PostgreSQL

Connect a PostgreSQL database and Nexus can run read-only SQL against it, checking the rows directly when that’s the fastest way to confirm a hypothesis. Is a record in the state the symptoms suggest? When did a value last change?

​

What we support

Nexus queries PostgreSQL with read-only SQL. It writes the query, runs it for the window that matters, and reads back the results: joining across tables, filtering and aggregating, and reading from tables, views, and materialized views.

​

Exploring your schema

Nexus doesn’t need you to describe your database. It learns the shape on its own: the tables you have, their columns and types, primary and foreign keys, and how tables relate. From there it explores progressively, starting with an overview of the database, then pulling fuller detail on the specific tables a question turns out to need, rather than loading everything up front. That keeps queries accurate against large schemas and grounded in tables that actually exist. Partitioned tables are handled too: a table split by month or by customer reads the same as an ordinary one. How Nexus learns and uses this structure is covered in How telemetry works.

​

Security

Nexus queries your database to read from it, never to change it, and that guarantee is enforced in several independent layers rather than left to trust. Every connection is read-only, and stays that way. Connections open in read-only transaction mode and re-assert it before each query, so a session can’t be flipped into a writable state mid-flight. We recommend connecting a read-only database user too, so the guarantee holds on your side as well. Every query is parsed and checked before it runs. Each query is parsed with PostgreSQL’s own parser and must be a single SELECT. Anything that writes or changes structure, bundles multiple statements into one request, or reaches a write path another way is rejected. The functions a query may call are allowlisted to safe, read-only ones, so anything unrecognized is blocked rather than allowed by default. Sensitive system tables are off limits. Queries can’t read PostgreSQL’s internal tables for credentials, roles, other sessions’ activity, or server configuration, regardless of what the connected user could otherwise see. Queries can’t overload your database. Every query runs under a timeout and returns a capped number of rows, so a broad or expensive query stays bounded rather than running away. We hold only a small number of connections to your database, close idle ones quickly, and recycle them regularly. And we give up quickly on a database we can’t reach. For more on how we handle your data during AI processing, see our Trust Center.

​

Connecting PostgreSQL

You can either connect a database directly with its connection details, or connect AWS and let it discover your RDS and Aurora databases.

​

Directly

What you’ll need:

• The host and port of your database (port defaults to 5432).
• The database name to connect to.
• A username and password.
• The SSL mode to use, and any client certificate, client key, or CA certificate your database requires. Mutual TLS is supported for environments that need it, and a client certificate can stand in for the password.

1. From the Nexus telemetry settings, add a telemetry data source and choose Postgres.
2. If the database isn’t reachable from the public internet, which is the common case, set Network access to Private network and route through a proxy you run in your network.
3. Enter the connection details and credentials, or paste a full connection string to fill the fields in one go, then test the connection. The test passes only when the user can log in and read at least one table.
4. Once connected, the database is enabled for Nexus. You can disable it at any time.

​

Through AWS (RDS and Aurora)

Connect AWS with RDS among the selected services, and your databases are discovered for you. One selection covers both RDS and Aurora: an aurora-postgresql cluster is discovered alongside plain RDS PostgreSQL instances, and both appear as PostgreSQL data sources rather than a separate Aurora type. Discovery finds the databases and their endpoints, but not a way to log into them. So each discovered database needs a login, which you choose per database once it appears:

• RDS IAM. Give the role or user incident.io authenticates as rds-db:connect on the database user you want it to log in as. Each connection mints a short-lived token, so there’s no password stored with us. Connections verify the server against the AWS RDS trust bundle, which we ship, so you don’t need to supply a CA. You’ll need IAM authentication enabled on the database and a database user set up for it. See AWS’s guide to IAM database authentication.
• Username and password. Provide credentials for a database user as you would for a direct connection.

If a database isn’t reachable from the public internet, attach a proxy to the AWS connection. Discovered databases are disabled by default, so review what’s found and enable the ones worth querying during an incident.

​

Best practice

• Connect a read-only user. Nexus only ever reads, and a read-only user makes that enforceable on your side.
• On a direct connection, point it at a replica rather than your primary, so investigation queries never compete with production traffic. Databases discovered through AWS currently always connect to the primary.
• Grant access to the tables Nexus should see and no more. It explores only what the user can read.
• On RDS and Aurora, prefer RDS IAM over a stored password, so there’s no long-lived credential to rotate.

​

Related