Introduction This document explains common problems while configuring EtherChannel between Cisco Catalyst Switches with Cisco Network Assistant (CNA). Requirements 1. Two switches Switch1 and Switch2 with Cisco IOS® Software Release 12.1  or later that is installed and powered on 2. You must have completed the steps in Configure  the Catalyst Switch with Cisco Network Assistant document for both  the switches 3. A straight- through Ethernet cable to connect your PC to the switch  local Catalyst 3560 switch 4. A PC with CNA software installed on it Note: This document is based on CNA 5.2. If your PC does not  have CNA installed, or if you need to upgrade your software, refer to Download and Install Cisco Network Assistant 5. Complete the LAN Addressing Worksheet Worksheets as instructed in the Site Survey Common Issues 1. Unable to launch Cisco network assistance (CAN) from the PC which is directly connected to the switch 2. The EtherChannel status shows Down(Layer 2) Solution Q. Unable to launch Cisco network assistance (CAN) from the PC which is directly connected to the switch A. Ensure that the power-on self test (POST) is completed successfully. The SYSTEM LED must be solid green. After you connect the switch to the     PC, you must wait for 30 seconds before you can try to connect. Wait for 30 seconds, and then try to connect again. Q. The EtherChannel status shows Down (Layer 2) A. EtherChannel or LACP (Link Aggregation Control Protocol) may not be currently configured on the remote switch or ports which need to be bundled  to form EtherChannel. Ensure all the remote and local switch ports which need to be bundled to form EtherChannel are up by entering shut and then no shut command under Interface configuration mode using Command Line Interface. Related Information Cable Descriptions Configure an IP Address on Your PC Add a New Ethernet Switch ... View more

Purpose This document provides solution for the issue where 4500E catalyst switch fails to run on new image after upgrading from ipbasek9-mz to entservicesk9.mz on 4500E Catalyst Switch Problem Description The upgrade is done on the switch using the following commands Set the boot to use the new image by using the following commands : "Switch(config)#no boot system flash cat4500e-baseipk9-mz.122-54.SG1.bin" "Switch(config)#boot system flash cat4500e-entservicesk9-mz.122-54.SG1.bin" Save the configuration and reboot the switch. The switch fails to pick the new image and picks the old image with a message “using the first file in the flash...” Solution Solution 1 To fix this issue, get in to the Rommon mode, Enter the command  "boot bootflash:cat4500e-entservicesk9-mz.122-54.SG1.bin" When the switch is booted completely loading the config file, delete the old image from the flash and reboot to confirm the switch is responding normal. Solution 2 We can achieve the same configuration without deleting the old image.  Change the config register value to 0x2102 and use the correct boot statement. This will make the switch to boot by picking the correct image without deleting the old image. Source: - https://supportforums.cisco.com/thread/2089464 ... View more

After the upgradation when we do a "sh ver" we can see only switch 1 says "Next reload license Level: ipservices", not switch 2. Switch 2 specifies "Next reboot licensing Level   : ipbase". Issue To enable the "IPservices" license on the 3750X stack we are using the below commands 3750-sw01(config)#license boot level ipservices switch 1 3750-sw01(config)#license boot level ipservices switch 2 In switch 2, we can't go into config t mode to enter the above commands and config t mode fails. 3750-sw01-2#conf t                      ^ % Invalid input detected at '^' marker. Purpose This document will attempt to track known reasons and workaround/fixes for this error. Output for " show license detail switch 1 " and " show license detail switch 2 " 3750-sw01#show license detail switch 1 Index: 1        Feature: ipbase                            Version: 1.0          License Type: Permanent          License State: Active, In Use          License Priority: Medium          License Count: Non-Counted          Store Index: 0          Store Name: Primary License Storage Index: 2        Feature: ipservices                        Version: 1.0          License Type: Evaluation          License State: Active, Not in Use, EULA accepted              Evaluation total period: 8 weeks 4 days              Evaluation period left: 8 weeks 4 days          License Priority: Low          License Count: Non-Counted          Store Index: 0          Store Name: Evaluation License Storage 3750-sw01#show license detail switch 2 Index: 1        Feature: ipbase                            Version: 1.0          License Type: Permanent          License State: Active, In Use          License Priority: Medium          License Count: Non-Counted          Store Index: 0          Store Name: Primary License Storage Index: 2        Feature: ipservices                        Version: 1.0          License Type: Evaluation          License State: Active, Not in Use, EULA not accepted              Evaluation total period: 8 weeks 4 days              Evaluation period left: 8 weeks 4 days          License Priority: None          License Count: Non-Counted          Store Index: 0          Store Name: Evaluation License Storage Resolution Only the show and debug commands are available in a CLI session to a specific stack member. You cannot go to the config mode of the member switch when the switch is in the stack. From the output of second switch"Index 2", you can see that the EULA is not accepted, which means that the evaluation license for ip services is not selected yet. Only when EULA enabled, the show version would say that on its next reboot ipservices will be selected. Check whether you are getting “End User License Agreement" (EULA) when type "license boot level ipservices switch 2"? If you do then it actually displays a very long message, and in the end it asks you if you would like to ACCEPT? (Yes/ No): --- if you choose "yes", it means that you have accepted the EULA, and then it displays a message saying on the next reboot ipservices will be used. If it doesn't display anything after you type "license boot level ipservices switch 2", would suggest you to unstack the member switches and upgrade the license to ipservices individually and the stack the switches. Source: - https://supportforums.cisco.com/thread/2080956 ... View more

