FAQs about common vswitches

Question: My machine is connected to other devices through a vswitch in the same CIDR block but cannot be pinged. What are the causes?

Answer:: It may be a hardware fault or a setup fault. In the case of hardware faults, check whether the switch display lights, power supplies, and connections are correct, and whether the switch is normal. If the IP address is set to a fault, check whether the vswitch has an IP address. If the IP address is not in the same CIDR block as other devices, delete it or set an IP address in the same CIDR block as other devices; then let's see if the VLAN settings are faulty. If the switch is configured with different VLANs, the ports connecting to the switch belong to different VLANs, so they cannot be connected, you only need to remove the configured VLAN.

Question: I want to use the FEC function. When I connect two pair ports of two vswitches with two lines at the same time, I find that each switch is always a normal port and the other port shows a red light, is the switch port faulty?

Answer:: Generally, in this case, the switch is normal. Because the two switches are connected with two ports, the switch determines that a LOOP exists and it will automatically disconnect one of them, drop the corresponding port (the port with a red light ). Solution: Enable the Spanning-tree function (which is enabled by default) to let the switch know that the two ports are FEC functions and logically one port.

Question: My switch can work normally on a 10/100 Mbps adaptive Nic. Now, the server connected to the 100Mpbs Nic has a red light and cannot communicate, is the vswitch faulty?

Answer:: This is caused by improper configuration. The port of the vswitch may be forcibly set to 10 Mbps. An error is reported only when the port is connected to Mbps. Solution: Under the port configuration, set the port speed to adaptive or force to 100 Mpbs.

Question: My vswitch has several VLANs. My machines in the same VLAN are not in the same CIDR block. Can they communicate with each other?

Answer:: No. The same VLAN can only be in the same CIDR block. Different CIDR blocks cannot be in the same VLAN. Otherwise, the switch reports an error.

Question: I want to connect Cisco Catalyst 3524 to an existing network connected by a vswitch. The primary switch is an ipvst00009, and its GBIC is already connected to the ipvst4006 switch. I decided to use port 3 of the Catalyst 6509 module and connect the Fast Ethernet port using a 5-way adapter cable, that is, port 6509 of the Catalyst 3/15 and port 3524 of the Catalyst 0/1, set the two ports to 100 Mbps, full duplex, the same VLAN1 management domain and the same VTP Mode Server mode ). The trunk mode and packaging of the two ports are set to non-negotiation and 802.2q respectively. So there is a problem: When I ping any IP address from the Catalyst 3524 console on the network, although the success rate of port 0/5 is obtained, there is no data transmission between the two switches. I suspect that the Catalyst 3524 Ethernet port does not have the trunk capability, so it is displayed on the Catalyst 3524 console, but not enough information. What is going on?

Answer:A trunk is used to connect to a vswitch. It transmits information to multiple VLANs through the Layer 2 network, and the VLAN has various ports on the vswitch or vswitch. As a Layer 2 device, a vswitch does not have the ability to determine the network address. As long as they package various frames during switching, a router must select Layer 3 (Routing) somewhere.

When you connect to multiple switches on a dry line, in order to transmit information of multiple VLANs through them, you must establish some layers for communication so that all switches can transmit information collaboratively, it can be implemented through the trunk line protocol and VTP domain.

First, all vswitches must support the common trunk protocol. If all switches are Cisco, you can use the ISL protocol as the trunk package type. If not all switches are Cisco, or other types of switches may be added in the future, it is best to select the IEEE 802.1q trunk protocol. It can be seen that the Catalyst 4000 used in multiple vswitch networks only supports the 802.1q trunk protocol, so you cannot select the ISL protocol.

As far as I know, In a Cisco environment, 802.1q is called "dot1q" packaging. After you select the 802.1q protocol, you should activate the trunk line. However, not all modules of the Catalyst 6500 series switch support trunk lines. To verify whether a specific port supports trunk lines, type "show port capabilities 3/15". On the Catalyst 6509, you will see the following output:

In the above content, the most important part is the line starting with "trunk encap type", which shows how the specific port may be used. If you do not see any information about the trunk line, you cannot connect it to the trunk line.

On the Catalyst 2900XL and Catalyst 3500XL, any 100Base-TX or Gigabit Ethernet port can be packaged as an trunk port in the ISL or IEEE 802.1q mode. For Catalyst 5000/6000, such a command is required to set the trunk mode "set trunk 3/15 on dot1q", and then "show trunk" is used to check the settings. On the Catalyst 2900XL and Catalyst 3500XL, the equivalent command is as follows:

Interface fa0/1
Switchport mode trunk
Switchport trunk encapsulation dot1q

Then, use "show interface fa0/1 switchport" to check the settings.

For VTP settings, you mentioned that the vswitches at both ends are set to the "server" mode. I don't know if this is required for your network. Generally, information is transmitted through servers on switches with changed VLAN structures. Generally, the allocation-level or access-level switches are set to "vtp client", so that they can receive and send the information, but do not actively send the information. If your Catalyst 3524 is set as a server mode and has Versions later than other access server vtp databases, it ignores data updates and may send exception messages to each server.

Note that on the 802.1q trunk line, all switches must share the same VLAN management domain without tags ). In addition, it should be noted that 80% of the connection problems occur at the physical layer, especially when the data has not been transmitted or there is no cdp after careful settings), you can unplug the cable, connect the cable detector or use any method you can think of to check the connection of the cable. It is also possible that the link lights on both ends are shining, but no data is actually transmitted from either side up.

Another point is that if you turn off the VTP branch when you use a Catalyst 6500 and a Catalyst 4000 Series Switch to dial when they are connected, there will be some technical problems that cause a large amount of information transmission between switches, and cause network instability.