| Any great network administrator will tell you that a successful management strategy includes not only proactive monitoring of the servers and network by use of a polling mechanism, but also a comprehensive fault management strategy, including the smart capturing of SNMP Traps. Advanced Fault Management products are now within reach for the average small and medium size businesses and their networks, with the inclusion of some hot new monitoring products, both windows and linux based. These monitoring systems can discriminate between the important and non-critical alarms, and notify the user when there is truly something worthy to look at. The rest is just noise, and good systems can easily filter that out. What are SNMP traps? They are packets of information sent over the network by managed and unmanaged agents that have something to say. For example, an Oracle database administrator would be interested in hearing if the database was about to run out of disk space, or if there were an excessive number of authentication failures. These little bits of information provide timely updates about bad (or sometimes good), things happening to critical services and processes on the network. So now you have your trap receiving system in place. Now what? How does this integrate with the rest of your processes? There is some setup involved, yet, once one spends some time getting used to the messages that arrive and doing some preliminary filtering and categorizing, the best systems actually start to "learn" from the user-choices. These learned behavior heuristics can be applied automatically by the system to find patterns and generalities in the data. Once these patterns are found, even more power is harnessed by the system as it can filter out unwanted information or become even more aware of critical information that may be hiding in the sea of messages being sent by hosts on the network.
|
| Networking | ||||||||||||||||||||||||||||||||||||||||||
|
| ||||
