DOC HOME SITE MAP MAN PAGES GNU INFO SEARCH PRINT BOOK
 
Configuring the Simple Network Management Protocol (SNMP)

Traps

Although SNMP is normally used in a synchronous manner in which systems poll each other periodically, it does include a mechanism for asynchronous events. In SNMP, these are referred to as ``traps.'' Traps are sent by agents to management stations to signal the occurrence of an asynchronous event.

Because traps are asynchronous messages signaling an event that may require attention, there are no responses from the management station to them. An agent may generate seven kinds of trap:


coldStart
Implies that the sending agent is reinitializing itself with significant changes in its configuration, for example, after a reboot.

warmStart
Implies that the sending agent is reinitializing itself with no changes in its configuration.

linkDown
Implies that a failure in one of the communication links of the sending agent.

linkUp
Implies that one of the sending agent's communication links has been restored to normal running status.

authenticationFailure
Implies that an instance of authentication failure has occurred at the sending agent.

egpNeighborLoss
Implies that an EGP peer relationship of the sending agent's EGP protocol has been lost.

enterpriseSpecific
Implies that an enterprise-specific event has occurred.
SCO SNMP includes two programs for testing the transmission and reception of traps: trap_send and trap_rece. See ``Testing trap delivery and reception'' for information on these commands.

See also:


Next topic: Authentication
Previous topic: Flow diagram of the trap operation

© 2003 Caldera International, Inc. All rights reserved.
SCO OpenServer Release 5.0.7 -- 11 February 2003