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

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

ACE:How to bind more VIPs' status, one VIP down, others also down

As we know, a WEB2.0 business is consist of more services, and those services communicte with each other.

If one service shutdown, and failover to another DataCenter, other services must talk with it cross datacenter, that will reduce the whole system's performance, so we need bundle failover those services.

One service has a VIP, my question is how to bind those services' VIP status.

I expect if one VIP down, the other VIP also down.

I appreciate your help.




Re: ACE:How to bind more VIPs' status, one VIP down, others also

Hi Smith,

There may be more than one way to accomplish this and I'm not sure which would be the best way, but here is one that comes to mind:

VIP-A  -->  Serverfarm-A --> Rserver-A

VIP-B  -->  Serverfarm-B --> Rserver-B

Let's say you have VIP-A and VIP-B that you want to failover together, even if only one of them goes down at a local datacenter.  Since you can apply more than one probe to a serverfarm, and that the default behavior is that if just one probe fails the rserver is down, then you could have Serverfarm-A and Serverfarm-B both probing Rserver-A and Rserver-B.  This way, if either Rserver-A or Rserver-B application goes down, both serverfarms will go down.  Now the ACE can switch the serverfarm for both VIPs to the backup serverfarm, which would reside at the backup datacenter.

Hope this helps,


New Member

Re: ACE:How to bind more VIPs' status, one VIP down, others also


Thank your help.

The method your provided is a good method for simple usage.

But most of WEB2.0 is more complex:

1. in Local datacenter, we have more than 3 services VIPs

2. one serverfarm have more than 20 servers

3. each serverfarm has different predictor policy, or session sticky

Your method have some problems:

1. the configuration will be very complex

2. to prevent select other serverfarm's realserver, group of realservers can not share one probe.

3. if one serverfarm already config more probes, and set fail-on-all, how to implements (probeA&probeB) | probeC ?

Could you give me more suggestion?

Thanks again