Hi,
I have a 3 node Exchange 2013 deployment (CAS and two node mailbox DAG). Having configured my Outlook Anywhere settings some 9 months (using split DNS to allow my OWA SSL cert to work both externally & internally, despite having a .local internal domain), everything was great until I started implementing some office 365 preparation ie Added External domain to Office 365 Portal and added external DNS records (autodiscover.outlook.com& msoid) set up ADFS server and UPN suffix for external domain) and after some time, my Outlook clients could no longer connect (kept prompting for credentials that were not accepted).
OWA, Activesync still work (internally and externally) but Outlook Anywhere only works externally now. I have played around with the Authentication settings (tried NTLM) on both Internal & External URL settings using powershell but no change.
Set-OutlookAnywhere -Identity “CAS name\RPC (Default Web Site)” -ExternalHostname “OWA.domain.com” -InternalClientAuthenticationMethod ntlm -InternallClientsRequireSsl $true
Set-OutlookAnywhere -Identity “CAS name\RPC (Default Web Site)” -ExternalHostname “OWA.domain.com” -ExternalClientAuthenticationMethod ntlm -ExternalClientsRequireSsl $true
When I try and create a new profile, Autodisovery finds the correct mailbox GUID and adds it in but then says the Exchange server is unavailable. Interestingly when I try and manually add the settings using the CAS server name, it resolves the server name to the mailbox_server_node_b (which is the location of the passive DAG copy) and then cannot connect with the resulting profile.
I cannot see how the Office365 Prep is related but equally cannot understand what can explain the issue