35

The answer depends on whether the switch is a Layer 2 or a Layer 3 switch. That is to say, is the switch only switching and relaying traffic on to a different device for routing, or, is it doing the routing decisions itself via SVIs (switched virtual interfaces). On a layer 3 switch, the port can be found by using a few simple commands on the device. ...


17

The cable pictured is a standard: IEC320-C15. The difference is in the temperature rating of the cable. These are rated to 120C, while the "ordinary" ones (IEC320-C13) are 70C. I believe the increased load for PoE switches requires the higher temperature rating.


13

Firstly, you need to get the MAC address, so get into a machine on the same VLAN and look at its neighbour table - Windows is netsh int ipv4 show neigh, Linux: ip nei Cisco: show ip arp x.x.x.x. Once you have that... If this is a discovery job on a layer 2 switch, do show mac address-table | i 0011.2233.4455 - replacing the mac address bytes as appropriate. ...


9

In theory, you'll want to migrate the clients to transparent before you remove the server from the equation. This makes absolute certain that things will continue to function without any gotchas. In practice, it doesn't matter. A server will just stop sending summary advertisements to the VTP domain and the clients will continue using the last seen ...


7

LACP gives you a significant advantage in that it verifies connectivity in both directions over each link. You can think of it as also combining the features of UDLD with link aggregation. For example, imagine you have the links configured in "mode on". Now lets say one of the links from the 3750 to the 2960 fails but in only one direction, say the ...


7

Unfortunately, you'll need the StackWise cables to stack 3750's. (And the StackWise modules as well depending on your switches.) See this Cisco white paper on Cisco StackWise for more information.


7

Regardless of whether your switch is performing frame forwarding (layer 2) or packet routing (layer 3), the following should work if the switch has a management IP address in the same subnet as the host you want to find: At the switch, ping the IP address you want to find. If the address is on the same subnet as the switch's management address, an ARP ...


6

A back-to-back SPAN session would seem to work. I can't say that is what was used in your environment but I just labbed it up and it worked with no problems. I was able to capture packets on the laptop that were from the gi0/7 interface on the 2960G. A poor mans RSPAN. Test layout and configurations below. No special configuration on the switchports at all, ...


6

First, you should probably not use VLAN 1 - it's generally not recommended, or if used, used for network device management traffic only. But, to answer your question, yes, you'd be using inter-vlan routing on the Cisco switch. See config below: ! enable layer-3 routing ip routing ! declare vlans vlan 1 name my vlan 1 vlan 21 name my vlan 21 vlan 2 ...


6

Your SSH client is saying "we can't agree on the cipher: server wants X but I'm configured not to speak X". This is a completely general SSH problem, not specifically related to this model of switch, which is that the SSH clients are regularly changed, usually to disallow older, weaker, ciphers. This isn't about key length, it's about cipher algorithms. ...


6

This happens when something downstream of the switch doesn't understand CDP and mishandles the traffic. So, something is connecting 2/0/5 to 4/0/6; chase the wiring and you'll find it. (someone has a dumb switch, or dumb VM leaking traffic)


5

You don't say which switch you are replacing nor specify the type of transceiver it has (you only specify the connector). However, the only 24 port Cisco switches I know of personally that have 100Mbps MT-RJ ports utilize 100base-FX, but make sure you double check. Will this switch support 100/1000 or 1000 base only? According to the Cisco 100-Megabit ...


5

In order for a VLAN interface to be up, the VLAN needs to be configured on a access port (Directly connected hosts) or the VLAN needs to be allowed over a trunk. Mind you that the interface or trunk need to be up for the VLAN interfaces to come up You can configure access ports assigned to those VLAN's and use the command no keepalive command to bring those ...


5

You should really set the IOS version to be the same before adding a switch to the stack. Basically, the switches need to have the same stack protocol version, and only switches with the same IOS version are guaranteed to have that. Switches with the same major version numbers probably have the same stack protocol version, but that is not guaranteed. Having ...


5

As already suggested by Stefan and Ricky, if your fiber ports are aggregated in an etherchannel, then you can safely remove any of the fibers and all the traffic will use the remaining fiber. In order to check if they are aggregated you can run show etherchannel summary and check the output lookinf or any PortChannel interface made of both fiber ports. ...


