I have 1 dell powerconnect 5324 that connects to a catalyst 3550 (call it catalyst 1), catalyst 1 connects to another catalyst 3550 (call it catalyst 2) that acts as the default gateway for catalyst 1.
I can get the 5324 and catalyst 1 to communicate no problem using a copper crossover cable. However, when I plug the 5324 into catalyst 1, it kills catlyst 1's connection to catlyst 2. Does anyone have any ideas or suggestions as to what might be causing the problem here...???
Does Catalyst2 plug into the network as a layer 2 device? It's possible that there is BPDU being sent by spanning tree and blocking the link. Does the status on either port between the two catalysts indicate that it is being blocked? Can you verify that there is no network loop being created.
Is catalyst1 able to get to catalyst 2, or is not getting beyond it? There is no implementation of cdp or lldp on the 5324, so it's not possible that the configuration would change. Can you try plugging another managed switch into this port on catalyst1?
There may be some other aspect involved in the configuration. If it turns out it's not a spanning tree related problem, add some of your configuration files to the reply and we'll take a deeper look.
as it turns out, I had to disable the spanning tree state on the dell switch. What I found strange was that I had the switch set up at my desk for testing purposes, and disabled the spanning tree. However, after rebooting the switch a few times for various other changes and moving it to the rack, the switch decided to re-enable the spanning tree (meaning that it reset that configuration property without my intervention); curiously enough all my other configurations were not effected. Is this standard behavior?
serverA is unable to ping serverB and serverB is unable to ping serverA. However, serverA IS able to access the dell powerconnect management webpage.
As it might be evident, I'm not exactly a professional network engineer here, sadly we have no one to fill that roll and the work needs to get done. Any insight to this problem would be greatly appreciated.
I just joined this forum,
This might be too late but,
The first thing I would check on the servers is the firewall state.
Make sure ICMP requests are enabled. You said you were new, ICMP is the tech term for PING.
New versions of Windows Servers turn this off by default.
eWisdom, thanks and I appreciate the information. The servers that I'm trying to communicate with that are behind the powerconnect switch are actually linux servers. I'm fairly certain that I didn't explicitly turn on the firewall or disable pinging when setting up these servers and I'm reasonably certain that those settings aren't what's causing the problem. I do appreciate the information though and I'll keep that in mind when dealing with the windows servers I have on the network. Thanks! Sadly I'm still having this issue and have pretty much given up on it for the time being as I certainly don't have the expertise to resolve it (and it'll probably turn out to be something simple to ;-)). So anyone with any advice or information, it would be greatly appreciated.
Make sure there is no packet filter set on any of the switches to drop ICMP packets.
Make sure both servers are on the same VLAN.
Cisco has recently release a new application to manage switches using the Cisco IOS.
It makes it easy to manage switches and moitor VLANS and port settings. Much better than using the Java based web tools.
The Cisco Network Assistant. Version 5.1 was just released.
In your first post, you mentioned Catalyst 2 acts as the default gateway for catalyst 1. This begs the question if your are purposefully trying to configure catalyst 2 as a "router" and trying to separate HOST A and HOST B across IP subnets... or... maybe theres simply an underlying concept you're misunderstanding about Switched Virtual Interfaces (SVI's) and host to host communication on the switched network.
If you don't mind, lets get the basics about your layer-3 topology out in the open.
1. IP Addressing. What IP addresses, subnet masks, and default gateways are you using for each device?
For windows use "ipconfig /all" and "route PRINT"
For Linux, use "ifconfig" and "route -n"
For Cisco, "show ip interface brief", and "show ip route"
For Dell PowerConnect, same as Cisco.
What I'm trying to determine here is if you have any IP configuration issues. Depending on which IP addresses are used, if you have the wrong subnet mask configured on one of the hosts, this may hinder a layer 2 discovery at a host level. So I would first completely verify all hosts have the exact same subnet mask if your design calls for it. Also, you want to make sure that there are no IP conflicts on your switch configurations.
Also, it would be helpful to understand your switch vlan's. The command line syntax on your Cisco switch (and I think Dell's as well) is "show vlan brief". This will give you a listing of your defined vlan's.
Also, regarding your lab and the change that mysteriously reverted back... unless you actually save your active configuration called your "running-config", the switch will revert back to what has been previously saved called your "startup-config". To save any changes made, you can issue the command "copy run star" or simply "write mem".
Have a nice day.
Message Edited by LinuxCruiser on 01-14-200703:00 PM
Message Edited by LinuxCruiser on 01-14-200703:01 PM
Dell-Steve A
48 Posts
455
0
Posted November 9th, 2006 17:00