SNMP - Simple Network Management Protocol
In May 1990 RFC 1157 was published, defining the SNMP version 1 (Simple Network Management Protocol). SNMP provides a systematic way to monitor and manage a computer network, quickly becoming a standard for network management.
In RFC 1441 to 1452 defined an improved version of SNMP (SNMPv2) that became an Internet standard.
The SNMP model
The SNMP model of a managed network consists of four components:
Managed nodes can be hosts, routers, bridges, printers or other devices. To be directly managed by SNMP, a node must be able to run an SNMP management process, called SNMP agent. Each agent maintains a local database that describes their state variables and history that affect their operation.
The network management stations is from managers, which are computers with special management software. The management station contains one or more processes that communicate with agents through the network, issuing commands and receiving responses.
SNMP Versions
Current versions approved by the IETF (Internet Engineering Task Force) are three:
SNMP Version 1 (SNMPv1) is the current standard SNMP protocol. It is defined in RFC 1157. Security in this version is based on community, which are nothing more than passwords: plain text which allows any SNMP based application to access information from the managed device, there are basically 3 communities in SNMPv1: read-only, read- writing and trap.
SNMP Version 2 (SNMPv2) is often referred to community-based work, is referenced by RFC 1905, 1906 and 1907, is an experiment in the IETF, but despite being experimental some vendors have started to use it in practice.
SNMP Version 3 (SNMPv3) This is the next version of the IETF standardization to achieve complete. This new version adds support for strong authentication and private communication between managed entities.
Managers and Agents
In the world of SNMP there are 2 types of entities: managers and agents. A manager is a server running some software that can handle administrative tasks on a network. Managers are often referred to as Network Managed Stations (NMS). An NMS is responsible for consulting (router, switch, Unix server, etc.) and receive traps, this information may be used to determine if any type of disaster has occurred. A trap is the way for the agent to tell the NMS that an event has occurred. The traps are sent asynchronously, not in response to requests from the NMS. The NMS is responsible for performing an action based on information received from the agent.
The second entity is the agent, is a piece of software running on the managed device on your network can be a separate program or may come built into the system eg Cisco System IOS in a router. Today, most devices come with an integrated SNMP agent type.
The agents provide manageable information on NMS, this is by tracking or monitoring of various operational aspects of the device. The NMS can query the status of each interface and take appropriate action if one of them this fall. When the agent knows that something bad has happened, it can send a trap to the NMS. This trap is originated in the agent and sent to the NMS, where this is handled properly.
It is important to keep in mind that both the petitions and the traps can happen at the same time. There are no restrictions on when the NMS can query the agent as the agent sends a trap to the NMS.
Manageable information structure (MIS) and management information base (MIB).
The Manageable information structure (MIS), provides a way to define managed objects and their behaviors. An agent has in his possession a list of objects that can be traced. One of these objects is the interface of the router (eg, up, down or test). This list collectively define the information to the NMS can use to determine the status of devices in which the agent resides.
Manageable information base can be thought of as an object database administrators who oversee agents. Any kind of state or statistical information can be accessed by the NMS if defined in the MIB. The MIS provides a way to define managed objects while the MIB is the definition (using syntax MIS) of the objects themselves. As a dictionary, which shows us how to pronounce a word and then gives its meaning, a MIB defines a textual name for a managed object and explain its meaning.
An agent may implement many MIBs, however all agents implement a particular MIB called MIB-II. This standard defines variables for things such as interface statistics (interface speed, MTU, octets sent, octets received etc.). The main purpose of the MIB-II is to provide general information manageable via TCP / IP. It does not cover all items that a seller may want to manage within their particular device.
If a vendor brings to market a new device and provides advanced features such as monitoring of work or other development, these features can not be defined on the MIB-II so the seller must create your own MIB to manage that device.
Naming a OIDs
Managed objects are organized into a hierarchical tree. This structure is the basis for SNMP management. An object identifier consists of a series of integers based on the nodes in the tree and separated by dots. However there are formats that are readable as any word which is a more friendly towards the user
In RFC 1441 to 1452 defined an improved version of SNMP (SNMPv2) that became an Internet standard.
