07-04-2007 06:38 AM - edited 03-05-2019 05:08 PM
Hello,
I am currently trying to extend our metro Ethernet with 1841 routers equipped with a 2mft-t1 card 4esw switch card.
I am trying this using point to point frame relay with separate bridge groups for each Vlan.
Currently the config is working but spanning tree is blocking one of my t1 connections.
Please Help!
Solved! Go to Solution.
07-04-2007 07:08 AM
HI,
But instead you can try out the mfr (multilink FR).
http://www.cisco.com/en/US/products/sw/iosswrel/ps1829/products_feature_guide09186a0080087079.html
Hope it helps, rate if does
Krisztian
07-04-2007 06:56 AM
Hi,
I think it is because of the spanning tree normal behavior. There are two links between the two bridges so one of them has to be blocked in order to keep the loopfree topology.
Krisztian
07-04-2007 07:00 AM
Hello,
I believe you are correct I was wondering if any one knew of a workaround. If this was a standard bridge group I believe that I could use circuit groups.
Frame relay does not allow me to apply circuit groups.
Thanks
07-04-2007 07:08 AM
HI,
But instead you can try out the mfr (multilink FR).
http://www.cisco.com/en/US/products/sw/iosswrel/ps1829/products_feature_guide09186a0080087079.html
Hope it helps, rate if does
Krisztian
07-04-2007 07:30 AM
Hello,
Thanks for the suggestion.
I have already tried this.
I did manage to get it working.
The speed was very slow. I think it would be better if i had tweaked it.
Spanning tree was still blocking the one link. This seemed very strange to me because ppp multilink works fine when configured with bridge groups.
interface MFR1
no ip address
!
interface MFR1.1 point-to-point
ip address 10.102.102.2 255.255.255.0
frame-relay interface-dlci 1000
bridge-group 1
!
interface MFR1.2 point-to-point
ip address 10.103.103.2 255.255.255.0
frame-relay interface-dlci 300
bridge-group 2
interface Serial0/1/0:0
no ip address
ip load-sharing per-packet
encapsulation frame-relay MFR1
no arp frame-relay
!
interface Serial0/1/1:1
no ip address
ip load-sharing per-packet
encapsulation frame-relay MFR1
no arp frame-relay
router2#show spanning-tree summary
Root bridge for: Bridge group 1, Bridge group 2, VLAN300.
PortFast BPDU Guard is disabled
UplinkFast is disabled
BackboneFast is disabled
Name Blocking Listening Learning Forwarding STP Active
-------------------- -------- --------- -------- ---------- ----------
Bridge group 1 2 0 0 1 3
Bridge group 2 0 0 0 2 2
-------------------- -------- --------- -------- ---------- ----------
2 Bridge Groups 2 0 0 3 5
Name Blocking Listening Learning Forwarding STP Active
-------------------- -------- --------- -------- ---------- ----------
VLAN300 0 0 0 1 1
-------------------- -------- --------- -------- ---------- ----------
1 VLAN 0 0 0 1 1
Blocking Listening Learning Forwarding STP Active
-------------------- -------- --------- -------- ---------- ----------
Total 2 0 0 4 6
07-04-2007 08:12 AM
Hi,
Why did you configure IP address on MFR interface? It should belong to bridge-group.
Krisztian
07-04-2007 10:26 AM
Hello,
I have gone back and retried the mpp frame relay configuration without the ip addresses.
I also made some other changes.
It seems like my spanning tree issue may be ok now but my speed is very slow for the most part but does occasionally burst above 2Mbps.
07-04-2007 11:02 AM
Hi,
Let's try to turn off the load-sharing per packet but I think it does affect only the layer3 load balancing. You may try to play with the frame-relay fragment.
Krisztian
07-04-2007 11:14 AM
Hello,
I have turned off the per packet load balancing. This did not seem to make a difference.
I cannot find how to change the fragmentation setting.
07-04-2007 11:16 PM
Hi,
Try something like this:
interface MFR1.1 point-to-point
frame-relay interface-dlci 1000
class frag
bridge-group 1
!
interface MFR1.2 point-to-point
frame-relay interface-dlci 300
class frag
bridge-group 2
map-class frame-relay frag
frame-relay cir 1536
frame-relay bc 12500
frame-relay be 0
frame-relay mincir 1536
frame-relay fragment 1600
Please note that with these settings both PVCs will get 1536k equally and will be fragmented to 1600 bytes/frame. It will lower serialization delay but obviously you won't have more bandwidth :)
Hope if helps
Krisztian
07-05-2007 05:20 AM
Hello,
Thanks for your response.
I tried your suggestion but still have very slow speeds.
If I disconnect one of the t1's I get a consistant ~1.4Mbs.
07-05-2007 07:16 AM
Hi,
What does show the: "show frame-relay multilink mfr1 detailed" output?
So it looks as the mfr is not working correctly. What is the IOS release, may be it is wise to look for the IOS bug on Cisco related to mfr.
