The most common 8 major faults and solutions for switch networking

There are many friends in the weak AC VIP group who have repeatedly mentioned network failures. Among them, there are many common faults when the switch is connected. In order to facilitate the troubleshooting of these faults, some common typical fault cases and solutions are introduced here.

Fault 1: The network cannot communicate when the switch is powered up

Fault phenomenon

When the switch is just turned on, it cannot connect to other networks. It takes a while to wait. In addition, after a period of use, access to other computers is fast, if you do not use the network for a while, the speed will slow down when you visit.

Failure analysis

Because this switch is a network manageable switch, the network management switch can enable the spanning tree protocol by default in order to avoid the network topology in the network. This way, even if there is a loop in the network, only one path will be reserved, and other links will be automatically cut off. Therefore, when the switch is powered on, each port needs to enter the state of monitoring, learning, and forwarding in turn. This process takes about 3 to 5 minutes.

If you need to start the switch quickly, you can start "PortFast" on the port directly connected to the computer, so that the port can be immediately switched to the forwarding state, so that the device can immediately connect to the network, avoiding the port from being monitored and learning to forwarding. The wait time required for a state transition.

Troubleshooting

If you need to implement data forwarding quickly after the switch is powered up, you can disable the extended tree protocol or set the port to PortFast mode. However, it should be noted that although the two methods omits the port detection process, once a topology loop is generated between network devices, network communication will be paralyzed.

Fault 2: The switch of 5 ports can only use 4 ports.

Fault phenomenon

There are 4 computers in the office, but there is only one information outlet, so a switch with 5 ports (one of which is an UpLink port) is configured. It was originally thought that four computers were connected to four interfaces, and one UpLink port was used to connect to the LAN. However, after accessing the network, the No. 1 port adjacent to the UpLink port could not be used normally.

Failure analysis

The UpLink port cannot be considered a separate port because it is actually a port with an adjacent port, but the applicable connection object is different. With the UpLink port, the hub device can be connected to the normal port of another hub device using a straight-through cable, eliminating the need for crossover cables.

The chips of switches and hubs are usually ×4, so most of the line device ports are 4 ports, 8 ports, 16 ports, 24 ports, etc. If you make 5 ports, you will waste 3 modules, which increases the cost.

Troubleshooting

Replace the 4-port switch with an 8-port switch to resolve the fault.

Fault 3: The "COL" indicator is on or blinking continuously, communication is not possible

Fault phenomenon

Computers on the LAN access the server through a hub, but one day it was discovered that all client computers could not connect to the server, and the Ping between clients was intermittent. Check the hub and find that the “COL” indicator is on or blinking.

Failure analysis

The "COL" indicator is used to indicate collisions and collisions in the network. The "COL" light flashes continuously, indicating that a collision has occurred; a long "COL" light indicates that a large number of collisions have occurred. The cause of a large number of conflicts may be a hub failure or a network card failure. In general, the possibility of a network card failure is relatively small, so the focus is on the exclusion of the hub.

Troubleshooting

Replace the hub and the network is back to normal.

Fault 4: After upgrading to Gigabit network, the server connection is intermittent

Fault phenomenon

The original server used a 10/100 Mbit/s network card and everything worked fine. However, after installing a 1000Mbit/s NIC and connecting it to the 1000Base-T port of the central switch, the connection between the server and the network is intermittent, the connection is extremely unstable, and normal network services cannot be provided. The network tester was used to test the network and found that the connectivity of the twisted pair link was no problem.

Failure analysis

The connection is normal at 100 Mbit/s, but it only fails when it is upgraded to 1000 Mbit/s. It seems that the cause of this failure may be a Category 5 cabling problem. Although in theory, the Super Category 5 system supports a transmission rate of 1000 Mbit/s, if the quality of the twisted pair, patch panel, network cable and other network equipment is not very good, or the termination process is problematic, 1000 Mbit cannot be realized. /s bandwidth.

Since 1000Base-T needs to use all four pairs of twisted pairs, the effective transmission rate of each pair is 250Mbit/s, and the full duplex transmission is completed. Therefore, the signal degradation of the 1000Base-T pair twisted pair is reduced, echoed, Back to the electrical properties of wear, crosstalk and anti-electromagnetic interference have higher requirements. If the performance of the twisted pair or other accessories is not good, serious crosstalk will occur between the online pairs, resulting in communication failure.

Troubleshooting

Considering that the performance of the Category 5 cabling system may not be able to meet the Gigabit network system, the fault is solved after replacing it with the Category 6 cabling product.

Fault 5: Although the Link light keeps flashing, the network speed is very slow.

Fault phenomenon

The Internet speed of the server is very slow. It is very slow to open the webpage at the beginning. Later, even the webpage cannot be opened, and Ping.com cannot resolve the address. At first thought it was a DNS setup or a server failure, but these are all working. Try to ping other computers and find that the packet loss rate is high. At this time, the Link indicator of the switch blinks continuously, and the data exchange is very frequent, indicating that the computer is constantly sending and receiving data packets. After the switch is turned off and then turned back on, the fault is alleviated, but this failure occurs after a while.

