Similiar Setups - Different Results

[BudT]

We have two branch offices, both utilize frame relay of 512kb with a CIR of 128kb. the they both have the same type routers, Cisco 2501, the same CSU, Astrocom SP100, both have similiar configurations, but one circuit is relatively slower than the other. Example, the average ping time on one is 46ms where the other is 84ms, almost double. Both terminate at the same location on different routers. I have had the telco vendor investigate the circuit but all they say is that we are only using part of what is allocated CIR and find nothing wrong. The people at the slower location are begining to notice the slowness and are starting to complain. I have checked everything I can think of. Any suggestions would be greatly appreciated.

 

[jeter]

You said they are program almost the same so I would look elsewhere.

Did you do a trace route from the router?
From the pc's? Is there a difference between the two?
The Telco company can run a test on your line but they will only go to the CIR 128k. They may advise to increase the bandwidth. Did you test your network for any problems such as broadcasts? “Reserve your right to think, for even to think wrongly is better than not to think at all”

Fisher CCNA

 

[tadiman]

Where are these two brach offices? If one is the same town and the other is 3,000 miles and four or five carriers away then you will see a difference. If the users are seeing a difference than something is pounding traffic on it, broadcasts, misconfigured port, etc. You say that the two routers "both have similiar configurations". The word similar scares me . Something there might be the problem.

Your port speed is 512k and the cir is 128k. What that means is that any traffic that exceeds 128k is now "DE" or discard eligable. If the carrier has a bottle neck in their network they can drop these DE packets. The local loop will test fine but the problem is beyond the local loop. This might explain the users noticing speed difference, but not necessarily the ping problem. If you want to post the interface config and describe the layout of this WAN network then maybe we can figure something out from there.

 

[BudT]

I say similiar, meaning thedy are the same except the IP addresses and router names. The good office is in Chicago and the other one is in Aspen, CO. Main office is in DC. From what I can see we never reach the 128KB CIR at any given point. I have tested the network there and find nothing out of the ordinary, we have even upgrade to a switched network. Here is the router configuration:

AspenCO#sho conf
Using 1253 out of 32762 bytes
!
version 11.3
service timestamps debug uptime
service timestamps log uptime
no service password-encryption
service udp-small-servers
service tcp-small-servers
!
hostname AspenCO
!
enable password XXXXXXX
!
appletalk routing
!
!
interface Ethernet0
description Ethernet Port
ip address 206.0.102.2 255.255.255.0
no ip directed-broadcast
no ip route-cache
no ip mroute-cache
appletalk cable-range 8000-8099 8000.154
appletalk zone Aspen CO
no mop enabled
!
interface Serial0
description 512K Frame Relay to Aspen DC
ip address 206.0.105.2 255.255.255.0
no ip directed-broadcast
encapsulation frame-relay
no ip route-cache
no ip mroute-cache
bandwidth 512
appletalk cable-range 2-2 2.10
appletalk zone Aspen CO
frame-relay map appletalk 2.2 100 broadcast
frame-relay lmi-type ansi
!
interface Serial1
description Not Currently Used
no ip address
no ip directed-broadcast
no ip route-cache
no ip mroute-cache
shutdown
!
router rip
network 206.0.102.0
network 206.0.105.0
!
ip classless
ip route 0.0.0.0 0.0.0.0 205.0.105.1
!
snmp-server community public RO
snmp-server host XXX.X.XXX.XXX
!
line con 0
exec-timeout 0 0
line aux 0
transport input all
line vty 0 4
password XXXXXXXXXX
login
!
ntp server192.168.1.28
end

 

[tadiman]

Initially I would have to say the problem is the distance. While the same carrier has the traffic (obviously ) they may route through different pops after the traffic hits the initial frame switch. Since the transport network is frame relay, trace route won't do any good to distinguish a carrier problem. However like jeter wrote in his (or her) reply a trace route from the router and then from a pc attached will indicate a problem in either ends network. If those results are similar than the problem is in the wan. Good luck getting anything done there. And it could be normal latency for the distance!!

Here's an example for you. My company is located in central PA. We have an office in Atlanta GA that pings at an average of 44 msec. Then we have an office in central IN that pings at 56 msec average. Now by the way of the crow the Indiana location is 200 miles closer!!!! The issuse is more than likely latency introduced in the frame providers network.

Hope any of this helped.

 

[jeter]

Post a copy of ( show int).

Also have you debug the interfaces?
Be careful with debugging it can and will overload your router so as a rule I turn only one on at a time depending on the router.

Have you checked the network in Aspen? “Reserve your right to think, for even to think wrongly is better than not to think at all”

Fisher CCNA

 

[BudT]

Telco vendor says that everything is correct on their end. How can i set the CIR on a Cisco 2501 router or tell what the CIR is?

 

[TTND]

Have you asked the Frame Relay provider what the route mileage is from your premise to the access switch? Many times this will be extended if the closest access switch is at full capacity. I have seen customers that have been long hauled 3000 miles to go from one location in NY to another location in NY. They can provide this information for you. Also, is the difference in delay being verified from the site router serial interface on side A to the site router serial interface on side B? This would prove if the problem was on your LAN or possibly the WAN and even possible to be the CPU utilization on the routers. The network providers can also ping across their network and provide you with the delay readings. Another thing, do you see any DE packets being dropped?
Hopefully that will get you started. First off though, I would get the provider to give you the milage differences and ping across their network.