[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: keytab - MIT Keytab Binary File Format Encoder / Decoder

On Wed, 03 May 2006 18:53:12 +0200
Love Hörnquist Åstrand <lha@kth.se> wrote:

> Michael B Allen <mba2000@ioplex.com> writes:
> > 2) What codeset are strings? Are they UTF-8 or locale dependant?
> "kerberos codeset", today, basicly ascii.

Good. Than it won't hurt to call it UTF-8 :->

> > 3) Are my assumptions about num_components correct: 1 for no service,
> >    2 with service?
> See code in Heimdal, depends on version. Hint, there is no such thing as
> "name" and "service" in kerberos5, its just a list of
> components. foo/bar/baz/kaka/cookie@EXAMPLE.ORG is a valid name.

Ahh, I see. Is the realm always present? So it's:

  struct keytab_entry {
      int32_t size;
      uint16_t num_components;
      counted_octet_string realm;
      counted_octet_string components[num_components];
      uint32_t name_type;
      uint32_t timestamp;
      uint8_t vno8;
      keyblock key;
      uint32_t vno;

  counted_octet_string {
      uint16_t length; 
      uint8_t data[length];

> The "vno32" as you name it, is a heimdal extention, and now adays java

Interesting. MS ktpass.exe enmits the 32 bit kvno.