Outlook connection status - mix of HTTPS and TCP/IP traffic

[Alok]

Hello,
SOmetimes i find in my outlook i get both HTTPS and TCP/IP in outlook
connection status.
Can somebody please help me understand under what circumstances do we see
such kind of traffic.
Our OWA/Backend mailbox servers are behind ISA servers.
Appreciate any help.
Regard
Alok

 

[neo [mvp outlook]]

Mixed connections can happen for a variety of reasons, but the easiest way I
have found to describe it is like this...

I show up in the office in the morning and logon to network/exchange. So at
this point I have a tradition TCP/IP connection. What I don't know is that
Exchange is located across the wide area network from my location. The
connection gets flakey for a moment and Outlook looses part of the
connection (could be to domain controller (gc) or exchange information
store). At this point Outlook tries to restablish the connection. If it
can't restablish the TCP/IP connection, it falls back to HTTPS. It then
tries that a few times and if it can't connect, it fails back to TCP/IP
after waiting a few minutes*.

/neo

* there is logic where this wait state between protocol fail back get wider
and wider apart.

 

[Alok]

Hello Neo,

I see something like
Type Con
Directory HTTPS
Mail TCP/IP
Public folders TCP/IP
Public Folders HTTPS

Is this normal ?How can i forcibly make all of them MAPI connection without
changing anything in RPC over https settings.
The objective is at office all Laptop users need to connect using MAPI and
at home connect using HTTPS but the currrent settings doesnot allow them.
Current setting is :
IN exchange proxy setting only slow network setting is ticked with basic
authentication and thats what we require.

Thanks Neo

 

[neo [mvp outlook]]

Again... yes it is normal if something happens to the connection between the
desktop and servers.


You can try disabling protocol fallback....

Registry location:
Software\Policies\Microsoft\Office\11.0\Outlook\RPC

ValueName:
DisableRpcTcpFallback

Type:
DWORD

Accepted Values:
1 = Disable fallback
0 = Allow fallback

 

[Alok]

Hello Neo,

Would you kindly explain me the below line which you had mentioned earlier.
* there is logic where this wait state between protocol fail back get
wider and wider apart.

Alok

 

[neo [mvp outlook]]

When Microsoft coded Outlook, they coded in some logic to cycle between the
two connection types. If both fail, Outlook takes a deep breath and then
tries the two protocols again. It is that deep breath in-between cycles
that gets longer before trying the connection again.

And before you ask... no, it isn't documented as far as I know.