[5166] in Kerberos
Re: Porting KRB5 to Linux?
daemon@ATHENA.MIT.EDU (Russ Taylor)
Fri May 12 21:46:44 1995
To: kerberos@MIT.EDU
Date: 13 May 1995 00:58:49 GMT
From: rtaylor@cie-2.uoregon.edu (Russ Taylor)
On 12 May 1995 17:38:05 -0400, eichin@MIT.EDU put forth into the Codex of Shub-Internet:
> Just to note -- I've been discussing things with Mr. Bass offline, and
> finally discovered something. The only problem he was having dealt
> with the -ll -lfl lex/flex confusion -- which the config scripts could
> deal with more gracefully, and I'll look into fixing that.
> What I at least had been trying to get out of him was more information
> on the "struct problem in std_rqs.c" which now that I check the
> archive, wasn't his problem at all, but was from
> rtaylor@cheetah.pacinfo.com (Russ Taylor) from whom we've heard
> nothing further. *That* problem is the one that is likely due to
> broken tools, which is a point I made earlier but was perhaps missed
> due to the smoke :-)
I quoted the file and line:
std_rqs, line 1: parse error before "ssu000001"
Then various execess elements errors.
Looked to be from a missing header file, btw
std_rqs.c is in utils/ss
Fortunately, once I downloaded V5R5 (came out the week after I started working
on this), the problem went away.
There were also a few lib errors (libkrb.a,libdos.a,libkdb.a)
I'd assume due to libss.a never finishing (didn't look into very hardbut
those were fixed by deleting the second definition, of course -- they also
didn't occur in V5R5.
Finally, ss/ss.h was missing, though I don't recall in which file
(the make.out file is on another machine, and I don't feel like looking it
up right now).
Since I've not had any problems compiling other software, it seems the tools
are fine and the source is to blame, unless V5R5 also changed it's sed/awk
calls.
--
"Why do little blue midgets keep beating me with fish?" -- the Tick
