I have a problem that looks to be slp related but i'm stumped as to how to fix it.

We have a mixed MS/NW environment with an ISA server for (amongst other things) VPN access and NW servers for file/print. The clients are all WinXP SP2 with client 4.9.1sp3. We use the MSGina as the primary and the NWGina sits behind it to allow passthrough authentication to the NW side of things. This works fine on the LAN (logging on from any one of the 3 subnets) and the netware client authenticates silently without you knowing it's there.

The problem comes with logging on over the VPN, the connection establishes fine but you then get "tree or server cannot be found" and the netware logon box appears, no matter what combination of tree/server/context/ip address i enter in here (all the usual suspects) it will see nothing.

We have a scoped SLP setup with 2 DA's all configured as per TID10059981 and the clients are set to receive their SLP options (78 + 79) via this. The ISA (2004) server is configured to allocate IP's for the clients (on a different subnet to the lan) but a limitation is that it cannot distribute dhcp options so i've configured the scope and da's statically on the VPN machines. When you run an SLPINFO the static entries are shown as NORSP.

On occasions if you wait 5-10 minutes at the novell login box and try again then you can get in (we're talking once or twice) and i've seen some posts and tid's referring to errors in SRVLOC over vpn connections causing problems with this because of timeouts.

Any help on this would be greatly appreciated before i lose the plot completely!


Marcus Kneale
eServices Server administrator
Carrick District Council