The SNMP model
The SNMP model of a managed network consists of four components:
- Managed nodes.
- Managed stations.
- Management information.
- A protocol of administration.
Managed nodes can be hosts, routers, bridges, printers or other devices. To be directly managed by SNMP, a node must be able to run an SNMP management process, called SNMP agent. Each agent maintains a local database that describes their state variables and history that affect their operation.
The network management stations is from managers, which are computers with special management software. The management station contains one or more processes that communicate with agents through the network, issuing commands and receiving responses.
SNMP Versions
Current versions approved by the IETF (Internet Engineering Task Force) are three:
SNMP Version 1 (SNMPv1) is the current standard SNMP protocol. It is defined in RFC 1157. Security in this version is based on community, which are nothing more than passwords: plain text which allows any SNMP based application to access information from the managed device, there are basically 3 communities in SNMPv1: read-only, read- writing and trap.
SNMP Version 2 (SNMPv2) is often referred to community-based work, is referenced by RFC 1905, 1906 and 1907, is an experiment in the IETF, but despite being experimental some vendors have started to use it in practice.
SNMP Version 3 (SNMPv3) This is the next version of the IETF standardization to achieve complete. This new version adds support for strong authentication and private communication between managed entities.
Managers and Agents
In the world of SNMP there are 2 types of entities: managers and agents. A manager is a server running some software that can handle administrative tasks on a network. Managers are often referred to as Network Managed Stations (NMS). An NMS is responsible for consulting (router, switch, Unix server, etc.) and receive traps, this information may be used to determine if any type of disaster has occurred. A trap is the way for the agent to tell the NMS that an event has occurred. The traps are sent asynchronously, not in response to requests from the NMS. The NMS is responsible for performing an action based on information received from the agent.
The second entity is the agent, is a piece of software running on the managed device on your network can be a separate program or may come built into the system eg Cisco System IOS in a router. Today, most devices come with an integrated SNMP agent type.
The agents provide manageable information on NMS, this is by tracking or monitoring of various operational aspects of the device. The NMS can query the status of each interface and take appropriate action if one of them this fall. When the agent knows that something bad has happened, it can send a trap to the NMS. This trap is originated in the agent and sent to the NMS, where this is handled properly.
It is important to keep in mind that both the petitions and the traps can happen at the same time. There are no restrictions on when the NMS can query the agent as the agent sends a trap to the NMS.
Manageable information structure (MIS) and management information base (MIB).
The Manageable information structure (MIS), provides a way to define managed objects and their behaviors. An agent has in his possession a list of objects that can be traced. One of these objects is the interface of the router (eg, up, down or test). This list collectively define the information to the NMS can use to determine the status of devices in which the agent resides.
Manageable information base can be thought of as an object database administrators who oversee agents. Any kind of state or statistical information can be accessed by the NMS if defined in the MIB. The MIS provides a way to define managed objects while the MIB is the definition (using syntax MIS) of the objects themselves. As a dictionary, which shows us how to pronounce a word and then gives its meaning, a MIB defines a textual name for a managed object and explain its meaning.
An agent may implement many MIBs, however all agents implement a particular MIB called MIB-II. This standard defines variables for things such as interface statistics (interface speed, MTU, octets sent, octets received etc.). The main purpose of the MIB-II is to provide general information manageable via TCP / IP. It does not cover all items that a seller may want to manage within their particular device.
If a vendor brings to market a new device and provides advanced features such as monitoring of work or other development, these features can not be defined on the MIB-II so the seller must create your own MIB to manage that device.
Naming a OIDsManaged objects are organized into a hierarchical tree. This structure is the basis for SNMP management. An object identifier consists of a series of integers based on the nodes in the tree and separated by dots. However there are formats that are readable as any word which is a more friendly towards the user
0 Comments:
Post a Comment