The Basics of Network Management with SNMP
SNMP (Simple Network Management Protocol) is a popular protocol for the management of networks, used for gathering and configuring information from network devices, such as hubs, printers, servers, switches, and routers on an IP (Internet Protocol) network, or any device which is network-capable. It is used for remotely monitoring and managing devices and their functions across a network, and has three components: SNMP agents, managed devices and network management systems (which are usually referred to as NMS). The managed devices are programmed to gather the information from the host device, to potentially store that information, and to be capable of forwarding the said information to the Network Management Station.
A SNMP agent is an item of software which is based on a host network device, for the purpose of collecting the programmed information and communicating it in the form of SNMP traps to the network management station. This station is used to monitor and control SNMP host devices by way of a system of simple messages, and it also conveys that information to the network monitoring team in the form of a type of report or alarm. The NMS is able to run several application processes which are designed to collect information about known devices on the network. This may be status information, for example the change in status of a router interface or switch port, changes in the device configuration or details relating to the performance of one of the managed devices.
It is often the case that other software applications which are loaded onto network management stations are proprietary, but they contain and operate TCP/IP (Transmission Control Protocol/Internet Protocol) standards based on SNMP. A good example of an SNMP application is HP OpenView.
Devices communicating with a NMS are quite often configured so that they belong to a known community, and they have a community string or password which provides the community authentication or identity. The later versions of SNMP comprise stronger forms of authentication and the encryption of SNMP messages. A NMS has to be a member of the community so that it can receive and view messages and or make changes in the configuration of the devices already in the community.
The agent software installed on to the host devices possess what is called a Management Information Base (MIB), which is a database with a list of the manageable objects linked to the device. When the device needs to report a change in the configuration or status, it sends a message, called a trap, to the community NMS, advising of the change.
There are a number of message types belonging to the NMS which it can use to make enquiries or make changes to the configuration on the managed devices, and these are as follows:
GET: This is used to ask for certain information from the managed device about an object on the device. It could be a request for the amount of disk space available on the device disk hard drive or for the status of a router interface.
GET-NEXT: This is used for requesting the value of the next managed object in the MIB belonging to the device.
SET: In this instance, the NMS can change the value of the managed object’s variable, if the object allows READ/WRITE access. Otherwise it will not be possible, if it is READ-ONLY.
The MIB on the managed device contains information organised in a hierarchical system which is identified by object identifiers, as they are known, which are universally recognised.
Picture courtesy of www.mytechportal.com