I am in the process of installing my new 2010 system this week. My intent is to have a 2 member DAG running MBX roles and a HUB/CAS server in front of it that will also act as witness. I have all my hardware built and 2008R2 Enterpriise installed on them. Yesterday I extended the schema and installed the CAS/HUB roles to the appropriate server. I came in this morning and installed the MBX role on the first server that will be in the DAG. I then switched to the second server to be in the DAG and it was at a login screen. Basically, after I left yesterday, it rebooted itself and there is a POST 1719 Controller failure event prior to this powerup message in HP Int. Logs. I really don't want to stop the testing and pilot but if I freeze where I am, all is on hold until I can get HP to check it. One solution is to install new firmware and call them if the problem reoccurs. I lean to putting on the firmware and let it run for a while before installing exchange and putting it into a DAG. So, can begin testing and pilot of the exchange system and later add the other mbx and create a dag or am I stuck until I trust this server enough to but it in the DAG?

You can add a server to your DAG at anytime. That is part of the beauty of Exchange 2010, the support of incremental deployment. See: http://technet.microsoft.com/en-us/library/dd638137.aspx#IDTim Harrington | MVP: Exchange | MCITP: EMA 2007/2010, MCITP: Lync 2010, MCITP: Server 2008, MCTS: OCS | Blog: http://HowDoUC.blogspot.com | Twitter: @twharrington

The only thing I would add is that what you should do is configure a CAS array at the start. It is more difficult to switch to a CAS array after the server has gone live. Simon.Simon Butler, Exchange MVP Blog | Exchange Resources | In the UK? Hire Me.

Any update?Gulab | MCTS-MCITP Messaging: 2010 | MCTS-MCITP Messaging: 2007 | MCC 2011 | Skype: Gulab.Mallah

I was hoping it was something simple, but after trying a firmware update HP suggested the keybd and mouse stopped working. They are now going to have to send someone to replace the main system board all those items are mounted on. I will probably rebuild/reformat the raid layout just to be sure so I will definitely go to pilot without the DAG because the CAS/HUB and the other MBX server are built and ready to go. As to the CAS array, I was under the impression that could wait so we will have to go with 1 CAS/HUB and then in the next month or so we should be implementing at least one more as well as a load balancer. We are thinking F5. Thanks so much for your help. I guess it's just terrible luck to have a bad server right out of the box.

Salam, I really recommend that you implement the CAS Array (whether NLB or a load balancer) before moving any of the mailboxes otherwise you'll have to spend alot of time reconfiguring outlook clients. On the other hand, why do you need to rebuild your raid layout? If replacing the motherboard fixes the problem, then your hard drives are not the cause of the problem, and your server should up and running as is. I wish you the best of luck. Kindest regards.For every action, there is a reaction!

Having the CAS array in from day 1 will save you a lot of work. If you implement the CAS array later, then you have to touch every client. They don't start using it automatically. The profile either has to be recreated or repaired to use the CAS array host. Instead just create the CAS array host name, point it to your live CAS and then configure the array. If you then decide to implement a load balancer of any description later, you do not have to change anything other than the DNS entry. The clients are already using that host name. I put a CAS array host name in to all Exchange 2010 deployments I do, no matter what the size. Simon.Simon Butler, Exchange MVP Blog | Exchange Resources | In the UK? Hire Me.