RPC over HTTP
Rickuk99
We have a legacy domain and exchange who however have
requested rpc over HTTP access for outlook.
The current set up, 3 Exchange 2003 servers all running SP2, 1 front end
server & 2 back end server. They have 2 domain controllers running as GC's,
not the exchnage servers, one running windows 2003 the other 2000.
Using the testexchangeconnectivity web site tool we pass all tests up to
NSPI 6004 and cant find a reason for this.
Testing NSPI Interface on Exchange Mailbox Server
An error occurred while testing the NSPI Interface.
Test Steps
Attempting to ping RPC Endpoint 6004 (NSPI Proxy Interface) on server
serverfqdn
Failed to ping Endpoint
Tell me more about this issue and how to resolve it
Additional Details
RPC_S_SERVER_UNAVAILABLE error (0x6ba) was thrown by the RPC Runtime
From the front end server I can telnet fine to port 6001, 6002 and 6004 on
the back end servers. The GC server has the registry settings for 6004
under the "NSPI interface protocol sequences" key and the front server also
has the rpc settings for the back end servers in the registry
HKEY_LOCAL_MACHINE\Software\Microsoft\Rpc\RpcProxy , one question here, does
the front end server need to know about the GC or do the back end server
proxy through to the GC? I have also confirgured the exchnage servers to
only use the 2003 GC instead of autodect GC servers.
Rich Matheisen [MVP]
wrote:
The E2K3 information store was a true end-point. It didn't proxy. The
E2K7 IS (port 6004) is a proxy.
>I have also confirgured the exchnage servers to
>only use the 2003 GC instead of autodect GC servers.
You can use the rpcping tools to break this down a bit by using it to
connect directly to the mailbox from a workstation. If that works, run
it from the FE server. If that works, run it from a workstation
through the FE server. At least you'll know which part breaks without
having to test the entire chain of machines.
---
Rich Matheisen
MCSE+I, Exchange MVP