svn commit: samba r8493 - branches/SAMBA_3_0/source/lib trunk/source/lib

[Gerald (Jerry) Carter]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Andrew Bartlett wrote:
> On Fri, 2005-07-15 at 13:24 -0500, Gerald (Jerry) Carter wrote:
>>-----BEGIN PGP SIGNED MESSAGE-----
>>Hash: SHA1
>>
>>>According to MSKB 141714 the %LOGONSERVER% variable should be
>>>substituted like our %L-variable.
>>> 		case 'L' : 
>>>-			if (local_machine_name && *local_machine_name)
>>>+			if (local_machine_name && *local_machine_name) {
>>>+				if (IS_DC)
>>>+					string_sub(p, "\%LOGONSERVER\%", local_machine_name, l);
>>> 				string_sub(p,"%L", local_machine_name,l); 
>>>-			else {
>>>+			} else {
>>Guenther,
>>
>>For some reason this makes me feel icky.  I know why you are doing it
>>but I'd rather see a generic means of support long name variables
>>using a consistent syntax.  Care to discuss this some more?
> 
> This seems odd to me too, but for a different reason.  I always assumed
> that this was evaluated on the client side, as a windows environment
> variable.
> 
> Can someone show this being set on SAMR, but being different in a
> netlogon reply, or some other way to show that the server munges it?

The problem is that we started expanding variables from the ldap passdb
backend and the %L in LOGONGSERVER was getting expanded to our %L.
Before we never touched it and so the client got %LOGONSERVER%.




jerry
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFC2EKQIR7qMdg1EfYRAqnAAJ4tO9A3DXoWQan0+rAz6aODUeKKsgCgjEca
k5wCMDFmCHVtrh0BOz1z22c=
=V31i
-----END PGP SIGNATURE-----