Raghavendra - why do you need to run your application on the PI server?
And how to you achieve that both computers have the same IP?
AFAIK it is not possible to have two computers on the same network with the same IP at the same time (a Microsoft Cluster does respond on a single IP, but that is in fact one computer owning the IP at a time) – and both computers need to be connected to form the collective – resulting in an IP conflict. So basically this translates for me to a network question rather than a PI question.
BTW, playing around with the IP on a production PI is not good practice, this could result in data loss.
Thanks for the response
here is our hardware configuration
Two windows servers hosted with -
- PI & AF Servers configured in collective
- SQL Server DB configured in cluster OR mirroring
- custom services
Now for custom services we need to achieve single IP address so that external apps can talk to this services.
If we configure Network Load Balance with Virtual IP adress for external application to talk to custom services.
Will there be any issue for PI & AF if NLB is configured.
As far as I know, the current PI High Availability platform is designed so that each member of a PI Collective has necessarily a different IP address. The same is true for the AF High Availability platform.
Would the installation of your custom services on computers separated from the PI+AF boxes be option in your case?
Thanks for the response.
We are planning to have two IP Address per server one will be a Physical IP Address and the other will be a Virtual IP Address.
In NLB we will be using Virtual IP Address and this will be same for both the servers, whereas for configuring PI & AF Collective we will be using Physical Address of the servers.
With this sort of environment will PI and AF Server work seamlessly without any conflicts?