Having a switch from Cisco: SG300-52, It functions only from a web gui, no cli available which has made configurations cumbersome. Issue We don't see a tab for assigning an IP Address to a VLAN under "VLAN Management". Purpose This document will attempt to track known reasons and workaround/fixes for this error. Our intended goal via the web gui is: Switch(config)#interface Vlan2 Switch(config-if)#ip address 10.x.x.x  255.255.255.0 Resolution Switch runs in Layer 2 mode by default.  If we want to put ip addresses on the vlans, need to change the switch mode from Layer 2 mode to Layer 3 mode.  We can ONLY do this in the console menu (not the web GUI).  It does indeed exist, you must access it using a baud rate of 115,200 (not the standard 9600), data bits = 8, parity = none, stop bits = 1, flow control = None.  It is also available via telnet (default username/password is cisco/cisco). Go to Switch Main Menu --> System Mode.  NOTE :  Switching between Layer 2 & Layer 3 modes will clear all other configurations that you have in place, so plan accordingly. You should be able to address the vlans after you've made this change. Source : - https://supportforums.cisco.com/message/3346316#3346316 ... View more

Issue We have 2 Bluecoat servers connected to the same switch that exchange multicast traffic. By default the switch may send multicast traffic to all ports. What is the best practice to optimize this traffic? Resolution If IGMP snooping is enabled and no multicast router is connected to the VLAN the switch may block this multicast traffic in an attempt to optimize traffic forwarding. If a multicast capable router is connected to the VLAN and regularly sends out IGMP queries then IGMP snooping will work well moving this multicast traffic only between really interested receivers. If no multicast router is connected to the VLAN/subnet then we should verify if our switch supports a feature called IGMP Snooping Querier that will send out the regular IGMP queries in order to make IGMP snooping to work correctly. Otherwise we need to disable IGMP snooping on that specific VLAN (if no multicast router is connected and no switch in use can act as IGMP Snooping Querier ) this will cause multicast traffic to be flooded on the VLAN like broadcast. If servers are connected to 2 different switches and there is a trunk between these 2 switches then each switch will intercept the IGMP report of the locally connected server and should propagate it on the trunk towards the multicast router/IGMP Snooping Querier . Source:- https://supportforums.cisco.com/message/3041967 ... View more

HSRP Behavior Issue: I have 2 Cisco 3560 Switch connected and a Trunk link is configured between them allowing all the VLAN. Also HSRP is configured with virtual IP address as 192.168.1.100. A server is connected with Multipath configured in its NIC. SWITCH 1 is ACTIVE & SWITCH 2 is STANDBY . By Chance if the trunk link between the switches goes down both the switches will become ACTIVE. In this scenario, how the traffic flow path is going to be? Since the virtual IP MAC will be given by both the switch, which path the Server NIC will choose? Resolution: If only one NIC is active then any traffic coming into the switch with the standby NIC connection destined for the server would need to be sent across the switch interconnect. If the interconnect fails between the switches this does not affect the active server NIC. The active server NIC will stay the same and therefore the server will continue to use the switch it was using before the failure. Obviously though return traffic may very well not reach the server as it may come in on the other switch. Source:- https://supportforums.cisco.com/message/3048197#3048197 ... View more

HSRP Behavior Issue: I have 2 Cisco 3560 Switch connected and a Trunk link is configured between them allowing all the VLAN. Also HSRP is configured with virtual IP address as 192.168.1.100. A server is connected with Teaming/Multipath configured in its NIC. SWITCH 1 is ACTIVE & SWITCH 2 is STANDBY . By Chance if the trunk link between the switches goes down both the switches will become ACTIVE. In this scenario, how the traffic flow path is going to be? Since the virtual IP MAC will be given by both the switch, which path the Server NIC will choose? Resolution: If only one NIC is active then any traffic coming into the switch with the standby NIC connection destined for the server would need to be sent across the switch interconnect. If the interconnect fails between the switches this does not affect the active server NIC. The active server NIC will stay the same and therefore the server will continue to use the switch it was using before the failure. Obviously though return traffic may very well not reach the server as it may come in on the other switch. Best practice is always to run more than one physical link between 2 switches doing HSRP, so you should definitely look to use Etherchannel for link redundancy. Source:- ... View more