Hi

On Wed, 5 Aug 2026 at 10:13, abhishek s <[email protected]> wrote:

> Hi team,
>
> Background
> The current PostgreSQL audit logs in Azure record are below
>
> Database user
>
> Database name
>
> Client IP (appears as the Zscaler egress IP)
>
> Application name (currently pgAdmin4)
>
> Problem
> Because the organization uses VPN or a similar proxy, PostgreSQL only sees
> the proxy's IP address instead of the actual user's workstation IP.
>
> As a result:
>
> The Client IP is not useful for identifying who accessed the database.
>
> The application_name is always "pgAdmin4", which is static.
>
> Security and audit teams cannot determine:
>
> Which workstation connected
>
> Which employee initiated the connection
>
> Which machine executed the SQL statements
>
> Proposed Enhancement:
> The request suggests that pgAdmin should automatically populate
> PostgreSQL's application_name connection parameter with client metadata
> when establishing the database connection.
>

You absolutely should NOT use application_name for security or audit
purposes. It's a client-side parameter that a user can override in any way
they like. Its purpose is to provide convenience when monitoring the server
by identifying what a connection is being used for. That's fine for day to
day monitoring with well behaved users, but you cannot rely on it for
audit/security purposes as an attacker can simply override it.

-- 
Dave Page
pgAdmin: https://www.pgadmin.org
PostgreSQL: https://www.postgresql.org
pgEdge: https://www.pgedge.com

Reply via email to