1. Each device in a network effects other devices and the network as a whole. Nothing is isolated when connected to a network.
2. In an enterprise network, it is important that the network staff knows its responsibilities These definitions are very important to a networking department. They affect the workload of each person, and the cost of network services for the enterprise.
3. Job responsibilities need to be defined and enforced on a department-by-department basis. The key to understanding this relationship is that making the responsibility area too large may overburden the resources of the department, but making the area too small may make it difficult to effectively resolve the problems on the network.
4. Network administration encompasses many responsibilities, including cost analysis. This means determining not only the cost of network design and implementation, but also the cost of maintaining, upgrading, and monitoring the network.
5. Here are some of the other cost factors that must be considered:
· Network growth over time
· Technical and user training
· Repairs
· software deployment
6. The cost of redundant equipment for mission-critical operations should also be added to the cost of maintaining the network. A wise network administrator might keep a spare router on the premises to minimize the time his company is offline.
7. Although there are many reasons for network monitoring, the two primary reasons are predicting changes for future growth and detecting unexpected changes in network status. Unexpected changes might include things such as a router or switch failing, a hacker trying to gain illegal access to the network, or a communication link failure.
8. One of the major differences in a WAN-LAN comparison is the physical placement of equipment. The placement and use of monitoring tools becomes critical to the uninterrupted operation of the Wide Area Network.
9. One of the most basic forms of connection monitoring takes place every day on a network. The process of users logging on to the network will verify that connections are working properly or the networking department will soon be contacted. This is not the most efficient or preferable method of connection monitoring available. Simple programs are available that will allow the administrator to enter a list of host IP addresses and these addresses will be periodically pinged. The ping test only says that the connection is down, not where it is down.
10. A
better way is to ping just a few of the important hosts, servers,
routers, and switches to verify their connectivity.
11. Traffic monitoring is a much more sophisticated method of network monitoring. It looks at the actual packet traffic on the network and generates reports based upon the network traffic. Programs such as Microsoft Windows NT network monitor and Fluke's Network Analyzer are examples of this type of software. These programs not only detect failing equipment but also determine if a component is overloaded or poorly configured. The drawback to this type of program is that it normally works on a single segment at a time and if data needs to be gathered from other segments the monitoring software must be moved to that segment.
12. SNMP (Simple Network Management Protocol) is a protocol that allows management to transmit statistical data over the network to a central management console. It is a component of the Network Management Architecture. The Network management Architecture consists of four major components.
·
Management
Station
·
Management
Agent
·
Management
Information Base
· Network Management Protocol
13. One of the greatest enhancements to SNMP is called Remote Monitoring (RMON). RMON extensions to SNMP give the ability to look at the network as a whole as opposed to looking at individual devices. Probes gather remote data in RMON. A probe has the same function as a SNMP agent. A probe has RMON capabilities; an agent does not.
14. Redundant
management consoles provide a valuable feature for the network. Redundant
management consoles provide two major benefits to network management processes.
First is the ability to have more than one network administrator in different physical
locations monitor and manage the same network; for example one in
15. The RMON extension to the SNMP protocol creates new categories of data. These categories add more branches to the MIB database.
· The Ethernet Statistics Group
· The History Control Group
· The Alarm Group
· The Host Group
· The Host TOPN (Top [number]) Group
· The Matrix Group
· The Filter Group
· The Packet Group
· The Event Group
· The Token-Ring Group
16. It is important to remember that RMON is an extension to the SNMP protocol.
17. Problems happen! Even when the network is monitored, the equipment is reliable, and the users are careful, things may go wrong. The test of a good network administrator is the ability to analyze, troubleshoot, and correct problems under pressure of a network failure that causes company downtime.
18. The first and most important thing is to use your engineering journal and take notes.
19. Another essential element of preemptive troubleshooting is labeling.
20. The first step in network troubleshooting is the definition of the problem.
21. The Process of Elimination and Divide and Conquer techniques are the most successful methods for networking troubleshooting.
22. Software tools can help in Local Area Network troubleshooting, but are especially helpful in a Wide Area Network troubleshooting situation. The commands that are available to a network administrator in most client software packages include:
· ping
· traceroute
· telnet
· Netstat
· ARP
· IPconfig (winipcfg)