Password sync plugin, and questions about plugin criticality

[Nicolas Williams]

On Mon, Jun 26, 2006 at 09:15:08AM -0400, Ken Hornstein wrote:
> >On Mon, Jun 26, 2006 at 01:33:38AM -0400, Ken Hornstein wrote:
> >> void pwupdate_init(void **context, krb5_context);

Also, void * sucks.

Use pointers to incomplete structs instead, as Will suggests.

> >No krb5_context should be needed.  One argument should suffice.
> >
> >Presumably it's up to the plug-in to be thread-safe.
> 
> But if I want to call krb5_* functions (which is supposed to work in
> the MIT plug-in architecture), how would I do that without a krb5_context?

Call krb5_init_context() :)  (you can stash that context in the
plug-in's handle...)

> >> int pwupdate_precommit_password(void *context, char *password, int pwlen,
> >> 				char *error_msg, int error_msg_len);
> >> 
> >> 	Update the password in the external store; called before password
> >> 	is stored locally.  In case of failure, return non-zero code and
> >> 	set error_msg to an error string.
> >
> >This mixes password quality checking and password updates.
> 
> I don't see how.  (One thing you may have missed; I had already worked
> out a password quality checking plugin seperately).

Oh, I misinterpreted the purpuse of the error_msg_* parameters.

> >Also, if you separate password quality checking from password
> >synchronization you can collapse these two functions into one.
> 
> I don't think I can; some people absolutely want external password sync
> to happen before the password gets written locally; other people
> want the exact opposite.

Why?  Do they expect atomicity?