CSS and a Sorry Server

Unanswered Question
Jun 30th, 2008

I have been trying to get my CSS 11506 to redirct to a Sorry Server when our content servers go offline. We thought that we had it working, but after some downtime it turned out that our configuration did not work.


After extensive reading I can't figure out what is wrong with my config, or if the problem lies else where. I am attaching my config below, can anyone tell me if they see any problems with what I have or if there is something that I need to do in addition to what I have. Thank you for you help, here is the config:


*************************** GLOBAL ***************************

no restrict web-mgmt

no restrict xml

bypass persistence disable


snmp community ******read-write

snmp name "******"

snmp contact "*******r"

snmp location "CSS11056"

snmp trap-host 10.20.1.4 ******


dns primary 10.20.1.2


ftp-record ******10.20.1.17 *** des-password


ibfebcgg6aheuc4h1hfcqhpcubwdxcjb cssgui


ip route 0.0.0.0 0.0.0.0 10.20.1.1 1 !


*************************INTERFACE*************************

interface 1/1

phy 1Gbits-FD-sym !


**************************CIRCUIT**************************

circuit VLAN1

router-discovery lifetime 1000


ip address 10.20.1.4 255.255.255.0

router-discovery


**************************SERVICE**************************

service Blade01

ip address 10.20.1.60

active

service Blade02

ip address 10.20.1.61

active

service Blade03

ip address 10.20.1.62

active

service Blade04

ip address 10.20.1.63

active

service sorry

ip address 10.20.1.41

active

!*************************** OWNER***************************

owner ***

email-address ******


content Content1

vip address 10.20.1.80

balance aca

add service Blade01

add service Blade02

no persistent

primarySorryServer sorry

active


content Content2

vip address 10.20.1.81

add service Blade03

add service Blade04

balance aca

active

!*************************** GROUP***************************

group content1nat

vip address 10.20.1.80

add destination service Blade01

add destination service Blade02

add destination service sorry


group content2nat

add destination service Blade03

add destination service Blade04

vip address 10.20.1.81

!**************************** ACL ****************************

acl 10

clause 5 permit any 10.20.1.60 destination content ****


sourcegroup ****

clause 6 permit any 10.20.1.61 destination content ICC/flippid


sourcegroup Content1

clause 99 permit any any destination any

clause 2 permit any 10.0.0.0 destination content ****


sourcegroup ****

apply circuit-(VLAN1)

clause 7 permit any 10.20.1.41 destination content ****


sourcegroup Content1


  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Gregory Scarlett Wed, 07/02/2008 - 01:20

One problem I can see is that you don't have any keepalives configured under the services, so they will default to a Ping. As long as they respond to ping, it will keep traffic going to those servers.


What services run on these Servers? We generally recommend you use as higher layer keepalive as possible, so if it is a web server for example, use a HTTP keepalive.


Have a look here for more info:


http://www.cisco.com/en/US/docs/app_ntwk_services/data_center_app_services/css11500series/v8.10/configuration/content_lb/guide/KAL.html


Actions

This Discussion