Breakout switches and IRB don't play well together

I've discovered another possible caveat when using dynamips for your labs, at least when using a breakout switch.  If you're using locally connected interfaces, or not using real switches it's not going to bite you though.

While working IPExpert Volume 1 Workbook Lab 7 which is a frame relay based lab focusing on bridging, I found that when using Dynamips IRB on ethernet interfaces doesn't work as expected when you're using a 3550 as your breakout switch.

This may be related to the one way CDP "feature" due to the 3550 not supporting l2protocol tunnelling on trunk port.

I found it confusing at the time since the bridging was working fine over the frame cloud but not over the ethernet.

This was the irb config I had on R2:

hostname R2

bridge 1 protocol ieee
bridge 1 route ip

bridge irb

interface Serial2/1.205 point-to-point

 description to R5 via Frame
 frame-relay interface-dlci 205
 bridge-group 1
!

interface BVI1
 ip address 150.100.221.2 255.255.255.0
!

This was the irb config I had on R5:

hostname R5

bridge 1 protocol ieee
bridge 1 route ip

bridge irb

interface FastEthernet0/1

 description to R6 and R7 via Cat3
 no ip address
 duplex auto
 speed auto
 bridge-group 1
!
interface Serial2/1.502 point-to-point

 description to R2 via Frame
 frame-relay interface-dlci 5012
 bridge-group 1
!

interface BVI1
 ip address 150.100.221.5 255.255.255.0
!

R6 and R7 are routers with interfaces on directly attached to the 150.100.221.0/24 subnet.

In the end I installed one of the many USB200M Ethernet Adapters I still have before going the breakout switch route in my Ubuntu Server - which was automatically installed as eth2, I edited the lab .net file to have R5 Fa0/1 use eth2, patched it to Cat3 replacing the feed from the breakout switch and fired things up and obtained the expected connectivity.

R2#ping 150.100.221.5

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 150.100.221.5, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 4/11/36 ms
R2#ping 150.100.221.6

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 150.100.221.6, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 4/10/24 ms
R2#ping 150.100.221.7

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 150.100.221.7, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 8/11/20 ms
R2#show bridge

Total of 300 station blocks, 297 free
Codes: P - permanent, S - self

Bridge Group 1:

    Address       Action   Interface       Age   RX count   TX count
ca05.06f6.0006   forward   Serial2/1.205     0          6          6
ca03.06f6.0006   forward   Serial2/1.205     0          5          4
ca04.06f6.0006   forward   Serial2/1.205     2          6          5

R2#sh arp
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  150.100.221.2           -   0000.0c81.46a0  ARPA   BVI1
Internet  150.100.221.5          46   ca03.06f6.0006  ARPA   BVI1
Internet  150.100.221.6          46   ca04.06f6.0006  ARPA   BVI1
Internet  150.100.221.7          25   ca05.06f6.0006  ARPA   BVI1
 
R5#sh bridge
Total of 300 station blocks, 297 free
Codes: P - permanent, S - self

Bridge Group 1:

    Address       Action   Interface       Age   RX count   TX count
ca04.06f6.0006   forward   FastEthernet0/1   0          6          5
ca05.06f6.0006   forward   FastEthernet0/1   0          6          5
0000.0c81.46a0   forward   Serial2/1.502     0         15         15
 
R5#sh arp
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  150.100.220.5           -   ca03.06f6.0008  ARPA   FastEthernet0/0
Internet  150.100.221.2          24   0000.0c81.46a0  ARPA   BVI1
Internet  150.100.221.5           -   ca03.06f6.0006  ARPA   BVI1
Internet  150.100.221.6          45   ca04.06f6.0006  ARPA   BVI1
Internet  150.100.221.7          24   ca05.06f6.0006  ARPA   BVI1


Not necessarily a show stopper but something to be aware of when you're doing a lab and things appear to be correctly configured but aren't working.  It's good to know that I'm not alone with making this observation.  There's something certainly to be said for having plenty of interfaces directly attached to your dynamips box and not needing the breakout switch.