Failure analysis

From the point of view of the failure phenomenon, this is a broadcast storm within the network. Broadcast storms can occur for a variety of reasons, such as worms, switch port failures, NIC failures, link redundancy without enabling Spanning Tree Protocol, network line sequence errors, or interference. Viewing the switch indicator light when a network failure occurs is a convenient way to visually check network connectivity and network traffic.

Troubleshooting

As far as the current situation is concerned, the worm is the most important cause of network paralysis. Update the system patch for the server in time, install the network version of the virus killing software, and update the virus database for the server in time. After the server installs the anti-virus client program, the fault is solved.

Fault 6: Server resource sharing failure

1. Unable to assign access rights to users

Fault phenomenon

The entire network uses the Windows domain and the client is Windows 2000 Professional. The IP address of the server is set to 192.168.0.1, the DNS is 127.0.0.1, and the internal IP address of the router is 192.168.0.1. All clients adopt the method of automatically obtaining an IP address and belong to the DomainUser group. When the server sets a shared file, although it can be specified, it cannot be accessed.

Failure analysis

In the Windows domain, NTFS permissions and share permissions are used to set access permissions for shared folders. However, the NTFS permissions are higher than the shared folder permissions, which means that you must first set the NTFS permissions for the folder you want to set as shared, and then set the shared folder permissions for it. If there is a conflict between the two, then the NTFS permissions will prevail.

Troubleshooting

Specify NTFS permissions for the user before specifying the shared folder permissions. For example, you need to create a shared folder TESTA for user A, so that the shared folder can be completely controlled by user A. To be accessed by any other user, you must first set the access rights of TESTA and specify the "full control" permission for user A. And set "read-only" permission for Everyone. Also, set this when setting the permissions of the shared folder.

2. Shared folders cannot be displayed in "My Network Places"

Fault phenomenon

Some folders have been shared, but they are not viewable in My Network Places, but some shared files on the same computer are visible.

Failure analysis

Since some shared folders are visible, the network configuration and connection of the computer is basically normal. And this is not a fault, but a type of configuration that belongs to a shared property. In Windows systems, there are two main types of shared files, one for system calls and the other for other users. Shared files for system calls are not present in "My Network Places", but can be displayed with commands such as "Net View"; shared files for other users can be seen in "My Network Places".

So how do you configure an invisible shared folder? Just put a dollar sign "$" after the shared folder name. For example, in the Windows Server 2003 system, the folder automatically created for each user is such a shared type folder, each user can only see his or her own user folder, and cannot see other people's user folders. There are also some disks. In Windows Server 2003, these disks are shared after installation, but their shared file names have a "$" symbol, so the client users cannot see them.

Troubleshooting

The "$" symbol after the shared file name is deleted, and the shared file that cannot be displayed can appear in "My Network Places".

Fault 7: Hub and router cannot share Internet access

Fault phenomenon

Multiple computers use broadband routers and hubs to share the Internet using hub expansion ports. After the connection is completed, 3 machines directly connected to the LAN port of the broadband router can access the Internet, but the computer connected through the hub cannot access the Internet. No crossover or parallel line between the router and the hub is used, and the hub and the router LAN port are not connected. The connected light does not illuminate. In addition, the computer on the hub cannot ping the router, nor can it ping other computers. What is the reason?

Failure analysis

1. The hub itself is faulty

The problem is that computers on the hub cannot ping each other, and it is even impossible to ping the router. This failure can only affect all computers connected to the hub.

2. Cascade failure

For example, the cascaded jumper between the router and the hub uses an incorrect line sequence, or a jumper connectivity failure, or an incorrect cascade port. The symptom is that the computers on the hub can ping through, but cannot ping the router. However, Internet access to computers directly connected to the router's LAN port will not be affected.

3. Broadband router failure

If the LAN port fails, the result will be similar to the cascading failure: if it is a routing failure, the result will be that the computers in the network cannot access the Internet, whether connected to the LAN port of the router or connected to the router.

Troubleshooting

From the point of view of the fault phenomenon, the computer connected to the hub can neither ping the router nor ping other computers, and initially concludes that it should be a connection failure between the computer and the hub. At this point, you can try to replace a network cable first. If you still can't troubleshoot it, you can replace it with a hub.

Fault 8: IP address conflict

Fault phenomenon

Recently, my computer often has the following situation, prompting "The system detected an IP address xxx.xxx.xxx.xxx and the network hardware address 00 05 3B 0C 12 B7 has an address conflict. The network operation of this system may be suddenly interrupted", Then I dropped the line for a minute or so and resumed the network connection. What is the reason for this, how to solve it?

Failure analysis

Such a system prompt is a typical IP address conflict, that is, the IP address used by the computer is identical to the IP address of another computer in the same network, resulting in communication failure. The MAC address of the network card that conflicts with this computer is "00 05 3B 0C 12 B7". Usually, the IP address conflict is caused by improperly assigned IP address of the network administrator or by other users setting the IP address privately.

Troubleshooting

Since the MAC address of the network card is unique, you can ask the network administrator to find the computer that conflicts with you by means of the MAC address and modify the IP address. Use the "IPCONFIG /ALL" command to view the computer's IP address and MAC address. Finally, use the "ARP-S IP address NIC physical address" command to bind this legal IP address to your NIC MAC address.