![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
I'm not a kerberos or a GSSAPI guru, simply one who has to clean up various bits of the puzzle that has been placed in front of me regarding the use of GSSAPI on CIFS and other Samba-implemented protocols. In particular, I'm concerned to try and get out of the GSSAPI game, and would some day love to put Samba back outside the 'implements some variant of GSSAPI' box. Currently, Samba implements a very shoddy GSSAPI wrapping, as well as SPNEGO, partly because it requires access to the raw Kerberos session key for use particularly in the CIFS protocol. CIFS uses the Kerberos session key for encrypting specific data portions on DCE/RPC named pipes, as well as to key the SMB signing system. My question is this (because I can't make heads or tails of the draft, sorry): Is the proposed PRF compatible with microsoft's existing use in this area, or will Samba forever-more be making calls to krb5_auth_con_getremotesubkey(context, auth_context, &skey) and krb5_auth_con_getlocalsubkey(context, auth_context, &skey)? Thanks, Andrew Bartlett -- Andrew Bartlett http://samba.org/~abartlet/ Authentication Developer, Samba Team http://samba.org Student Network Administrator, Hawker College http://hawkerc.net
Attachment:
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Kitten mailing list Kitten at lists.ietf.org https://www1.ietf.org/mailman/listinfo/kitten