From: kevin gannon (kevin@gannons.net)
Date: Sat Oct 01 2005 - 10:29:42 GMT-3
I have no real world SNA experience so the below maybe seem
daft. I have setup up the following
R1
------------------
interface FastEthernet0/0
no ip address
duplex auto
speed auto
dlsw transparent redundancy-enable 9999.9999.9999
dlsw transparent map local-mac 4000.0000.0099 remote-mac 0009.b7d1.6e20
R5
-------------------
interface FastEthernet0/0
no ip address
duplex auto
speed auto
dlsw transparent redundancy-enable 9999.9999.9999 master-priority 1
dlsw transparent map local-mac 4000.0000.0002 remote-mac 0090.ed8b.7604
Both R1 and R5 are attached to the same switch in the same VLAN. On this
VLAN I have also setup a DSPU listener and on another router I have setup
a DSPU host. The DSPU traffic is travelling in DLSW and hits R1 in this case
as this is the router which won the circuit. 0090.ed8b.7604 is the host by the
way.
I can see the MAC address translation taking place on R1 :
R1#
*Mar 1 03:38:42.779: DLSW-ER:action_c(): origin mapped from (wan)
0090.ed8b.7604 ---> 0200.0000.0099
So all is well.
So in the real world if the mainframe recieves the SNA from the MAT'd
address in my
case 0200.0000.0099 the MAT it should respond to the 0200.0000.0099 ?
If this is the case then in my configs MAT seems to work in only one direction.
In fact if R1 receieves an explorer for the MAC 0200.0000.0099 it doesnt
respond and traffic is dropped.Testing using DSPU commands again a
real SNA host may behave differently ?
Maybe I should be thinking about this like static IP NAT.
Thanks & Regards
Kevin
This archive was generated by hypermail 2.1.4 : Sun Nov 06 2005 - 22:00:49 GMT-3