SQL Server connectivity guide

Port 1433: Test SQL Server Connectivity Safely

Learn what SQL Server port 1433 does, verify the real listening port, test the complete network path, interpret failures, and avoid treating a reachable socket as proof that the database is ready.

12-minute readVerified July 24, 2026
Client traffic follows a protected route through a firewall to a database server
On this page

What is port 1433 used for?

Port 1433 is the conventional default TCP port for a default Microsoft SQL Server Database Engine instance. It is not guaranteed to be the active port. A named instance may choose a dynamic port, and an administrator can assign a different static port. Confirm the real listening port before changing a firewall rule.

TCPDatabase session transport
1433Default instance convention
UDP 1434Browser discovery
1 pathTest from the real client

Port 1433 vs 1434: test the correct protocol

The similar numbers cause a common diagnostic mistake. SQL Server database connections normally use a TCP port. For a default instance, that is commonly TCP 1433. SQL Server Browser listens on UDP 1434 and tells a client which TCP port a named instance uses. Browser discovery is not the database session itself.

ComponentTypical portPurposeWhat to test
Default instanceTCP 1433Database connectionsTCP path from the application
Named instanceDynamic or assigned TCP portDatabase connectionsDiscover the actual port first
SQL Server BrowserUDP 1434Instance-to-port discoveryOnly when discovery is required

What an open port 1433 proves—and does not prove

A successful TCP test proves

The tested client can resolve the target and complete a TCP connection to the selected host and port at that moment.

It does not prove

SQL authentication, database permissions, TLS compatibility, database names, application connection strings, query execution, or production readiness.

Use reachability as the first layer of a layered diagnosis: DNS, route, firewall, listening socket, TLS, authentication, authorization, and application behavior. This prevents the misleading conclusion that “the port is open, so SQL Server is healthy.”

Collect four facts before testing port 1433

  1. Exact hostUse the same DNS name or private IP the application uses. Testing localhost answers a different question.
  2. Actual instance portCheck SQL Server Configuration Manager or the error log. Do not assume a named instance uses 1433.
  3. Real client network pathRun the test from the application host, subnet, VPN, container, or build agent that experiences the failure.
  4. AuthorizationTest only approved systems. A port test needs a host and port—not a password, connection string, or production secret.

How to check if port 1433 is open

  1. Confirm the listenerReview TCP/IP settings and the error-log entry that reports the listening address and port. Restart the Database Engine if a configuration change requires it.
  2. Test from the affected clientReplace the placeholder below with an authorized DNS name or private IP. A true result confirms TCP reachability only for that source, destination, and moment.
  3. Use an explicit endpointTo bypass instance discovery, use tcp:server.example,1433. If the instance listens elsewhere, substitute the verified port.
  4. Continue above the network layerIf TCP succeeds but the application fails, inspect TLS, login method, account state, permissions, database name, driver, and connection-string options.
PowerShell
# Run from the same network path as the application.
Test-NetConnection -ComputerName db.internal.example -Port 1433

# Inspect the network-layer result:
TcpTestSucceeded : True

Do not publish real hosts, IP addresses, credentials, or connection strings. Use sanitized values when sharing diagnostic evidence.

Interpret common port 1433 test results

ResultLikely meaningNext check
TcpTestSucceeded: TrueTCP reached a listenerTest TLS, login, database selection, and permissions
Connection refusedThe target responded, but no listener accepted the port or a device actively rejected itVerify service, port, binding, and local firewall
Timeout / filteredA firewall, route, VPN, security group, or ACL may drop packetsTrace source-to-destination policy from the correct subnet
DNS failureThe client cannot resolve the hostCheck private DNS, VPN, suffix, and spelling
Login or TLS error after TCP succeedsThe network works; the failure is higher in the stackCheck certificates, encryption, authentication, driver, and permissions

Why SQL Server port 1433 is not listening

A missing 1433 listener does not automatically mean SQL Server is down. The instance may be healthy on another port. Check these causes in order:

  • Named instance: it may use a dynamic TCP port.
  • TCP/IP disabled: the protocol may not be enabled for this instance.
  • Pending restart: a port change may not take effect until the service restarts.
  • Wrong IP binding: the expected interface may be disabled or configured differently.
  • Port conflict: another process may already own the configured port.

