[2020] in Kerberos_V5_Development

home help back first fref pref prev next nref lref last post

Re: Final issues before 1.0 code freeze

daemon@ATHENA.MIT.EDU (Marc Horowitz)
Mon Nov 25 14:42:33 1996

To: "Theodore Y. Ts'o" <tytso@MIT.EDU>
Cc: krbdev@MIT.EDU
From: Marc Horowitz <marc@cygnus.com>
Date: 25 Nov 1996 14:41:34 -0500
In-Reply-To: "Theodore Y. Ts'o"'s message of Mon, 25 Nov 1996 12:45:52 -0500

"Theodore Y. Ts'o" <tytso@MIT.EDU> writes:

>> 	* Build of kadmind (with static libraries only) fails on NetBSD
>> 		unless you specify your own resolver library.  
>> 
>> 		Given that it's only one OS, and we hvae a work-around,
>> 		and it doesn't matter if we're using shared libraries,
>> 		it seems like we shouldn't half to worry about it.

What is the source of this problem?  I don't remember seeing a message
about it, but I could have missed it.

>> 	* make install blows out for NetBSD
>> 
>> 		This is mentioned in the README file, with a pointer to
>> 		the build notes, but there doesn't seem to be anything
>> 		in build.texinfo, nor was a PR opened on this.  Que pasa?

I can file a PR, but Sam knows more than I do about the problem.
NetBSD make, for reasons which are lame, will use an "obj"
subdirectory if one is present, instead of using the current
directory.  This causes Bad Things to happen.  We can just decide to
state that bsd make is broken (which it is), and that bsd people
should use gmake (which works fine), or we can just remove the install
rule from the db in the krb5 tree, since it doesn't get used, anyway.

		Marc

home help back first fref pref prev next nref lref last post