Sounds like a documentation bug... :(
When I can connect to sourceforge, I'll try and file a bug report for
this.. Feel free to beat me to it.
[BTW. I'd guess that you are not initializing data=NULL before making
this call, so if pam_get_item doesn't alter it, you'ld get random
behavior in the first incarnation of your code.]
Cheers
Andrew
Mike Glover wrote:
>
> In the process of coding pam_sql, I was trying to figure out
> how pam_get_item and pam_set_item work. From the docs (on kernel.org)
> it looks like pam_get_item should either return either a pointer
> to my data, or NULL if it couldn't be found. So I should be able
> to do something like this:
>
> retval = pam_get_item (pamh, KEY, &data);
> if ( ! data ) {
> ...
> }
>
> Alas, when I did this, I wasn't detecting when the data
> wasn't found, and my program segfaulted. I ended up doing
> this:
>
> retval = pam_get_item (pamh, KEY, &data);
> if ( ( retval != PAM_SUCCESS ) || (! data ) ) {
> ...
> }
>
> This *seems* to work, but the docs aren't very clear on this
> point. So am I doing it right?
>
> -mike
>
> --
>
> GnuPG key available at http://devel.duluoz.net/pubkey.asc
> Key ID = 1024D/9A256AE5 1999-11-13 Mike Glover <mpg4@duluoz.net>
> Key fingerprint = EF6E 8BCB 4810 E98C F0FD 4596 367A 32B7 9A25 6AE5
>
> _______________________________________________
>
> Pam-list@redhat.com
> https://listman.redhat.com/mailman/listinfo/pam-list
