[917] in Kerberos_V5_Development
Re: dbm.h vs
daemon@ATHENA.MIT.EDU (Mark W. Eichin)
Fri Jul 21 00:05:40 1995
Date: Fri, 21 Jul 95 00:04:48 -0400
From: "Mark W. Eichin" <eichin@cygnus.com>
To: krbdev@MIT.EDU
In-Reply-To: <9507202357.AA18904@perdiem.cygnus.com> (eichin@cygnus.com)
On a related note, the berkeley "db_dbm" interfaces that kerberos uses
(as defined in src/util/berk_db/hash/krb5_ndbm.c) depend on having DBM
and datum defined (presumably from a system dbm.h or ndbm.h.) I don't
think we want to be checking for ODBM if we're using the new
interfaces, especially as it proliferates the number of untested
combinations.
I'd even argue that the combinations should be consolidated (have
*one* with-dbm flag, not seperate ones for aname and rcache and kdb)
but that's mostly due to the frustration I've been having getting
*either* option to work with linux :-)
