> In looking at the code, I don't understand why adding sys/types.h at that > point would correct the problem. The only reason for a header file to be > included at that point is if we want HAVE_header_name defined, and no Bacula > code explicitly checks on HAVE_SYS_TYPES_H
The problem is that autoconf tests instead do check it. So, on one hand a test fails because something is defined in sys/types.h (which is not included in the test because HAVE_SYS_TYPES_H is not defined); on the other hand Bacula unconditionally includes sys/types.h and we crash because we're getting two sets of definitions (the one from sys/types.h and the one supplied by config.h -- which assumes no sys/types.h is present). Andrea ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Bacula-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/bacula-devel
