Bad Eth card on 2620?

[datadan]

I have a 2620 where the ethernet stopped working after a facility move (we moved switches/cable/equipment/ ip subnet stayed the same). Things came up incrementally over the course of a weekend and it failed after 1-2 days of working. How can I tell if the card is bad or if I have a configuration issue?

I can console in and
show int
doesn't reveal any obvious errors.

Any assistance is appreciated.
Thanks,

 

[wybnormal]

normally there will something in the logs or counters that will give it away. Look at CRC errors.. sizing errors.. runt packets etc. I'm assuming you have run the "show interface" and the port shows up? Can you ping out of the port using an extended ping? Have you put a sniffer on the port? What about running a debug on the port?

There are many ways to look at the port and deciper if the port is gonzo.

MikeS


Find me at

http://www.packetattack.com

"Take advantage of the enemy's unreadiness, make your way by unexpected routes, and attack unguarded spots."
Sun Tzu

 

[datadan]

wybnormal:

Thanks for the input. I have a 2620[10.10.12.2] (1 eth 1 csudsu) and a 2600[10.10.10.10 & temporarily 10.10.12.2] (2 eth).

I am using the 2600 2nd ethernet card 'cause I can't get the 2620 eth to work.

When I move the cable from swith into eth 0/1 (of 2600[10.10.12.2]) into the 2620 eth 0/0 I get the following error on the console for ~2 minutes:

1w1d IP-EIGRP neighbor 10.10.10.10 not on common subnet for Fast Ethernet 0/0.

show int shows line and protocol up.

I can't ping beyond the router.
WHat is an extended ping?
When you say port are your referring to the cisco port or the 3com switch port?
How do I run a debug on the port?
THanks,

 

[datadan]

I found this article:

http://www.cisco.com/en/US/tech/tk3...h_note09186a0080094613.shtml#trouble_eigrp_nc

However after following steps I can't get the routers to see eachother unless I put a secondary route in for each router for the other subnet.

Any ideas?

How do I reset EIGRP tables (reset restart?), I have powered off and on numberous time with no effect.

 

[wybnormal]

clear ip eigrp neighbors

The extended ping is easy... just type in PING and then follow the prompts. There are quite a few options to choose from and you can force the PING out a certain interface which is very handy for troubleshooting.

MikeS


Find me at

http://www.packetattack.com

"Take advantage of the enemy's unreadiness, make your way by unexpected routes, and attack unguarded spots."
Sun Tzu

 

[datadan]

When I walk through the prompts I get to source interface.

It only let's me type an ip which I type 10.10.12.2
Type of service I choose the default = 0
Validate Data Reply = Yes
Data Pattern = 0xABCD
Verbose
Sweep range of sizes = N

This works without a problem, I suspect because i have a secondary IP on ethernet 0/0.

Is specifying the 10.10.12.2 enough to ensure that the ping goes through that interface?

Is there some EIGRP limitation that prevents ethernets plugged into the same switch from seeing eachother?

Can someone point me to a

http://www link

that can walk me through turning off EIGRP and turnning on RIP?


Thanks,

 

[gaveeve]

I'm confused about the setup. Do you have 2 different subnets going into the same switch without vlans? How are your 2 routers communicating? Through the switch?
To answer the following question "Is there some EIGRP limitation that prevents ethernets plugged into the same switch from seeing each other?"
If you have the eigrp processes labeled with different #s on each router they will need to import routes to see each other.
Do a show run: router eigrp 100(or some other #) will come up. The number should be the same on each router.

 

[datadan]

Yes I have two subnets in one building.

Two routers do indeed plug into the same switch (12port 3com 3300 superstack).

Vlans? I've had this configuration for 2-3 years without being concerned about Vlans, why would I need to concern myself with them now? The switch has always handled this before. VLAN 802.1Q learning is enabled. 802.1p is disabled. IGMP Multicast learning is disabled.

All EIGRP numbers are the same on all routers throughout the network. I only unplugged everything in location A and plugged it all back in at location B.