12-14-2003 06:36 AM
I'm currently having some problems getting a VRRP failover config to work. Basically,i've got three VLAN's, each of which has the same critical service for their respective VRIDs. This critical service only monitors the internal switch to which the CSS is directly connected.
When the master CSS is disconnected from the internal switch, the internal VRIDs are correctly failing over, but the externals are getting in some kind of loop..going from Master to Backup infinitely...
i'd be grateful for any help!
Here is some sample output from the master:
CSS11050# sh chas
Configuration for CSS 11050:
Name: CSS 11050 SW Version: 5.00 Build 105
HW Major Version: H0 HW Minor Version: 0
Serial Number: JAB0724059H
Base Mac Address: 00-0c-ce-13-fe-9f
Module Number Module Name Status
1 FEM primary
2 GENIC primary
5 SCFM-PLUS primary
Port Number Port Name Status
1 e1 online
2 e2 online
3 e3 online
4 e4 online
5 e5 online
6 e6 online
7 e7 online
8 e8 online
--More--- 9 e9 online
CSS11050# sh ver
Version: ap0500105 (5.00 Build 105)
Flash (Locked): 5.00 Build 105
Flash (Operational): 5.00 Build 105
Type: PRIMARY
Licensed Cmd Set(s): Standard Feature Set
I tried to look for a bug in the IOS version, but i
could not find anything in particular....
any pointers ?
12-14-2003 11:44 AM
can you give us the config part with the redundancy
and the config of the critical services.
Also, what's the status of the critical services after failover.
[we need both side].
Gilles.
12-14-2003 02:29 PM
please see below for the redundancy config.. sorry, i masked the ip addresses partly because these might be sensitive.. i can mail you the complete config...
Status after failover of the critical service is flipping (recalling from memory now) from alive to dying to down to alive..
this is on the master:
!************************** CIRCUIT **************************
circuit VLAN1
description "Internal LAN"
ip address 10.a.b.102 255.255.255.128
ip virtual-router 1 priority 240 preempt
ip redundant-interface 1 10.a.b.126
--More--- ip redundant-vip 1 10.a.b.100
ip critical-service 1 VRRP-KAL
ip address 10.a.b.189 255.255.255.192
ip virtual-router 4 priority 240 preempt
ip redundant-interface 4 10.a.b.190
ip redundant-vip 4 10.a.b.186
ip critical-service 4 VRRP-KAL
ip address 10.a.b.253 255.255.255.192
ip virtual-router 3 priority 240 preempt
ip redundant-interface 3 10.a.b.254
ip critical-service 3 VRRP-KAL
circuit VLAN2
description "Operator LAN"
ip address 10.a.c.4 255.255.255.240
ip virtual-router 2 priority 240 preempt
ip redundant-interface 2 10.a.c.6
ip critical-service 2 VRRP-KAL
circuit VLAN3
description "SMSC and management LAN"
--More---
ip address 10.a.d.102 255.255.255.192
ip virtual-router 5 priority 240 preempt
ip redundant-interface 5 10.a.d.100
ip critical-service 5 VRRP-KAL
service VRRP-KAL
ip address 10.a.b.111
keepalive type script ap-kal-pinglist "10.a.b.111"
active
This is on the slave:
circuit VLAN1
description "Internal LAN"
ip address 10.a.b.101 255.255.255.128
ip virtual-router 1 priority 230
ip redundant-interface 1 10.a.b.126
ip redundant-vip 1 10.a.b.100
ip critical-service 1 VRRP-KAL
--More---
ip address 10.a.b.188 255.255.255.192
ip virtual-router 4 priority 230
ip redundant-interface 4 10.a.b.190
ip redundant-vip 4 10.a.b.186
ip critical-service 4 VRRP-KAL
ip address 10.a.b.252 255.255.255.192
ip virtual-router 3 priority 230
ip redundant-interface 3 10.a.b.254
ip critical-service 3 VRRP-KAL
circuit VLAN2
description "Operator LAN"
ip address 10.a.c.5 255.255.255.240
ip virtual-router 2 priority 230
ip redundant-interface 2 10.a.c.6
ip critical-service 2 VRRP-KAL
circuit VLAN3
description "SMSC and management LAN"
ip address 10.a.d.101 255.255.255.192
--More--- ip virtual-router 5 priority 230
ip redundant-interface 5 10.a.d.100
ip critical-service 5 VRRP-KAL
service VRRP-KAL
ip address 10.a.b.110
keepalive type script ap-kal-pinglist "10.a.b.110"
active
The .110 and .111 are the 2950 switches to which the CSS's are directly connected.
Is this sufficient ?
br,
Roland
12-19-2003 01:50 AM
We figured it out. The switches were also reachable through the exernal network. Consequently, the internal VRID was going slave, and the external ones where telling to go master again...
solved by adding a route to blackhole...
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