Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

CSS ACTIVE/ACTIVE SCENARIO WITH JUST TWO SERVERS ??? POSSIBLE??

Hi

I'm gonna have a setup of TWO CSS11503 Content Switches with standard WEBNS feature set

in an ACTIVE / ACTIVE VIP and Virtual interface redundancy scenario for load-balancing just

two web servers initially.

Can I hv this setup up & running if I configure the two servers with different default

gateway addresses on the private side and two static routes in the private side Layer3

for two different VIP addresses in the public side ??

Any better suggestions for this scenario.

Thanx

3 REPLIES
Cisco Employee

Re: CSS ACTIVE/ACTIVE SCENARIO WITH JUST TWO SERVERS ??? POSSIBL

Why do you want active/active ?

What do you expect from this ?

What are the requirements for the design ?

Do you have 2 NICs on the servers ?

How many websites/Virtual IP ?

I would personally not go for active/active but active standby or box-to-box redundancy

Regards,

Gilles.

New Member

Re: CSS ACTIVE/ACTIVE SCENARIO WITH JUST TWO SERVERS ??? POSSIBL

Hi Gillies

thanx for your feedback....

The current client requirement is to have ACTIVE / ACTIVE CSS Scenario for

load balancing their current intranet report servers (two nos).

The client has seen the ACTIVE / ACTIVE design from

the website and prefers the same for two servers currently.

Yes.... Both the Servers have DUAL NIC to connect to two L2 switches in the

server side.

Currently therez only one Virtual IP address...

BR

Rajesh

New Member

Re: CSS ACTIVE/ACTIVE SCENARIO WITH JUST TWO SERVERS ??? POSSIBL

Firstly - what Gilles said.

Having said that, I'm using some content switches in active/active modes in a couple of places in a geographically distributed gateway. Active/Active lets us improve our redundancy characteristics and allow for device failures as well as link failures between the gateways.

There are lots of complexities that arise if you take this path - you will need to do a lot of logical math and testing about traffic symmetry under all of the different failure conditions, because you introduce the possibility that response traffic could come back at L2/L3 through a different CSS than the request traffic.

129
Views
0
Helpful
3
Replies