SNMP is a simple network management protocol, which belongs to the application layer protocol in the TCP/IP five-layer protocol and is used for network management. Because the SNMP protocol is simple and reliable, it has been welcomed by many manufacturers and has become the most extensive network management protocol at present.
SNMP is divided into two roles:
- SNMP management station (manager, our native 127.0.0.1)
- SNMP agent (agent, the machine we want to operate, such as 192.168.1.144)
A management station refers to a server running software that can perform network management tasks. It is usually called a network management station (NMS). The NMS is responsible for sampling the information of the agent in the network and receiving traps from the agent. The agent is the part used to realize the SNMP function in the actual network equipment. The agent receives NMS read and write request messages on UDP port 161, and the management station receives agent event notification messages on UDP port 162. Therefore, once the access authority of the device (community, public by default) is obtained, the device information can be accessed, and the device parameters can be modified and configured. Due to the UDP protocol, there is no need to maintain a connection between the agent and the management station.
SNMP currently has three versions: v1, v2, and v3, all are supported on Ursalink routers and gateways:
- v1 is the original version of the SNMP protocol, but it is still the basic way for many manufacturers to implement SNMP.
- v2 is usually referred to as community-based SNMP V2. Community is essentially a password.
v3 added support for authentication and ciphertext transmission.
The management information (MIB) library can be understood as the management object database maintained by the agent. Most of the status and statistical information of the management objects defined in the MIB can be accessed by the NMS. MIB is a tree structure organized in a hierarchical structure. Each managed object corresponds to a leaf node of the tree structure, called an object, and has a unique digital identifier.
MIB data objects are organized in a tree-like hierarchical structure, each branch in this tree-like structure has a dedicated name and a numeric identifier. The branches of the structure tree actually represent the logical grouping of data objects. The leaves, sometimes called nodes, represent various data objects. The use of subtrees in the structure tree represents increased intermediate branches and increased leaves.
Using this tree-like hierarchical structure, MIB browsers can access the entire MIB database in a convenient and concise way. MIB browsers are such a tool that can traverse the entire MIB structure tree, usually in the form of graphical display. Represents each branch and leaf object, the last NODE is also the OID we usually say.