#yyds dry goods inventory #Switching equipment serious packet loss troubleshooting

Reason for packet loss

Terminal problems (local or peer computer, server problems)

Terminal-to-switch cable problems (too long, damaged, noisy)

Switch problems (performance, interface, power supply, long usage, high CPU, high temperature, inexplicable failure)

Intranet problems (arp spoofing, BINGDU, etc.),

Intermediate transmission equipment problems (CPU, memory, QoS, burst traffic; may be the fault of P2P),

Transfer intermediate firewall problem

Configuration problem (data link layer encapsulation protocol is different, duplex, speed are different, see the port)

Loop (check logging, check MAC, check CPU)

Process ideas

1. Determine the scope: is it a device, a floor, a device connected to a switch, or a vlanEquipment, it is easy to start after determining the range;

2. Process according to the range:

Single device pass

A switch is usually a problem with the switch. Log in to the device to check, it may be the switch is faulty, or the uplink port of the switch is faulty.Question;

A device of a vlan is partly due to configuration reasons, and the other is due to ARP BINGDU in the vlan, or other threats.A lot of it is ARP gateway spoofing;

On a certain floor, it is generally the problem of aggregation. The major reason is equipment failure. The aggregation is generally double-chain.On the way, it is unlikely that both links will fail at the same time, but also check the port, and

It may be a board problem, for example, the aggregation switch is 4500, and the redundant links are all on one board;

If it is this network failure, then generally start by looking at the core, mainly in the following aspects:

1. Check the CPU and logs of the device, mainly to see if there is a loop and serious alarms on the device;p>

2. Look at the equipment temperature and other environmental factors;

3. To see if the egress link of the device is normal, you can check by show interface XX , ping the next hop gateway of the egress ISP, etc.

4. Check if the board is ok, you can check it through show mod, etc.

5. Check whether the configuration has been modified, there are errors, check the routing, etc.

There is an operation method: ping, connect the laptop to the network cable of the faulty PC:

1. Ping the gateway, if the gateway ping packet loss is serious, then look at the PC---access switch---Convergence---Where is the fault between the cores

2. Ping the gateway is normal, then the segment from PC to gateway is ok, exclude this segment;p>

3. Ping gateway next hop, general router, if the ping is normal, go down and gradually ping, if not normal, then it can be determined that it is the fault between the core gateway and the next-hop Layer 3 device, locate and eliminate it;