From: Kyle Galusha (kgalusha@xxxxxxxxx)
Date: Fri Feb 02 2001 - 20:37:57 GMT-3
I could use another set of eyes on this.
I have configure two dynamic dlsw peers. R1 is one peer and r8 is another
peer. R3 is an intermediate router that connects to r1 via frame and r8 via
ISDN. I have OSPF as the routing protocol with ip ospf demand between R3 and
R8. I am running netbios from Win95 machines on the ethernet segments on r1
and r8. The routers peer fine and by using "find computer" on the win95 PCs I
can get a DLSW circuit established.
So it looks like this
PC----ether---R8---isdn---R3----f/r------R1---ether--PC
The problem is my isdn link stays up (it shouldn't) after the circuit goes
down. OSPF hellos or LSAs are not the problem. It looks like a DLSW tcp
(2065) session keepalive is keeping the session up but I've configure keep of
0 on both peers.
Any thoughts?
Kyle
Config and Debugs follow.
r8#sho dlsw cir (this is normal)
Index local addr(lsap) remote addr(dsap) state uptime
1342177382 0003.f20e.0277(F0) 0003.f226.652c(F0) CONNECTED 00:00:19
Total number of circuits connected: 1
r8#sho dlsw peer (this is normal)
Peers: state pkts_rx pkts_tx type drops ckts TCP
uptime
TCP 144.16.1.1 CONNECT 90 72 dynam 0 1 0 00:02:47
Total number of connected peers: 1
*******************
selected parts on r8s config
r8
dlsw local-peer peer-id 144.16.1.8
dlsw remote-peer 0 tcp 144.16.1.1 keepalive 0 timeout 90 dynamic (this should
stop keeps)
dlsw bridge-group 1
!
!
interface FastEthernet2/0
ip address 144.16.8.8 255.255.255.0
no ip directed-broadcast
bridge-group 1
!
bridge 1 protocol ieee
************************
Selected parts of r1s config
!
dlsw local-peer peer-id 144.16.1.1
dlsw remote-peer 0 tcp 144.16.1.2 priority cost 5
dlsw remote-peer 0 tcp 144.16.1.8 keepalive 0 timeout 90 dynamic no-llc 1
dlsw bridge-group 1
!
interface FastEthernet0/0
ip address 144.16.11.1 255.255.255.0
no ip directed-broadcast
ip pim sparse-mode
ip igmp join-group 224.10.10.10
duplex auto
speed auto
bridge-group 1
!
bridge 1 protocol ieee
***********************
r8#sho tcp bri (this tcp session should drop but does not)
TCB Local Address Foreign Address (state)
61101228 144.16.1.8.11128 144.16.1.1.2065 ESTAB
r8# debug dialer pack This shows dlsw is keeping dialer up but why?
06:47:16: IP: s=144.16.1.8 (local), d=144.16.1.1 (Dialer0), len 40, sending
06:47:16: TCP src=2065, dst=11324, seq=3966165290, ack=0, win=20004 RST
06:47:16: IP: s=144.16.1.1 (Dialer0), d=144.16.1.8, len 40, rcvd 4
06:47:16: TCP src=2065, dst=11128, seq=3847272988, ack=3966124550, win=2010
2
ACK
r8#sho dlsw pee (after circuit drops connect stays up 120 seconds then drops
then reestablsihes
repeatedly???? This connection should stay down until needed.
Peers: state pkts_rx pkts_tx type drops ckts TCP
uptime
TCP 144.16.1.1 CONNECT 17 14 dynam 0 0 0 00:00:57
Total number of connected peers: 1
Total number of connections: 1
debug dlsw peer (looks like it wants to disconnect but doesn't)
06:42:39: DLSw: START-TPFSM (peer 144.16.1.1(2065)): event:CORE-DELETE CIRCUIT
s
tate:CONNECT
06:42:39: DLSw: dtp_action_v(), peer delete circuit for peer 144.16.1.1(2065)
06:42:39: DLSw: END-TPFSM (peer 144.16.1.1(2065)): state:CONNECT->CONNECT
Kyle Galusha
System Engineer
Cisco Systems, Inc.
5757 Corporate Blvd
Suite 102
Baton Rouge, LA 70808
(225) 218-2001
(225) 218-2055 FAX
(800) 365 4578 Pager
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:28:34 GMT-3