I. Working principle of a switch
The working principle of a switch (Layer 2 switch): A switch, like a bridge, is a networking device that operates at the data link layer. Each of its ports has bridging capabilities, and each port can connect to a LAN, a high-performance website, or a server. It can learn the device connection status of each port through self-learning. All ports are controlled by a dedicated processor, and information is forwarded via a control and management bus.
It can also be centrally managed using dedicated network management software. In addition, switches generally support multiple methods to improve data exchange speed and efficiency.
(1) Store-and-forward:
All conventional bridges use this method. Before sending data frames to other ports, they store the received frames completely in their internal memory, verify them, and then send them to other ports. This results in a latency equal to the sum of the time to receive a complete data frame and the processing time. If the cascading is long, this can lead to serious performance problems, but this method can filter out erroneous data frames.
(2) Cut-in method:
This method only checks the destination address of the data frame, which allows the data frame to be transmitted almost immediately, thus greatly reducing latency.
Its disadvantage is that erroneous frames will also be transmitted. When the probability of erroneous frames is low, the segmentation method can be used to improve transmission speed. When the probability of erroneous frames is high, the store-and-forward method can be used to reduce the retransmission of erroneous frames.
II. Switch Fault Classification
1. Power supply failure
Unstable external power supply, aging power lines, or lightning strikes can damage the power supply or cause the fan to stop, preventing normal operation. Power supply issues can also frequently lead to damage to other internal components.
If the POWER indicator light on the panel is green, it indicates that the switch is functioning normally; if the indicator light is off, it means that the switch is not receiving power. These types of problems are easy to spot, easy to resolve, and also the easiest to prevent. To address this type of fault, the first step should be to ensure a stable external power supply.
Generally, an independent power supply is provided by introducing a dedicated power line, and a voltage regulator is added to avoid instantaneous high or low voltage fluctuations. If conditions permit, a UPS (Uninterruptible Power Supply) can be added to ensure the normal power supply to the switch. Some UPSs provide voltage stabilization functions, while others do not; this should be considered when selecting one. Professional lightning protection measures should be installed in the computer room to prevent lightning damage to the switch. Many professional companies now offer lightning protection engineering services; these should be considered when implementing network cabling.
2. Module failure
A switch is composed of many modules, such as stacking modules, management modules (also called control modules), and expansion modules. The probability of these modules failing is very small, but if a problem does occur, it can result in significant financial losses. Careless insertion or removal of modules, impacts during switch handling, or unstable power supplies can all lead to such failures.
Of course, all three modules mentioned above have external interfaces, making them relatively easy to identify. Some can even be identified by indicator lights on the modules. For example, a stacking module has a flat trapezoidal port, or some switches have a USB-like interface. The management module has a CONSOLE port for connecting to the network management computer for easy management. If the expansion module uses fiber optic connections, it will have a pair of fiber optic interfaces.
When troubleshooting this type of fault, first ensure that the switch and modules are powered normally. Then check that each module is plugged into the correct position. Finally, check that the cables connecting the modules are intact. When connecting the management module, also consider whether it uses the specified connection speed.
3. Cable fault
In theory, this type of fault is not a fault of the switch itself. However, in actual use, cable faults often cause the switch system or ports to malfunction. Therefore, this type of fault is also classified as a switch hardware fault.
For example, loose connectors, incorrect or non-standard cable arrangement during fabrication, using straight-through cables instead of crossover cables, cross-linked optical fibers in fiber optic cables, and network loops caused by incorrect wiring connections.
