Cisco 1720, fastethernet problems

[walxom]

I am having a rather puzzling issue with my 1720. My fast ethernet connection keeps randomly dropping. When the connection is dropped, i check the state of the fa0 interface and it states that it's "up, up". However no communication can be established between the router and the rest of the network. What's bizzare is that it only requires a reboot of the router and we're back up. Cisco replaced the router twice thinking it was a hardware defect, but apparantly after 2 replacements we've ruled that out.

This damned 1720 is the only hardware on the network that behaves this way. If for instance it was a problem with one of my hubs, I'd think other nodes on the network would experience simliar problems.

The router is config'd with NAT allowing all private connections. Any comments or tips as to why this could be happenening? Thanks!

need help..
Brad, MCSE

 

[svermill]

Don't know enough about what types of services you are implementing. If you are doing nothing fancy, then I would look at something other than the router. However, often times when rebooting is the only solution, you are overutilizing memory or the proc. Do some debug to see what is going on. I like these:

'sh proc cpu'
'sh proc mem'

These will tell you if something is killing your router. Those 1720s are great, but they are somewhat meak. For example, do you have hundreds or thousands of NAT translations that are building up and eventually sucking up all of your memory? So on and so on.

Good luck. Let us know what it turns out to be.

 

[wybnormal]

I would suspect the IOS load.. get a new load on it. I had something like this which happend only because of a bunch of random events that I happened to have happen. A updated IOS *fixed* the problem.

MikeS
Find me at

http://www.packetattack.com

"The trouble with giving up civil rights is that you never get them back"

 

[ChrisAC]

Oh yes!! I know about this one!! This happened to me today! I had set up a 1720 on a dsl connection for a two day promotional show where they need internet access for the visitors! The IT company were setting up the network and I'd already got the router up and working and was sat in one office with my laptop happily browsing the web waiting for these guys to install the servers, when .... it all stopped working!! 4pm on a Friday afternoon and it stopped working!! ARRRHHH!!! I ran up to the comms room expecting to find the router powered off or lying on the floor in pieces or something like that, but no!! Consoled in, pinged out .. no problem! Ethernet was up up!! Strange! Couldn't ping the fastethernet port with my laptop directly connected!

So, I figured that my NIC had gone tits up and so connected it to the network and pinged another machine! Success ... which meant a problem with the ethernet port on the router! DoH!! One reload later and all was well and I could breath again!! Problem is, the show starts tomorrow and it could happen again!!

I think I'll switch my mobile off this weekend!

Chris.
************************
Chris Andrew, CCNA
chrisac@gmx.co.uk
************************

 

[jeter]

Its possible that you could have a bad NIC however I would look at the way your ethernet port is setup. Auto since, 10 or 100, full or half. Change them around I have run into several problems with the router in an auto mode. Jeter@LasVegas.com
Fisher CCNA
Sprint-Data

 

[tuks]

I agree with you Jeter. I had this similar problem a month ago. The router fastethernet port was connected to a 10BaseT hub. So I force the speed to 10 and it worked.

 

[walxom]

Thanks for all your input guys.

Svrmill, it makes perfect sense. I know this becuase back when nimda took a handful of us by storm (especially those running iis) one of my servers flooded my router. The router's cpu was at 99% which caused it to crash repeatedly.
However, the problem has been resolved in that respect.

Both Cisco and us "concluded" that it infact was a defective ethernet port on the router itself. It's integrated and not modular so they replaced the entire router...TWICE!.

Hrm...There's only about 5 NAT translations. Fastehternet is running at half-dulplex, connecting to a 24 port 10/100 autosensing hub (It used to be connected to an intel 16pt switch but it crapped out on me).

In one of the offices, we were short on some wallports so they uplinked a 10bT 8 port hub for connections in that particular office. Maybe that could be a problem...I'm going to play around with duplex settings and replace that 10baseT hub. I'll let u guys know what happens. thanks again guys!

 

[webnetwiz]

I've got a 1720 that routes between 2 ethernet networks. the only way that I could get this thing to do that (to support a second ethernet interface) was to load a special version of IOS, not a GD (general deployment version) version. I got this info from Cisco after wasting 2 hours trying to figure out why the darn thing sees the 2nd interface in IOS 12.x GD, and doesn't see it at all in older versions. The resolution could be a special version of IOS.

Just my five cents...

David