Bridge_Aggregation and loop

[VinodGupta]

Hi

I have 4(A,B,C,D say) HP_5900 Switch, Sw A & B are in Stack_1 and C&D are another Stack_2(IRF)
Now we need to make connectivity between Stack_Switch1(A&B) to Stack_2(C&D) for that 2 Bridge_Agg created (45 & 46) at both end and passing common Vlan(12) on all interface associated with Bridge_A(45&46)>>>>>Vlan_interface 12 is created and assign a /30 ip

will this scenario work, without any Port(Bre_AGG) blocking or this will create l2_loop,
what happen if "undo stp globle enable" used 

=======================COnfigurtion of Stack1 is below_similar used at other end============
vlan 12

interface Vlan-interface12
ip address 10.19.27.66 255.255.255.252

#
interface Bridge-Aggregation45
port link-type trunk
undo port trunk permit vlan 1
port trunk permit vlan 12
link-aggregation mode dynamic
#
interface Bridge-Aggregation46
port link-type trunk
undo port trunk permit vlan 1
port trunk permit vlan 12
link-aggregation mode dynamic
#

interface GigabitEthernet1/0/45 & 2/0/45
port link-mode bridge
port link-type trunk
undo port trunk permit vlan 1
port trunk permit vlan 12
port link-aggregation group 45
#
interface GigabitEthernet1/0/46 & 2/0/46
port link-mode bridge
port link-type trunk
undo port trunk permit vlan 1
port trunk permit vlan 12
port link-aggregation group 46
#

[parnassus]

Two BAGGs (per IRF) interconnecting two different IRF deployments, both carrying the same VLANs, are going to form a loop...something to avoid.

Why two BAGGs? Why - instead - not using a single BAGG on each IRF deployment made of four member ports? that way you will have both IRF interconnected with four links all well distributed (link 1 from A to C, link 2 from A to D, link 3 from B to C and link 4 from B to D).

[VinodGupta]

Hi Parnassus

Thanks for your valuable reply

The scenario mention by me as above came from Planning team 

we have testing above in lab with two BAGG with same VLAN

case 1:- when "STP global enable" we found that at one side 1 BAGGs are in ALT state>>>>that is OK and I was expecting the result result>>>>means STP is working properly

Case2:-BUT when we disable STP at one side by "undo STP global enable" >>>I was expected there should be loop>>>but result is different i.e we not notice any loop both the BAGGs working fine and Other side Both BAGGs are showing in forwarding state >>>>unable to find where is GAP, why loop is not created

[Vince-Whirlwind]

As Parnassus says, you should be creating one BAGG with all 4 ports in it.
Allowing STP to block ports is something you do when you don't have BAGG functionality.

As far as the loop not occurring goes - check your switch log when you make the second BAGG switchports active to see what's happening.