Strange Router Lockups

[someone6162]

I've got a Cisco 1811.

Pretty Simple Set. A single Cable connection WAN feeding a LAN of about 60 clients using NAT. For some strange reason, after about every 2-3 days, the router lockups up, everyone's internet stops working, and the WAN's throughput goes to 0 (nothing gets through). When in this state, I can't telnet into the router, use the GUI, or anything. It is totally locked up.

As soon as I manually go and flip the router off, and then flip it back on, everything starts working again.

I have no clue what could be causing this. Here are my questions:

1) What could be causing this to happen??

2) How do I prevent it from happening?

3) Is there a way I can program the router to reboot itself automatically when it goes enter into such state?

Thanks

 

[baddos]

Can you terminal into the unit through it's serial console port while it's "locked up"?

 

[JOAMON]

Check you default route. If it is using something like ip route 0.0.0.0 0.0.0.0 fastethernet0/1 then you may have the problem that your arp cache is growing so large as to crash the router. Use next hop router IP address as default.

 

[someone6162]

Hi Joamon, I think you are exactly right. I looked on cisco's site and found the following about ARP cache tables:

http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a00800ef7b2.shtml

I had a thorough read over it, but wasn't able to deduce the steps involved in implementing what you were suggesting.

Could you please let me know what I need to type in my CLI to implement what you mentioned? I'm a newbie

Thanks

 

[JOAMON]

Your ISP should have given you a default gateway ip address. Get rid of the one aimed at ethernet whatever by putting a no in front of the statement from cinfig mode and type :
ip route 0.0.0.0 0.0.0.0 XXX.XXX.XXX.XXX (ip gateway of ISP)

 

[someone6162]

Okay,

I checked things out. Turns out there was never a default route specified. Would the router route 0.0.0.0 0.0.0.0 to FastEthernet0 by default if nothing was specified? (I'm using simple NAT on FE1). Could the arp overflow problem still occur without a default route specified?

I added in the command ip route 0.0.0.0 0.0.0.0 xxx.xxx.xxx.xxx, where xxx is the IP address of my ISP's gateway. Internet still functions, but I guess I'll have to wait a few days to see if the router still crashes. Now, the issue is that my ISP assigns dynamic IP's. Furthermore, the gateway address changes from time to time when a new IP address is obtained. What should I do in a situation like this?

 

[JOAMON]

I would think that even though it may change occasionally your ISP should have a static gateway for you to use. I would contact them and see if they have one. You might also check and see if the following automatic import for DHCP may help out.

http://www.cisco.com/en/US/products/ps6600/products_white_paper09186a0080080106.shtml