Krisztian
07-05-2007 07:46 AM
Hello
This is the show frame-relay multilink mFR 1 command on both routers. I have tried few versions of ios now. Currnetly I am running 12.4(16)advanced ip the newest one.
I have run a debug and I am getting some errors.
*Jul 5 15:03:12.011: frame_relay_input: F8812000 2B4C5B7, size= 330
*Jul 5 15:03:12.019: frame_relay_input: 48C12000 2B4C4B7, size= 330
*Jul 5 15:03:12.091: MFR1/300: Lost fragment 546 (all links out of sequence)
*Jul 5 15:03:12.491: MFR1/300: Lost fragment 57A (all links out of sequence)
*Jul 5 15:03:12.695: MFR1/300: Lost fragment 594 (all links out of sequence)
The other versions I have tried are
124-6.T7
124-13c
124-11.T2
router1# show frame-relay multilink mfR 1 detailed
Bundle: MFR1, State = up, class = A, fragmentation disabled
BID = first-bundle
No. of bundle links = 2, Peer's bundle-id = first-bundle
Rx buffer size = 24096, Lost frag timeout = 200
Bundle links:
Serial0/1/1:1, HW state = up, link state = Up, LID = Serial0/1/1:1
Cause code = none, Ack timer = 4, Hello timer = 10,
Max retry count = 2, Current count = 0,
Peer LID = Serial0/1/1:1, RTT = 0 ms
Statistics:
Add_link sent = 2, Add_link rcv'd = 1,
Add_link ack sent = 1, Add_link ack rcv'd = 2,
Add_link rej sent = 0, Add_link rej rcv'd = 0,
Remove_link sent = 0, Remove_link rcv'd = 0,
Remove_link_ack sent = 0, Remove_link_ack rcv'd = 0,
Hello sent = 951, Hello rcv'd = 949,
Hello_ack sent = 949, Hello_ack rcv'd = 942,
outgoing pak dropped = 0, incoming pak dropped = 0
Serial0/1/0:0, HW state = up, link state = Up, LID = Serial0/1/0:0
Cause code = none, Ack timer = 4, Hello timer = 10,
Max retry count = 2, Current count = 0,
Peer LID = Serial0/1/0:0, RTT = 0 ms
Statistics:
Add_link sent = 2, Add_link rcv'd = 1,
Add_link ack sent = 1, Add_link ack rcv'd = 2,
Add_link rej sent = 0, Add_link rej rcv'd = 0,
Remove_link sent = 0, Remove_link rcv'd = 0,
Remove_link_ack sent = 0, Remove_link_ack rcv'd = 0,
Hello sent = 946, Hello rcv'd = 1052,
Hello_ack sent = 1052, Hello_ack rcv'd = 946,
outgoing pak dropped = 0, incoming pak dropped = 0
router2#show frame-relay multilink mFR 1 detailed
Bundle: MFR1, State = up, class = A, fragmentation disabled
BID = first-bundle
No. of bundle links = 2, Peer's bundle-id = first-bundle
Rx buffer size = 24096, Lost frag timeout = 200
Bundle links:
Serial0/1/1:1, HW state = up, link state = Up, LID = Serial0/1/1:1
Cause code = none, Ack timer = 4, Hello timer = 10,
Max retry count = 2, Current count = 0,
Peer LID = Serial0/1/1:1, RTT = 0 ms
Statistics:
Add_link sent = 2, Add_link rcv'd = 2,
Add_link ack sent = 2, Add_link ack rcv'd = 1,
Add_link rej sent = 0, Add_link rej rcv'd = 0,
Remove_link sent = 0, Remove_link rcv'd = 0,
Remove_link_ack sent = 0, Remove_link_ack rcv'd = 0,
Hello sent = 892, Hello rcv'd = 893,
Hello_ack sent = 893, Hello_ack rcv'd = 894,
outgoing pak dropped = 0, incoming pak dropped = 0
Serial0/1/0:0, HW state = up, link state = Up, LID = Serial0/1/0:0
Cause code = none, Ack timer = 4, Hello timer = 9,
Max retry count = 3, Current count = 0,
Peer LID = Serial0/1/0:0, RTT = 0 ms
Statistics:
Add_link sent = 2, Add_link rcv'd = 2,
Add_link ack sent = 2, Add_link ack rcv'd = 1,
Add_link rej sent = 0, Add_link rej rcv'd = 0,
Remove_link sent = 0, Remove_link rcv'd = 0,
Remove_link_ack sent = 0, Remove_link_ack rcv'd = 0,
Hello sent = 991, Hello rcv'd = 891,
Hello_ack sent = 891, Hello_ack rcv'd = 991,
outgoing pak dropped = 0, incoming pak dropped = 0
07-05-2007 08:42 AM
Hi,
Are the two links from the same provider, because if the delay of the links are different problems can occur?
Krisztian
07-05-2007 08:57 AM
Hello,
This is actually a back to back link to extend our metro ethernet service.
Currently the 2 routers are on my desk and have 2 t1 crossovers connected between them.
There should be no difference in delay.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide