How does a cisco switch find a loop? How to solve Loop Problems, cisco

How does a cisco switch find a loop? How to solve Loop Problems, cisco

How does a cisco switch find a loop? Loop Problem Solution

How can we intuitively and quickly determine whether a loop exists on a cisco switch? And quickly locate the switch loop? Is this a problem?

Currently, there is no best way to view the loop. You can only view the loop through some information. In addition to the Spanning Tree, there are many preventive loop commands in the cisco environment. The loop will generate a broadcast storm, which will cause serious contact congestion. When the loop appears, the switch will send a prompt message, which can be found through the network management software, view the log server, or directly log on to the switch, in the cisco environment, you can use show cdp to find the loop. Alternatively, you can intuitively check whether the switch interface indicator is flashing, and then click show interface + an interface under the command line to view the traffic information of the interface. Alternatively, you can directly input the debug arp command. If N-plus ARP information appears continuously on the screen, note that it appears quickly. If it is not occasional, it indicates that there must be a loop. If no debug arp occurs, it can be switched off, then, check the ARP information displayed on the screen. The information will certainly include interfaces. Then, check these interfaces and finally confirm. We recommend that you use the last one.

Technical support:

1. If a loop occurs in the access layer, for example, if the HUB is short-circuited, the switch in the distribution layer is bound to be reflected, which leads to a high CPU usage of the distribution layer switch. :

Command:

# Show processes cpu history

2. we recommend that you use the Solarwinds tool to monitor the CPU Of all distributed layer switches. When users in a certain area are slow, first check the CPU utilization of the distributed switch in this area, and then check whether such a record exists in the log:

Command:

# Show log

Feb 29 08:47:54 BeiJing: K_EBM-4-HOSTFLAPPING: Host 00: 15: C6: 0D: 3F: BFin vlan 21 is flapping between port Po1 and port Gi2/1

3. Check which switch is the neighbor of the distribution layer switch (G2/1) and log on to the access layer switch:

Command:

# Show cdp neighbors

Check the log to check whether there is any err-disable log. Of course, the premise is that this detection function is configured. If this prompt is not displayed

Use

# Show interfaces | in (is up | 5 min) to see which interface has a huge volume of packets. If the port with a huge volume of packets is not a cascade port, congratulations, you have basically found the root cause.

4. If there are many vswitches in the distribution layer, repeat the preceding command multiple times to find the root location.

5. At this time, you can:

1. shutdown

2. Implement Storm Control

3. spanning-tree host

BGDUGARD enabled in global Mode

You can select the above three methods

Of course, the network devices in your LAN are Cisco. The detection methods for other devices may be different, but the idea is similar.