Hide Forgot
Description of problem: postgresql-odbc64 fails the rebuild test on ppc and ia64 happens on stablesystems - the brew build obviously passed ... Version-Release number of selected component (if applicable): postgresql-odbc64-09.00.0200-1.el5.src.rpm How reproducible: always Steps to Reproduce: 1. rpmbuild --rebuild /mnt/redhat/brewroot/packages/postgresql-odbc64/09.00.0200/1.el5/src/postgresql-odbc64-09.00.0200-1.el5.src.rpm Actual results: build fails Expected results: build succeeds Additional info: I believe the relevant log snippet is this: gcc -DHAVE_CONFIG_H -I. -I. -I. -I/usr/include -I/usr/include -Wall -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -c info.c -fPIC -DPIC -o .libs/info.o In file included from /usr/include/sql.h:19, from psqlodbc.h:62, from info.c:19: /usr/include/sqltypes.h:56:2: error: #error "Needs to know how big a long int is to continue!!!"
Since it builds fine in brew, I'm inclined to close this as NOTABUG. We can't promise that SRPMS will rebuild correctly in any random environment. In particular, I'm guessing that this has something to do with having devel RPMs for multiple architectures installed, something that isn't really feasible to deal with.
It builds again with newer unixODBC64-2.2.14-3.el5.ia64, tested on ia64 only. See bug #757126.
(In reply to comment #2) > It builds again with newer unixODBC64-2.2.14-3.el5.ia64, tested on ia64 only. > See bug #757126. I can confirm that ia64 build goes fine with that new version however, there are still doubts about ppc, and the testing machine is down (ticket already filed)
ok, ppc now builds fine too, closing