I have a public folder db on my CCR. Every time I turn off the active servers the CCR fails over but the public folder fails to start. Is there a way to resolve this issue. Would it be better to house my public folders on a server that is not clustered? My public folder is not that big. The main use is for calendars and a couple of folders that recieve group mail.

Its determined on the high availability type of the public folder you want. In the CCR environment, public folder will use continuous replication to accomplish the high availability; and in the non-cluster environment, public folder can also obtain the high availability by using the public folder replication among multiple public folder servers Check info: 1. Is there any error event in the application on the nodes when the public folder database failed to start? 2. Could you manually start the public folder database? If cant, is there any error info occur? Is there any relevant event in the application log? 3. Are the resources online in the cluster admin console? 4. Does the replication work for public folder database? Please compare the number and size of database and log files between nodes 5. What shows in the Get-StorageGroupCopyStatus? Can you run it and check output by using the listed conditions in this article? 6. In order to verify the current functionality of CCR, please run the cmdlets below: Get-ClusteredMailboxServerStatus Identity CMSName | Fl 7. Please run ExBPA against the CMS for health check

In addition to what James said, can I ask whether you have any other additional public folder databases outside of the one housed on the CCR environment? Also, when you say you "turn off" the active server, do you mean a clean shutdown or are you simulating a failure of the active node? If so, the outcome is to be expected since a public folder database will fail to mount in lossy failover scenarios where other public folder databases exist within the organisation.Neil Hobson, Exchange MVP

What I did was shut down the active server to simulate a failover. Everything came back up fine except the second storage group (public folders). I could not get it to start. What I had to do was to move the services to the other server and everything came back up. I only have one public folder database. I am running a 2003 environment now and am trying to get things moved into 2007.

So, theres no error event in the application log? All the resources are online? The output of the cmdlets is fine? No error info in the ExBPA as well? You said that you cant get the public folder database started, is there any error info or error event? Quote: What I had to do was to move the services to the other server and everything came back up What service? Could you describe more details? Quote: I am running a 2003 environment now and am trying to get things moved into 2007 So, its mixed environment? Could you describe the current exchange topology?

I think I figured it out. I am migrating to 2007 from 2003 so I have them both running. I read that using CCR public folders will not fail over if there are more then one public database. I have two. The on in 2007 and the 2003 database. I think once I decommision the 2003 environment I will not see this problem. What is happening now is that when I turn off the active node all exchange services start except public folders. Once the server comes back on line I can then start public folders.

Well, public folder replication will break in the cluster, if thats what you encountered. Please post at here if theres any further update