Hi All,
We have the following hardware for our RD Web solution:
1 Server – RD Gateway, RD Licensing, RD Web Access, RD Connection Broker
1 Server – RD Session Host 1
1 Server – RD Session Host 2
Our environment consists of the following:
1 Windows Server 2012 R2 domain controller at the main office, 4 additional 2003 R2 servers scattered throughout our 4 remote offices. Domain functional level of 2003.
Although I have found some really good walkthroughs on the internet that have helped me for the most part in setting up an RD farm, I can’t seem to find definitive answers to my questions. My questions are mainly with the connection broker, load balancing and high availability.
I am hoping that someone could help me find the holy grail of RD web! Please avoid responding with links to technet articles as I have read several of them already. What I am instead looking for is personal knowledge and experience, thank you!
- Will high availability work with just one connection broker or do we need to purchase an additional server to use as a connection broker?
- Will my ONE connection broker automatically roll a client’s connection over to RD session host 2 in the event that RD session host 1 gets powered off?
- What exactly does high availability do? My interpretation of it from a user’s perspective is, they log on from the comfort of their living room, check their email then open and edit their word document. They decide to go jump in the shower so they close everything out without logging off. High availability now kicks in and keeps track of what the user was doing with their session when they closed and powered off their computer. User returns to their computer, powers it on and logs back intohttps://remote.company.com/rdweb to find that their email and word document that they were editing is still open and running just like they left it. This all makes if the users are using VDIs. However, we are only using remote apps at the moment. Assuming that my interpretation of HA is correct, how does it benefit users who are using remote apps?
- Can I deliver both VDI and remote apps to my users with my existing hardware?
- Does load balancing work right "out the box" or is there some configuration required?
- When configuring high availability mode, I am prompted for the DB connection string, database store, and DNS round robin name. This of course is asking for all of the SQL input information necessary to enable HA. I have installed and configured SQL Express 2012 SP1 with the native client beautifully onto the broker server. The broker server’s computer account is part of a security group that has dbcreator rights to SQL. Port 1433 is open on the local firewall and it has been configured in SQL configuration manager IP settings. But for some oddball reason, I get “could not create the database” (Event ID: 32776) when attempting to proceed with the HA configuration. Could this be because:
- The SQL instance is on the connection broker itself?
- My farm needs to have more than one broker server?
Any personal experience and words of wisdom would be greatly appreciated, thank you for your time!