SYSKEY, TNG freeze, 2.0.x->TNG merge and other thoughts

[Luke Kenneth Casson Leighton]

> Actually luke, don't mix up the RPC server/client stub code
> with the RPC *implementation* code in 2.0.x. The implementation
> code in 2.0.x is much more reliable than the TNG code (not
> the actual server/client functions though).

cvs -t co -r SAMBA_2_0 samba
cvs -t diff -u -r "1.8" source/rpc_server > foo

i'm just reviewing srv_lsa.c diffs.  it shows, for example that the
implementation of lsa_open_policy() is this:

for (i = 4; i < 20; i++)
{
	r_u.policy_hnd[i] = i;
}

you are not telling me that this is more reliable than the tng
implementation, i take it?

what i _do_ like about 2_0 is this:

api_reply_lsa_open_policy()
{
	if (!lsa_io_q_open_policy(...)
	{
		return False;
	}
	...