Full guide to commonly used port numbers: from query to security practice
Port number is the "gate number" in TCP/IP communications. IP location host, port location host service. Rapid confirmation of the clearance of the transport, development and security clearance 80、443、22、3306、6379 The equivalent port correspondence can significantly reduce the time taken to locate the failure. This page provides a common local searchable port chart, which complements the scopeing, the use of the scene and the recommendations for exposure control.
How do you write down the port number three?
- 0-1023 Known Port:System-level services are commonly used, such as HTTP 80, HTTPS 443, SSH 22, DNS 53. The listening usually requires higher privileges.
- 1024-49151 Port of registration:Applications and intermediates are common, such as MySQL 3306, PostgreSQL 5432, Redis 6379, RDP 3389.
- 49152-65535 Dynamic/private ports:The client has a large number of temporary source ports, which can also be used for custom services, but are clearly registered in the document and firewall.
How do you run the problem with this tool?
- Enter port numbers or service keywords (e.g. mysql, ssh) in the search box.
- Reduced scope with classification pages: Web, databases, remote access, mail, file transfer.
- Confirms whether the protocol is TCP, UDP or both, and then cross-checks the server listening and safety group rules.
- The list of ports that will be "must be open to the public" will be opened, the rest will be rejected by default.
Typical scenes and protocol elements
The website is not open:80/443 First, we check the security team, firewall and reverse agent. This tool helps you to confirm the standard port and avoid mismatching HTTPS to the wrong port.
Database exposure risk:The direct opening of ports such as MySQL/MongoDB/Redis to the public network is extremely risky. Source IP, priority Intranet + VPN/Beauty.
Remotely managed face:RDP 3389, SSH 22 is a high frequency scanning target. You can change non-standard ports, enable key login, and limit source segments.
TCP and UDP:TCP reliable for connection and suitable for Web/mail/database; UDP faster without connection, suitable for DNS and part of real time operations. Release rules match the port with the protocol.
List of security recommendations
- Minimum exposure: the public network only opens the required port of entry for operations (commonly only 80/443).
- Management segregation: SSH/RDP/Databank is not released from any source throughout the network.
- Changes can be traced: reports are updated simultaneously, files are updated, surveillance and alarms are monitored.
- Periodic check: Review of the actual open list against port forms and clean-up of zombie services.
Capacity boundaries
This page is a port knowledge search that does not scan the host and does not judge whether a port is currently open. Physical connectivity testing is based on authorized professional processes. The content is intended only for study and daily service.
Supplement: Port position in network stack
The transfer layer (TCP/UDP) uses the port to distinguish multiple sessions from the same host. The common error order is whether DNS is deciphering the correct security team/firewall and whether the process is released and whether the target port is being monitored and the correct webcard is being bound. View the port meaning separately from the " listening state, access control" to avoid simply reconfiguring the configuration while ignoring the network layer rejection.
Check out common problems.
Why do services write TCP/UDP at the same time?
DNS 53, for example, may use two types of protocols in different scenarios. The firewall is configured to be released by actual agreement and cannot be used only by TCP.
Change the SSH port, is it safe?
The port changes only reduce the probability of being swept through by default scripts and cannot be a substitute for key login, source limitations and intrusion detection. It should be a layer of deep defence, not the only measure.