Re: Hedwig List, Apache Compile (was Re: Response to Mr. Steve Colie)

[Rick Forrester <rickf glacier jpl nasa gov>]

mhw wittsend com said:
> 	I tried to subscribe when I discovered that the apache source rpm 
> from 6.0 won't even compile on 6.0.  Yes I applied all the RedHat 
> patches and even figured out the screwy configure command "./
> configure --with-layout=RedHat --enable-module=all 
> --enable-shared=max" to get it to match the installation.  (It would 
> be nice if convoluted configure commands are required for a package 
> that those commands would be documented somewhere).

> 	First it gives me compile errors like this:

> mod_auth_dbm.c:77: ndbm.h: No such file or directory make[4]: *** 
> [mod_auth_dbm.so] Error 1 make[3]: *** [all] Error 1 make[2]: *** 
> [subdirs] Error 1 make[2]: Leaving directory `/usr/src/apache_1.3.6/
> src' make[1]: *** [build-std] Error 2 make[1]: Leaving directory `/
> usr/src/apache_1.3.6' make: *** [build] Error 2

> 	Then (after working my way around that one) I can't get half of the 
> modules to load because of an undefined symbol "stat".

> 	I'm begining to think that there have been too many changes made in 
> the RedHat way of tinkering with locations.  Now the source packages 
> are not even consistant with the distributed binaries.  That means, I 
> go back to 5.2 or find another distribution.  This may be acceptable 
> for people who only want to ever install from binary rpms and never 
> recompile something for themselves (I really don't want the DSOs in 
> Apache and I've got to compile some other stuff in) but it is not 
> acceptable to people like me who do have to add things back in on 
> top. 

Mike, I'll start a test run on the 6.0 source rpms this evening when I
get home.  I was one of the beta testers, and as one test, precisely 
because of this sort of thing, I ran a build test on the source RPM's
which were in the 5.9.7 beta.  The beta & QA people at Red Hat are
concerned about this sort of thing; they burned a special set of cdroms,
source and binary, just so I could do this easily.  There are always
changes in the move to the final release, though, and a problem probably
crept in at that point.

I'll let you know what happens.

best
   rickf
-- 
Rick Forrister                 <Richard Forrister jpl nasa gov>

"To get something done a committee should consist of no more than
 three people, two of whom are absent."  Robert Copeland