Hello there, I need to do this in my test environment and I want to make sure it is correct. I also need some help.
Currently, I have four servers. I will configure two servers as UI hosts, and the other two as Batch hosts.
For the UI hosts, I will configure a load balancer with sticky sessions. I will not change the ServiceDefinition for the batch hosts — they will run without a load balancer.
Then, from the debug page, I will update the ServiceDefinition for the Request Scheduler and Task Scheduler on both batch servers so they run as batch servers.
My question is: Will this make the UI servers handle UI only, or will tasks still run on them?
Hi @IslamElkhouly , We also have same kind of setup in our current environment where we have 2 UI servers and 2 host servers and we have defined them in service definition for tasks and UI. Also, we have Load balancer configured for UI servers. And SailPoint by default manages the load as well just make sure you have all the servers up and running and having same configuration in case of some connection over TLS. Also you can have one server each for Ui and task in case of DR
but you will required network team or load balancing team to point the ip to DR servers specially for Ui
hi @Samar123 After reviewing the doc, I found that I need to modify the ServiceDefinition on all four servers to include the two batch servers, as shown in the photo. By default, the other two servers will act as UI servers since they are not listed in the ServiceDefinition. I would like to confirm if this is correct
Hi @IslamElkhouly , you do not need to modify these files on each server as you will sharing the same database and change on one server will be replicated on each server. Also, as a best practice create a DNS entry something like iam.yourorganisation.com and point it to your load balancer IP then load balancer will take care to distribute loads across the UI servers.
Please mark it as solution if i am able to answer your questions