (see the table at the top of this page)
  • Your database user has permission to connect from an external host

Use a custom port

If your database runs on a non-standard port, enter your custom port in the Port field of the connection form, and make sure that port is open inbound from Mammoth's IP.

Report an issue · support@mammoth.io