Common faults and solutions of Ethernet switches
Ethernet switches, as core network devices, may cause local or global network paralysis due to their failure. From the followingHardware failures, configuration errors, network attacks, environmental factorsFour dimensions, systematically sorting out common fault phenomena, causes, and solutions.
1、 Hardware malfunction
1. Port malfunction
-
phenomenon：
- The port indicator light does not light up (LINK light off) or flashes frequently (UP/DOWN).
- The port is unable to communicate, but other ports are functioning properly.
-
cause：
- Poor cable contact (crystal head oxidation, cable breakage).
- Physical damage to the port (such as lightning strikes, electrostatic breakdown).
-
solution：
-
Step 1Replace the cable and test its continuity (using Fluke DSX-8000).
-
Step 2If the cable is normal, try connecting the faulty port to another device (such as a PC) to confirm if it is a hardware issue with the port.
-
Step 3If the port is damaged, it is necessary to replace the switch or module (such as Huawei S9235 supporting hot swappable modules).
2. Power failure
-
phenomenon：
- The switch cannot start or suddenly loses power during operation.
- The power indicator light is off or flashing abnormally.
-
cause：
- The power module is damaged (such as capacitor bulging or fan stalling).
- Loose power cord or unstable voltage (such as voltage fluctuation>± 10%).
-
solution：
-
Step 1Check the power cord connection, replace the power cord for testing.
-
Step 2Measure the output voltage of the power supply (normal range: AC 100-240V, DC -48V ± 20%).
-
Step 3If the power module is damaged, activate the redundant power supply (if any) or replace the module with a new one.
3. Heat dissipation failure
-
phenomenon：
- The switch restarts frequently or some ports are unable to communicate.
- Abnormal increase in body temperature (>60 ℃).
-
cause：
- The fan stops running or the speed is insufficient (<2000RPM).
- Blockage of heat sink (accumulation of dust).
-
solution：
-
Step 1Clean the dust on the heat sink (blow with compressed air).
-
Step 2Test the fan speed, and if it is abnormal, replace the fan (such as H3C S6850 fan with a lifespan of about 3 years).
-
Step 3Improve environmental ventilation (such as installing cabinet fans).
2、 Configuration error
1. VLAN configuration error
-
phenomenon：
- Devices within the same VLAN cannot communicate.
- Unexpected interoperability between devices in different VLANs (broadcast storm).
-
cause：
- VLAN division error (such as assigning the finance department and R&D department to the same VLAN).
- The trunk port is not properly configured (such as allowing all VLANs to pass through).
-
solution：
-
Step 1Check VLAN configuration（
display vlan）Confirm that the port VLAN allocation is correct. -
Step 2Check the Trunk port configuration (such as Cisco switches that require configuration)
switchport trunk allowed vlan）. -
Step 3Clear erroneous configurations and redivide VLANs.
2. STP protocol failure
-
phenomenon：
- The network loop caused a broadcast storm, resulting in a 100% CPU usage rate of the device.
- Some ports are blocked by STP, but there are still loops present.
-
cause：
- STP protocol not enabled (if not configured)
stp enable）. - STP root bridge election exception (such as multiple switches with the same priority).
-
solution：
-
Step 1Enable STP protocol (such as
stp mode rstp）. -
Step 2Manually configure root bridge priority (e.g
stp priority 4096）. -
Step 3: Use
display stp briefCheck the port status.
3. Routing configuration error
-
phenomenon：
- Cross VLAN communication failed.
- The default gateway is unreachable.
-
cause：
- The third layer switch is not configured with VLANIF interface IP.
- Static routing configuration error (such as reversing the destination network segment).
-
solution：
-
Step 1Configure VLANIF interface IP (such as
interface Vlanif 10，ip address 192.168.10.1 24）. -
Step 2Check static routing (such as
ip route-static 192.168.20.0 24 192.168.10.2）. -
Step 3: Use
pingandtracerouteTest connectivity.
3、 Network attack
1. MAC address flooding attack
-
phenomenon：
- Switch CAM table overflow (such as Huawei switch CAM table capacity of 8K).
