[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Seg Fault during make tests on AIX (ITS#1134) (fwd)

I did the two things Bryan suggested: 1) ./configure --enable-shared=no 
				      2) vi include/portable.h replacing
				         the HAVE_GETADDRINFO line before
					running make depend and make.

I am able to run all the tests successfully now!

   Ric Miller           (ric@colostate.edu)
   Academic Computing & Networking Services

---------- Forwarded message ----------
Date: Thu, 21 Jun 2001 14:12:40 -0400 (EDT)
From: Bryan Zeli Kartzman <bryan@ymail.yu.edu>
To: Ric Miller ACNS <ric@yuma.acns.ColoState.EDU>
Subject: Re: Seg Fault during make tests on AIX  (ITS#1134)  (fwd)

should have included you in this mailing as well...


---------- Forwarded message ----------
Date: Wed, 20 Jun 2001 20:10:24 -0400 (EDT)
From: Bryan Zeli Kartzman <bryan@ymail.yu.edu>
To: JeromePerreard@lyon.msi-sa.fr
Cc: kurt@openldap.org
Subject: Re: Seg Fault during make tests on AIX  (ITS#1134) 

I saw your post about AIX 4.3.3 generating seg. faults at the "make test"
step and was having the same problem. I did the following and was able to
get it to pass the tests: I configured with the --enable-shared=no (no
shared libraries). Still I was having a problem with getaddrinfo so I went
into the file include/portable.h and took out the definition of
and replaced it with 

I then did the make depend,make, and make test and it passed all the
tests. ( I then went into the configure script and hardwired getaddrinfo
to no, reran the configure --enable-shared=no,make depend,make, and make
test. If you want instructions for that ask and I'll send them). I haven't
yet actually set up my slapd.conf file yet or created my database, but
passing those tests is a good sign that it will work. I'm using gcc with a
gdbm backend. I had installed ssl and sasl support but for now I've
removed those libraries until I see that what I've done so far actually

Why don't you try the same thing on your box and if you are also
successful we can report back to openldap that we have a workaround for
the problem.