Thx a lot!
On Fri, Mar 19, 2010 at 8:14 PM, <[email protected]> wrote: >> On Fri, Mar 19, 2010 at 1:27 PM, <[email protected]> wrote: >>>> Hi, >>>> >>>> I am still trying to make an sasl bind. Now I use ldap_sasl_bind_s >>>> with 'CRAM-MD5'. The server sends a 'LDAP_SASL_BIND_IN_PROGRESS' back, >>>> but as far as I can see the server cred are empty. How do I make the >>>> next call with ldap_sasl_bind_s? >>>> >>>> In the documentation I have seen that the use of ldap_sasl_bind is not >>>> recommended as client use, instead the ldap_sasl_interactive_bind_s >>>> should be preferred. Has anyone a working example of how to use this >>>> method, or some api doc? I have no clue how to set 'flags' the >>>> LDAP_SASL_INTERACT_PROC, an the 'defaults' params.. >>> >>> You first say that you're using ldap_sasl_bind_s(), and then you note >>> that >>> ldap_sasl_bind() is not recommended. Do you realize that despite some >>> similarities in the name, the two functions are profoundly different? >>> By >>> no means ldap_sasl_bind_s() can return LDAP_SASL_BIND_IN_PROGRESS. >> >> Yes, I know that those methods behave differently - as I have read in >> the docs, I just search for the simplest alternative as possible for >> ldap_simple_bind, since I have seen that the ldap_simple_bind methods >> are deprecated. >> >> Nevertheless ldap_sasl_bind_s returns LDAP_SASL_BIND_IN_PROGRESS which >> I understand as a challenge which should be returned back to the >> server. This one seems to had a similiar issue: http://bit.ly/awT4D4 >> >> But I think I have to look at the examples for better understanding - >> thx for the tip! > > ldap_sasl_bind_s() can be used, passing LDAP_SASL_SIMPLE, in lieu of > ldap_simple_bind_s(). ldap_sasl_interactive_bind_s(), only need to be > used for those methods that require multiple steps; > LDAP_SASL_BIND_IN_PROGRESS indicates that a further step is expected, > which never happens when performing a simple bind. See the code snippet > in slap_client_connect(), in servers/slapd/config.c for a complete example > of how both functions can be used in the most complete form. > > p. > >