The SQL Server error log records which IP addresses and ports the server is listening on. That evidence is more reliable than assuming the default.

Configure access without overexposing port 1433

An inbound rule should be a narrow business exception, not blanket internet exposure. Define the verified destination and TCP port, approved source ranges, environment, owner, review date, and removal condition.

Preferred controls

Private networks, VPN or peering, source allowlists, segmentation, TLS, least-privilege accounts, monitoring, and expiring change records.

Avoid

Opening TCP 1433 to every source, embedding credentials in scripts, or leaving temporary rules active after testing.

Validate every layer after port 1433 becomes reachable

A successful TCP connection is a useful boundary, but it is not the finish line. It shows that the tested client reached a listening process at the resolved address and port during that attempt. It does not identify the process as the intended SQL Server instance, prove that encryption policy can be negotiated, authenticate a login, select the expected database, authorize an operation, or show that a query returns correct results. Preserve the endpoint, source network, timestamp, resolved address, and command output so the result can be compared with application logs rather than remembered informally.

Validation layerEvidence of successIf it fails
Name resolutionThe hostname resolves to the approved address for the intended environment.DNS suffix, private zone, VPN context, stale cache, and split-horizon records.
TCP reachabilityA connection to the verified host and configured SQL Server TCP port succeeds.Listener binding, routing, firewall, security group, network ACL, and source allowlist.
TLS and pre-loginThe driver and server agree on encryption and the certificate is valid for policy and hostname.Driver version, trust chain, certificate name, expiry, protocol version, and encryption settings.
AuthenticationThe intended Windows, SQL, federated, or workload identity is accepted.Authentication mode, account state, credential source, clock, identity mapping, and sanitized error code.
Database authorizationThe login maps to the intended database user and can perform only the required operation.Default database, user mapping, roles, explicit denies, object grants, and row-level policy.

Run the application-level test from the same execution context as the failing workload whenever possible. A successful administrator test from a laptop may use a different DNS resolver, route, source address, driver, certificate store, identity, or connection string. Match the real workload before concluding that port 1433 is healthy. If a temporary firewall exception is used to isolate the cause, give it an owner and expiry, record the before-and-after result, then remove it promptly.

Check the SQL Server endpoint before deeper debugging

Use the InfiniSynapse DB Compatibility Checker to verify TCP reachability for an authorized SQL Server host and port. For SQL Server, this check is limited to the network endpoint: it does not validate credentials, TLS, permissions, or query execution.

Open DB Compatibility Checker Use only authorized, sanitized endpoint information. Do not paste secrets.

A practical decision tree for port 1433 failures

  1. Can the client resolve the hostname?If no, fix DNS, VPN, or the endpoint before inspecting SQL Server.
  2. Is the verified TCP port reachable?If no, distinguish refusal from timeout, then inspect listener, routing, firewall, security group, and source allowlist.
  3. Can the SQL client complete TLS and authentication?If TCP passes but the client fails, focus on certificates, encryption, login method, account state, and driver behavior.
  4. Can the account access the intended database?Validate the database name, default database, mappings, roles, and least-privilege permissions.

Port 1433 FAQ

What is port 1433 used for?

TCP port 1433 is the conventional default port for a default Microsoft SQL Server Database Engine instance. Named instances may use dynamic ports, and administrators can assign another static port.

Is SQL Server port 1433 TCP or UDP?

Database traffic on 1433 normally uses TCP. UDP 1434 is associated with SQL Server Browser and named-instance discovery.

How do I check whether port 1433 is open?

From the application's network path, run Test-NetConnection with the authorized host and verified port, then inspect TcpTestSucceeded.

Why is SQL Server port 1433 not listening?

The instance may use a dynamic or custom port, TCP/IP may be disabled, a restart may be pending, or the test may target the wrong interface.

Does an open port 1433 prove that SQL Server works?

No. It proves TCP reachability only, not authentication, permissions, TLS, database names, or query execution.

Should port 1433 be exposed to the public internet?

Avoid broad public exposure. Prefer private networking, strict source allowlists, encryption, least privilege, and monitoring.

Official Microsoft references

Port behavior and troubleshooting guidance were verified against Microsoft documentation. InfiniSynapse product claims are limited to the tool's stated TCP reachability scope.