Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Cannot ping Switch?

Status
Not open for further replies.

Guest_imported

New member
Jan 1, 1970
0
0
0
First lemme start off by saying I'm new to CISCO. Had a week long ICND course and that's about it. The question: I set up my Cat 3524 switch, gave it an IP address, Name, and the like. I can get into the switch thru the console port but I am unable to get in thru Telnet (which I did PW protect but never get the prompt) or Hyperterminal from my Workstation, I cant even ping it!!! I have gone in thru console and double checked the IP address, subnet mask and everything seems cool. I also downloaded the Java Pulgin to be able to manage from a web browser and all to no avail. I have tried both IE5.5 and Netscape version 4.5. This all makes no difference as I CANT EVEN PING IT! The switch is working fine but I would still like to be able to get to it from my workstation. Any Ideas? Did I leave anything out? Thanks.

Alex
 
Are you getting a link light when you connect your lan? Make sure that you are making a good connection to the switch and also make sure your interfaces are turned up.
 
I get the Link light and the Switch is working the way it should. It is already in production, but I am unable to get to it from anywhere but the Console port. Any Suggestions?
 
I have to ask if you happen to have the command "No ip directed-broadcast" or any other ICMP filter in place?

is the switch telnet IP on the same VLAN that you are on and if not, is it being routed?

What happens when you traceroute to the switch's IP that you use for telnet?

Mike S
 
Did you set the default gateway to point to your default router?
Can you get to the switch from the same subnet, but not from different subnets? That would point to something like a default gateway or netmask issue.
Also, are you messing with vlans, or just leaving all the ports in default mode?
 
Default Gateway is set to my Router (only one here), no VLANS and the entire network is on the same subnet. I posted the thread regarding integration with token Ring. I am able to see all clients on both sides, that is coming off without a problem. No filters in place, only set up the IP configuration, Default Gaetway and passwords, everything else is Vanilla. Tracert gives the default gateway and then times out. I dont understand?
 
I just thought of something.. I happened to configure a 2500 series router once without the enable password set.. it would not let me telnet to the router without the password set. It did give an error message but I wonder if the switches pull the same trick? My 6508s dont but they are a bit different then the *lowly* 3500 series ;)

have you checked the password config just to be sure ?

Mike S
 
TAMCO, let me take a step back. You wrote that you posted the Token Ring question. Is the device that you are trying to ping from on the token ring side? Do you have full IP connectivity throughout the rest of the segment? What I am trying to do is isolate the problem. I want to make sure that you not being able to ping the switch is not related to some other network problem.
 
wybnormal, I have checked it and everything seems cool. If that were the case I think at the very least I would be able to ping it?

TimothyR, No,I am directly connected to the Ethernet Switch from my workstation. I can see, ping, etc. everything from the ethernet side to the T/R side, and vice versa. If I werent the person posting and I read this,I would think there is something wrong with the IP address or Subnet Mask but that's all in order. At the very least I should be able to ping it, assuming that the IP configuration is right, right? I guess it could be much worse but I am still a bit frustrated.
Thanks,
Alex
 
If there were an access-list that dumped ICMP into a NULL interface or just the bit bucket, you would not get any response back from the target. Which is why I was asking, I have some security freaks around that have done this on occasion :)

Sure starting to sound like a misconfigured subnet/mask..

or...

Is this switch configured as a "command switch?" for use in a cluster? If it's a stand alone, it should be a "command switch"

Or...

do you have the correct default gateway set for the switch? It may get the packet but with an incorrect gateway, it might not be able to return it

or..

I verified that the 1900s will not let you telnet unless you have the enable password set.. but I was not able to find out if this is true with the newer 3500 series..


or.. re-setup the telnet using this guide..

Step 1

Attach a PC or workstation with emulation software to the switch console port.

The default data characteristics of the console port are 9600, 8, 1, no parity. When the command line appears, go to Step 2.

Step 2
enable
Enter privileged EXEC mode.

Step 3
config terminal
Enter global configuration mode.

Step 4
line vty 0 15
Enter the interface configuration mode for the Telnet interface.

There are 16 possible sessions on a command-capable switch. The 0 and 15 mean that you are configuring all 16 possible Telnet sessions.

Step 5
password <password>
Enter a password.

Step 6
end
Return to privileged EXEC mode so that you can verify the entry.

Step 7
show running-config
Display the running configuration.

The password is listed under the command line vty 0 15

Step 8
copy running-config startup-config
(Optional) Save the running configuration to the startup configuration

Mike S

 
TAMCO, you are correct. Just reading from my end it looks as if it is an IP or Subnetting problem. It is especially tricky since you are on the same network segment. Could you post your running-config MINUS the specific port entries and PLEASE either x-out or delete your passwords.

wybnormal, it should not matter whether TAMCO has a passwd on the vty or has clustering configured concerning the inability to ping, should it?
 
I finally got pissed enough to kill the switch and run thru the setup again. I had gone thru the setup before without resetting with no success, but after resetting the switch and going thru setup again all seems right. I know the reset should not make a difference and I am a little confused but I'll settle for no explanation and the darn thing running right than nothing at all. I realise it will look like an IP configuration error to most, but I'll chalk it up to Murphy. Thanks to all who helped, I really do appreciate it.

Alex
 
Tim-

On the 1900 series, if there is not a enable password set, you can not telnet to the switch and I wondered if the 3500s were the same. YOu can still get through the console but not the telnet. This is in the 1900 docs on the Cisco site. I dont know if is version dependent.

The clustering is a more interesting question that I can not answer as I do not cluster cisco switches ( yet). In the cluster one switch is the master and will answer the pings and telnets but not the others if I read the PDF files correctly. Perhaps I'm mistaken but I thought to toss out the idea in the hopes someone may have a definitive answer.

The reloading which Alex makes much more sense. I would guess his inital config was or is corrupted for some reason. A reload and reconfigure clear whatever problem it was and now it's &quot;happy&quot;. I have seen 3com switches go stupid enough that nothing short of a system reset will clear it enough to reboot and reload it. Why? who knows..

Good news Alex and good luck.

Mike S
 
wybnormal, I currently cluster 35xx switches. You give the cluster master an IP address and you can or cannot give the cluster members separate addresses. If you give the members IP addresses, then you can telnet to them like normal. If you don't assign them addresses, then you telnet into the master and use the &quot;rcommand&quot; command to go between the switches. Or, use the gui in your browser.

The only drawback to clustering that I have found as of yet is with SNMP. If the members don't have their unique addresses, then they won't be read via SNMP applications. But, I have just started working with that, so there may be something that I am missing.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top