deano Posted January 22, 2014 Posted January 22, 2014 Hi all, Just need to query something regarding 2012 RDS farms and thin client devices. I have create a new session collection in my RDS environment and added two session host servers. These are appearing in the properties for the collection under load balancing. I have one session host server to use but what address would the thin client machines use to connect to the farm and where is it setup? Do i need to create DNS entries for the farm name with each IP of the session host servers? But would this not bypass the session broker?
maark Posted January 22, 2014 Posted January 22, 2014 Am setting up similar here for sims web apps. Set it up on one server and added two session servers - I let them get addresses from DHCP - did not configure anything in DNS and when you connect to the name of the first one then it automatically uses the next session server for each user accessing the remote app.
localzuk Posted January 22, 2014 Posted January 22, 2014 (edited) Best way to do it is with DNS records (same name for each server, pointing at each IP) - that way the initial 'landing' before the RDS load balancer kicks in is equally spread as well as the actual RDS load. In reality, the amount of load this causes will be very minor in a school environment. It also means you have a level of failover too, otherwise if the 'main' server goes down, you lose your entire farm, but if you have an RDS record specific to the farm it will be easy to remove the faulty IP address from it until it is fixed. Edited January 22, 2014 by localzuk
deano Posted January 22, 2014 Author Posted January 22, 2014 Best way to do it is with DNS records (same name for each server, pointing at each IP) - that way the initial 'landing' before the RDS load balancer kicks in is equally spread as well as the actual RDS load. In reality, the amount of load this causes will be very minor in a school environment. It also means you have a level of failover too, otherwise if the 'main' server goes down, you lose your entire farm, but if you have an RDS record specific to the farm it will be easy to remove the faulty IP address from it until it is fixed. So on the thin client machines you would point them at the DNS name that i create for the servers. Would this not bypass the broker?
localzuk Posted January 22, 2014 Posted January 22, 2014 (edited) RDS Brokering is done silently when you connect to an RDS server, the client doesn't know about the broker at all. This is our setup for student RDS servers: RDSSB - Session Broker RDS1 - RDS Server RDS2 - RDS Server RDSFarm - CNAME DNS record to RDS1 RDSFarm - CNAME DNS record to RDS2 RDS1 and RDS2 set up to use RDSSB as the load balancer/broker. Works perfectly. Edited January 22, 2014 by localzuk 1
deano Posted January 22, 2014 Author Posted January 22, 2014 RDS Brokering is done silently when you connect to an RDS server, the client doesn't know about the broker at all. This is our setup for student RDS servers: RDSSB - Session Broker RDS1 - RDS Server RDS2 - RDS Server RDSFarm - CNAME DNS record to RDS1 RDSFarm - CNAME DNS record to RDS2 RDS1 and RDS2 set up to use RDSSB as the load balancer/broker. Works perfectly. When you say setup to use broker, you mean just the initial configuration of the session collection. No other configuration required somewhere
TheScarfedOne Posted January 22, 2014 Posted January 22, 2014 Your session hosts need to be told from the session host config that they are part of a farm. That's where the farm name is specified. You also need to install the connection broker role on your connection broker server
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now