Description of problem:
sendmail-8.13.6-1.src.rpm does not compile on Red Hat 9, Fedora Core 1 nor Red
Hat Enterprise Linux 3
sendmail.spec has been modified following:
%define with_sasl1 yes
%define with_sasl2 no
%define enable_pie no
%define old_setup yes
Above changes enables building with sendmail-8.13.3-1.src.rpm
Build stops with following error message:
Making in /usr/src/redhat/BUILD/sendmail-8.13.6/obj.Linux.2.4.22-
cc `-O2 -g -march=i386 -mcpu=i686' -I. -I../../sendmail -I../../include -
I/usr/include/db4 -I/usr/kerberos/include -Wall -DXDEBUG=0 -DTCPWRAPPERS -
DNETINET6 -DHES_GETMAILHOST -DUSE_VENDOR_CF_PATH=1 -
D_FFR_WORKAROUND_BROKEN_NAMESERVERS -D_FFR_SMTP_SSL -DSASL=1 -
D_FFR_UNSAFE_SASL -DSM_CONF_LDAP_MEMFREE=1 -DNOT_SENDMAIL -
Dsm_snprintf=snprintf -D_FFR_MILTER_ROOT_UNSAFE -D_REENTRANT -DXP_MT -c -o
/bin/sh: -c: line 1: unexpected EOF while looking for matching ``'
/bin/sh: -c: line 2: syntax error: unexpected end of file
make: *** [main.o] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.93020 (%build)
Version-Release number of selected component (if applicable):
8.13.5 and forward doesn't work
Reproduced in Red Hat 9, Fedora Core 1 and Red Hat Enterprise Linux 3.
Steps to Reproduce:
1. download and install src.rpm
2. modify sendmail.spec
3. rpmbuild -ba sendmail.spec
Resolution found. Change sendmail.spec confOPTIMIZE value back to pre 8.13.3-
Resolution verified with Fedora Core 1 and Red Hat Enterprise Linux 3.
Change of confOPTIMIZE value does not affect Fedora Core 4 ability to build
sendmail.spec changelog has following entry:
* Tue Mar 1 2005 Thomas Woerner <firstname.lastname@example.org> 8.13.3-1.1
- fixed gcc4 build: use double quotes for confOPTIMIZE to avoid m4 confusion
That fix apparently broke the srpm for systems using gcc3.
Red Hat Linux 9 and Fedora Core < 4 are deprecated since longer time now and
shouldn't be used anywhere except with Fedora Legacy. What you're trying to
do is heavily unsupported. And Red Hat Enterprise Linux was never and will
never be a playground system for such obvious backports. Therefore suggesting
to close with WONTFIX/DEFERRED.