Few months ago we bought Cisco 3825 router with 2 WIC-2T interface cards. Our intension was to replace old central router (Bay Networks ASN dual node stack) in company which was working since 1997. We have about 70 locations which are connected via 3 Frame Relay ports (2 MB CIR, 2 MB throughput)in classic hub and spoke toplogy. On locations we have mix of Nortel and Cisco routers (most are Nortel AN and Passport 2430 and some Cisco 805). All three ports are multipoint and that worked fine with old router. Most of locations on problematic FR port have 8k CIR PVC with 128k throughput, some are 32k and 64k and couple are with 256k CIR.
I configured Cisco in the same manner as ASN (3 multipoint interfaces) to avoid downtime and travelling to locations. After replacement with Cisco 3825 some odd things begun to happen:
- locations on two ports (0/0/0 and 0/1/0) are working OK most of time but locations on port 0/1/1 have problems. On start up all seems to be alright but after some time on few locations response time rise to 5-9 SECONDS making that links unusable. After some random time response time drops to normal but same thing happens on some other locations on that port.
Port utilisation is between 20% - 45%. I noticed large number of discarded and delayed packets and that problem mostly happens on locations with 8k CIR but not exclusive on them. Sometimes (not quite often) happens with locations on 64k CIR. As I can see traffic shaping is not dropping packets and interface has no errors.
Show process CPU gives arround 1% occupation.
Only difference between that 3 ports is that problematic one is not connected to Nortel FR switch but on some different equipment. FR provider doesn't have any record of errors and claims that everything is OK.
I tried with and without adaptive shaping (BECN and interface congestion and only traffic rate), moved port from 0/0/1 to 0/1/1 (second port on second interface card), changed X.21 cable. In last attempt I used static mapping for DLCIs and I have run ot of ideas.
When I put Nortel Passport 2430 router on that port problems dissapear, which confirms claims from FR provider that their equipment is OK.
On first port simmilar thing happens but very rare (once in 4-5 hours) and usually with one location.
I know that Cisco recomends point-to-point and that multipoint can cause problems but I can't believe that Nortel is working without any problem in same topology.
I attached part of Cisco config and show for ser0/1/1 and 0/0/0
can you, if possible, find out what brand of frame relay switch your provider is using on that port ? I have a feeling that your problem is connected to the interaction between the Cisco and the ´unknown´ frame switch...
I found out that some kind of server is doing FR switching! FR provider also claims that they have several companies connected on that "FR switch" which have Cisco equipment and that they work without any problems. They informed me that this server is due to be replaced in few months...
[toc:faq]The ProblemOn traditional switches whenever we have a trunk
interface we use the VLAN tag to demultiplex the VLANs. The switch needs
to determine which MAC Address table to look in for a forwarding
decision. To do this we require the switch to do...
[toc:faq]Introduction:Netdr is a tool available on a RSP720, Sup720 or
Sup32 that allows one to capture packets on the RP or SP inband. The
netdr command can be used to capture both Tx and Rx packets in the
software switching path. This is not a substitut...
IntroductionOSPF, being a link-state protocol, allows for every router
in the network to know of every link and OSPF speaker in the entire
network. From this picture each router independently runs the Shortest
Path First (SPF) algorithm to determine the b...