- The switch enters Hub mode and broadcasts all traffic.
-
cause：
- The attacker forged a large number of MAC addresses, exhausting the CAM table space.
-
solution：
-
Step 1: Enable port security（
port-security enable）. -
Step 2Limit the maximum number of MAC addresses for a port (such as
port-security max-mac-num 3）. -
Step 3Configure static MAC address binding (such as
port-security mac-address sticky 001a.2b3c.4d5e）.
2. ARP spoofing attack
-
phenomenon：
- Users frequently disconnect or are redirected to phishing websites when accessing web pages.
- ARP table exception (such as gateway MAC address being tampered with).
-
cause：
- Attackers forge gateway ARP responses and hijack traffic.
-
solution：
-
Step 1Enable ARP anti spoofing function (such as
arp anti-attack check user-bind enable）. -
Step 2Configure static ARP binding (such as
arp static 192.168.1.1 001a.2b3c.4d5e）. -
Step 3Deploy ARP detection tools (such as Arpwatch).
3. DDoS attacks
-
phenomenon：
- The CPU usage rate of the switch continues to be 100%, and the port traffic is abnormally high.
- Normal business cannot be accessed.
-
cause：
- The attacker sends a large amount of SYN/UDP flooding traffic, exhausting switch resources.
-
solution：
-
Step 1Enable ACL to restrict abnormal traffic (such as
acl 3000，rule 5 deny tcp source any destination any eq 80）. -
Step 2Deploy traffic cleaning devices (such as Huawei AntiDDoS).
-
Step 3Upgrade switch firmware and fix known vulnerabilities.
4、 Environmental factors
1. Power interference
-
phenomenon：
- Frequent restarts of switches or port false alarms.
- The power indicator light is flashing abnormally.
-
cause：
- The power cord is laid in parallel with the strong current line (such as in the same slot as the air conditioning power cord).
- Poor grounding (grounding resistance>4 Ω).
-
solution：
-
Step 1Reposition the power cord to avoid crossing with strong electrical lines.
-
Step 2Check the grounding resistance and install grounding electrodes (such as copper bars buried 1.5 meters deep underground).
-
Step 3Enable UPS uninterruptible power supply to prevent voltage fluctuations.
2. Electromagnetic interference
-
phenomenon：
- The port error rate increases (such as an increase in CRC errors).
- Wireless devices (such as Wi Fi) have abnormal signals.
-
cause：
- Switches and strong electromagnetic equipment such as microwave ovens and frequency converters are located in the same room.
- Not using shielded Ethernet cables (such as Cat6a STP).
-
solution：
-
Step 1Move the switch to the electromagnetic shielding room (or install a shielding cover).
-
Step 2Replace with shielded Ethernet cable (STP) and ensure good grounding at both ends.
-
Step 3Increase the distance between devices (such as maintaining a distance of more than 3 meters from the microwave).
3. Excessive humidity
-
phenomenon：
- The oxidation of the switch interface resulted in poor contact.
- Condensed water has appeared on the fuselage.
-
cause：
- The humidity in the computer room is greater than 90% (such as during rainy season).
- The dehumidification function of the air conditioner has failed.
-
solution：
-
Step 1Install a dehumidifier to control the humidity between 40% and 60%.
-
Step 2Replace the oxidation interface (such as RJ45 crystal head).
-
Step 3Regularly check the dehumidification function of the air conditioner.
5、 Troubleshooting Process Table
| step | Operation content | Tools/Commands |
| 1 | Confirm the scope of the fault (single port/single device/entire network) |
pingThetraceroute
|
| 2 | Check physical connections (cables, port indicator lights) | Fluke DSX-8000、 visual inspection |
| 3 | View logs and alerts (CPU usage, error count) |
display logbufferThedisplay cpu-usage
|
| 4 | Test basic functions (VLAN, STP, routing) |
display vlanThedisplay stp briefThedisplay ip routing-table
|
| 5 | Isolation of faulty equipment (replacement method) | Backup switch, module |