Three goals to be implemented for network fault diagnosis

Network fault diagnosis should be achieved three purposes, a friend entrusted me to him to solve the problem of network fault diagnosis, asking a lot of people, there is no satisfactory answer. The following is a similar solution to the problem found on the Internet.





Network fault diagnosis should be realized in three aspects: to determine the fault point of the network, to restore the normal operation of the network, to find out the deficiencies of network planning and configuration, to improve and optimize the performance of the network, to observe the operation of the network and to predict the quality of network communication in time.





Network fault diagnosis is based on the knowledge of network principle, network configuration and network operation. From the phenomenon of fault, the Network diagnostic tool is used to obtain the diagnosis information, to determine the network fault point, to find the root of the problem, to troubleshoot and to restore the normal operation of the network.




There are several possible
network failures: physical devices in the physical layer fail to connect to each other or problems with the hardware and the wiring itself; Interface configuration problem of data link layer network equipment, Network Layer Network protocol configuration or operation error, transmission layer device performance or communication congestion problem;





three-tier Cisco IOS or network application error. The process of network fault diagnosis should go up from the physical layer along the OSI seven-layer model. First check the physical layer, then examine the data link layer, and so on, try to determine the failure point of communication failure until the system communication is normal.





Network Diagnostics can use a variety of tools, including LAN or WAN Analyzer:





Router diagnostics commands, network management tools, and other troubleshooting tools. Cisco provides the tools that are adequate to troubleshoot most network failures. Viewing the routing table is a good place to start troubleshooting network Diagnostics. ICMP ping, trace commands, and Cisco Show commands, debug commands, are network tools to get useful information for troubleshooting.





we usually use one or more commands to collect the appropriate information and, in a given case, determine what command to use to get the information that is needed. For example, the usual way to determine whether a device can be reached through an IP protocol is by using the ping command. Ping emits an ICMP packet from the source point to the destination, and if successful, the returned Ping packet confirms that all physical, data link, and network layers are functioning properly from source to destination.





How to understand its information after the Internet operation, to understand the normal operation of the network, monitoring and understanding of the network under normal conditions to run the details to understand the failure of the situation. What about the surveillance stuff? The show interface command makes it easy to get information about each interface you want to check.





Another show buffer command provides a regular display of buffer size, use, and usage conditions. The show proc Command and shows proc MEM commands can be used to track processor and memory usage, which can be collected on a regular basis and used for diagnostic references in the event of a failure.





Network fault diagnosis is manifested as a symptom, and symptoms include general (like a user not being able to access a server) and more special (such as routers not in the routing table). Use specific troubleshooting tools and methods for each symptom to find one or more failure reasons. General troubleshooting mode is as follows:





When analyzing network fault diagnosis, it is necessary to know the fault phenomenon first. The symptoms and potential causes of the failure should be explained in detail. To do this, determine the specific behavior of the failure, and then determine the type of cause of this failure phenomenon. For example, the host does not respond to customer request services. Possible causes of failure are host configuration issues, interface card failures, or the loss of router configuration commands.





collects the information that is needed to help isolate possible causes of failure. Ask users, network administrators, managers, and other key people about problems that are related to the failure. A wide range of information from Network Management Systems, protocol analysis and tracking, Router diagnostics command output reports or software manuals.





consider possible causes of failure based on the information collected. Some failure reasons can be excluded according to the situation. For example, depending on some data, you can troubleshoot a hardware problem and focus on software reasons. Any opportunity should try to reduce the possible cause of the failure, so as soon as possible to plan an effective troubleshooting plan.





establishes a diagnostic plan based on the last possible cause of failure. Start the diagnostic activity with only one of the most probable causes of failure, which makes it easy to revert to the original state of the failure. If you consider more than one fault at a time, it is much more difficult to try to return to the original state of the failure.





perform the diagnostic plan, carefully test and observe each step until the symptom disappears.





Each change to a parameter confirms its result. The analysis results determine whether the problem is resolved, and if not, continue until it is resolved.





Network fault layered diagnosis technology





physical layer and its diagnostic




The
physical layer is the most basic layer in the OSI layered structure system, which is based on the communication media, realizes the physical interface between the system and the communication media, transparently transmits between the data link entities, and provides services for establishing, maintaining and dismantling the physical connection between the computer and the network.




The failure of
physical layer is mainly manifested in the physical connection of the equipment, whether the connection cable is correct, and the configuration and operation of MODEM, CSU/DSU and other equipments are correct. The best way to determine whether a physical connection to a router port is intact is to use the show interface command to check the status of each port, interpret screen output information, view port status, protocol build status, and EIA status.





Data Link layer and its network fault diagnosis




The main task of the
Data link layer is to make the network layer obtain reliable transmission without knowing the characteristics of the physical layer. The data link layer is used to package, package, error detection and correction of data through the link layer, and coordinate the shared media. Before the data link layer exchanges data, the protocol is concerned with forming frames and synchronizing devices.





to find and troubleshoot the data link layer, you need to look at the configuration of the router and check the encapsulation of the same data link layer for the shared port. Each pair of interfaces has the same encapsulation as the other devices it communicates with. Check its encapsulation by looking at the configuration of the router, or use the show command to view the encapsulation of the corresponding interface.





Network layer and its network fault diagnosis




The
Network layer provides the means to establish, maintain and release network layer connections, including routing, flow control, transmission acknowledgement, interruption, error and failback. The basic way to troubleshoot a network layer is to follow the path from source to destination, view the Router routing table, and check the IP address of the router interface.





If the route does not appear in the routing table, you should check to determine whether the appropriate static route, default route, or dynamic route has been entered. Then manually configure some of the missing routes, or troubleshoot some of the dynamic routing processes, including RIP or IGRP routing protocol failures. For example, for IGRP routing information to exchange data only between systems of the same autonomous system number (AS), view the matching of autonomous system numbers for router configuration