Hello
Can someone help me with the following question please,
If I am a client computer, I query my configured DNS server looking for an LDAP server, once I have the IP address of the LDAP server I can connect over port 389
Now then LDAP v3 supports simple authentication (username/password) or SASL (simple authentication and security layer) which I believe kicks off Kerberos negotiation (possibly via SPNEGO)
The reason I ask is I want to put LDAP queries via a load balancer, so if a client want to query LDAP they point to a load balanced VIP.
However Kerberos (port 88) does not play nicely or at least easily with load balancers (due to the SPNs)
So I want to understand when does a client go from its initial LDAP connection to a DC (port 389) to negotiating and using Kerberos (port 88) ?
I am assuming LDAP is used to querying AD and updating AD (adding a user to a group) or are updated to AD done over RPC by the client?
I am a bit confused about all this, I understand Kerberos and TGT and Service tickets, the PAC in the Kerberos ticket is turned into a Windows token, which is then used to check for authorisation. In which case logically I will need a Kerberos service ticket
for the LDAP service on the DC so it can be turned this Kerberos ticket into a Windows token to check the ACEs (in the ACL)to see if I have rights to add a user to a group for example.
However I understand LDAP on its own supports CRUD operations (create, read, update, delete), so logically I could to without Kerberos or NTLM if I can do a simple bind to the DC an use pure LDAP to make the updates?
As you can see I am a bit confused on how this all fits together please help! :)
CXMelga