5

I am debating between taking all my servers and putting them on one of the 5 switches, or distributing them over the first few ports of all 5 switches? Best thing to do is distribute them over all five switches.If you config on one switch there may be downtime if any failure on that specific switch. if I distribute the ports across the stack, too ...


5

Could this be as simple as copying and pasting "show run?" In general, yes. You will probably need to edit the interfaces because the stackable switches will be something like GigabiteEthernet1/0/1 (Switch/Slot/Interface), but older switches may be something like FastEthernet0/1 (Slot/Interface). You can verify the interface name differences ahead of time. ...


4

You'll have to look at the configuration of the switches to see how those ports are being used -- etherchannel/bonded, spanning-tree redundancy, etc. You'll need to look at the config of both ends to see if it's safe to use the "backup" fibers at all. If it's just simple STP, then the blocked port will have an orange status led.


4

If you made an ether channel / port channel, don't worry, the traffic will go over the remaining fiber. Otherwise you could consider to make a channel. To be on the safe side, run show interface <name> several times for interfaces to both fibers on one switch and watch the counters. The one which doesn't change at all, while the other one does, should ...


4

Like @Brett said, StackWise is not possible without the use of stack ports. However, there is also really really old Cisco switch feature called cluster which uses just ethernet ports. I personally would never run this cluster feature and would steer away StackWise as well unless I have extremely good justification for it. Usually higher software ...


4

If what is important to you with this server is the availability and the uptime of the server, you should distribute the ports across the stack so if a switch goes down, you don't lose connection to your server. Also, I don't think the stack modules would be overloaded by the traffic at all. If you choose to put all the cables from the server in one switch,...


4

You can use the reload at <time> command to schedule a reboot of the whole stack but there is no way that I'm aware of to reload an individual stack member at a specific time, without resorting to third-party tools to issue the reload slot <number> command at a specific time. Edit 1: To set your stack member priorities, you can use the command ...


3

This looks normal to me. I think you may have a misunderstanding about the nature of this command. This should help, from Cisco documentation: The lacp port-priority interface configuration command determines which ports are bundled and which ports are put in hot-standby mode when there are more than eight ports in an LACP channel group.


3

As @Brett said, StackWise cables are required. Depending on what your goal is for the stack, you may be fine just creating an port-channel (Etherchannel) using up to 8x100Mb (800Mb) or 8x1Gb (8Gb) between 3750 switches. If you have certain 3750 models with 10/100Mb ports and 1000Mb uplinks, you could channelize up to 4 ports to get 4Gb.


3

Sadly portchannels on Cisco are a pain when adding new interfaces. Sometimes the only way to "recover" a Port-Channel is to completeley deconfigure and reconfigure it. Steps to do so: Reset all channel interface members to their default config default interface TenGigabitEthernet1/1/2 default interface TenGigabitEthernet2/1/2 Remove the channel interface ...


3

In Cisco catalyst 3750 switch configure Switch virtual interface (SVI) For 3 Vlans and enable " iP routing ". Communication among different Vlan routing is enable .hence different Vlan can communicate . For example in cisco 3750 multi layered switch Switch(config) # interface Vlan 1 Switch(config)#ip address 172.30.1.1 255.255.255.0 Switch(config)#...


3

Converting comment to an answer. You can upgrade from version 12.2(35)SE5 directly to the newest IOS version for the switch 3750 series found on Cisco download center. The upgrade process will automatically detect your system and upgrade various onboard controllers accordingly to support the new version. There's no need to do a midway upgrade.


2

I think your username is wrong in your commands, try substituting testusername for ALL snmp-server view TESTVIEW mib-2 included snmp-server group TESTGROUP v3 auth read TESTVIEW snmp-server user testusername TESTGROUP v3 auth sha testpassword


2

I got a response from Cisco: “Because of hardware limitation, "no ip redirect" does not work in Catalyst 3560 and 3750. If routing is performed on access switch and the next hop of the packet is a host on the same subnet of the ingress layer 3 interface, the frame is still punted to the CPU.” And there is also this thread https://supportforums....


2

Define the local group for the secondary AAA method for authorization, that way if the servers are down it'll use the local priv level for the user account (make sure it's set to 15 for full access) and continue from there.


Only top voted, non community-wiki answers of a minimum length are eligible