If you add a file to the project, please remember to add it to the appropriate Makefile.am as well. Failure to do so *will not show up* in our standard smoke/regression tests because those do "make install" but they will prevent RPMs (and probably equivalents on other distros/platforms) from building. To see why, let's review how building real packages works. * First there's autogen and configure, which operate in your original source directory turning .am and .in files into real platform-specific makefiles and scripts. * Then there's "make dist-gzip" and friends, which package up *only files specified in makefiles* into a tarball. * The rpmbuild "prep" stage unpacks this tarball and applies patches, typically in ~/rpmbuild/BUILD. * The rpmbuild "compile" stage does what you'd expect within BUILD. * The rpmbuild "install" stage does a "make install" in BUILD, which populates BUILDROOT. There's more, but those are the key parts. With that in mind, we can look at various options for adding new source files. * xxx_SOURCES go into the dist tarball and become available in BUILD for the compilation phase. xxx_HEADERS do likewise, and also get populated into BUILDROOT during the install phase. * noinst_HEADERS inhibits propagation into BUILDROOT (making headers act like sources), and is generally what you'd want for headers that are not specifically intended to be in -devel packages. * nodist_xxx_SOURCES can be used to specify files that are needed to build xxx, but will not be in the dist tarball. This is what you'd use for generated files. * Anything not specified in any of these ways never even makes it to the dist tarball. There are different rules for other things like Python scripts, but I don't remember those very well. Also, I might have gotten some of the above slightly wrong; corrections from people who know this stuff even better than I do (Kaleb?) are welcome. The most important point is this: "git add" is not enough! _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-devel