I just setup a nice new nework with OES for 25 users. All running well except for the login lag.
With tree name specified we see exactly 45 seconds of delay before the login script processing starts.
But with an IP address in place of the tree name, the login is almost instant so this must be an SLP issue right?

Server: OES2 SP3 SLES 10 SP4
Client Win XP with client 4.91 SP5 (IR1)

I've spent hours on this, so I hope hope someone can give me a pointer or two.
Here are some of my checks and configs.

Windows PC:
On the client side everything is set to static
C:\>slpinfo /d

************************************************** ***
*** Novell Client for Windows NT ***
*** Service Location Diagnostics ***
************************************************** ***

SLP Version:
SLP Start Time: 9:36:31pm 3/3/2012
Last I/O: 8:08:09pm 3/4/2012
Total Packets: Out: 161 In: 82
Total Bytes: Out: 12545 In: 2782

DA IP Address Source(s) State Version Local Interface Scope(s)
--------------- --------- ----- ------- --------------- --------------- CNFG UP SLPV2 DEFAULT CNFG UP SLPV2 DEFAULT

On the server:
slptool findsrvs service:

server1:~ # slptool findsrvs service:ndap.novell

# grep -v "#" /etc/slp.conf | grep -v ";" | more

net.slp.DAAddresses =
net.slp.interfaces = # this is the LAN subnet
net.slp.traceDrop = true
net.slp.useScopes = DEFAULT
net.slp.isDA = true
net.slp.DASyncReg = false
net.slp.isDABackup = true
net.slp.DABackupLocalReg = true