On Wed, 12 Dec 2012, James Page wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA256 > > On 11/12/12 23:00, Gary Lowell wrote: > > On Dec 11, 2012, at 2:06 AM, James Page wrote: > > > >>> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 > >>> > >>> On 11/12/12 06:32, Gary Lowell wrote: > >>>>>>>>> > >>>>>>>>> I assume you are building with "dpkg-buildpackage" > >>>>>>>>> ? > >>>>>>>>> > >>>>>>>>> The manpage shows: > >>>>>>>>> > >>>>>>>>> "-B Specifies a binary-only build, limited to > >>>>>>>>> architecture dependent packages. Passed to > >>>>>>>>> dpkg-genchanges." > >>>>>>>>> > >>>>>>>>> "-A Specifies a binary-only build, limited to > >>>>>>>>> architecture independent packages. Passed to > >>>>>>>>> dpkg-genchanges." > >>>>>>>>> > >>>>>>>>> So on the i386 and amd64 machines you'd run with -B > >>>>>>>>> and sync them to ceph.com > >>>>>>>>> > >>>>>>>>> On one of the machines you'd also run with -A which > >>>>>>>>> should produce the architecture independent > >>>>>>>>> packages like libcephfs-java. > >>>>>>>>> > >>>>>>>>> That's the theory, I haven't tested it :) > >>>>>>>>> > >>>>>>>>> Wido > >>>>> Thanks Wido. We're using pbuilder, but it looks like it > >>>>> has similar options, or can pass an option string to > >>>>> dpkg_buildpackage. I'll do some testing. > >>> > >>> "--binary-arch" will limit a pbuilder build to the target > >>> binary architecture only; I would recommend you use this with > >>> the amd64 build and build the arch: all packages out of the > >>> i386 build; this is what happens in the official Ubuntu > >>> builders. > >>> > > Hi James, > > > > I thought this was going to be the easy solution, but on running a > > quick test, we are already calling pbuilder with the --binary-arch > > option and it its building the java package anyway. It looks > > like there is a deeper issue in that we building the java package > > in the default target. It looks like for this to work, we need to > > move the java library build to it's own target in the Makefile, and > > build that target from the binary-indep target in the debian rules > > file. Does this sound like I'm on the right track ? > > Gah - this will bite when I do the next upload to Ubuntu as well then; > Can I suggest that we rework debian/rules for debhelper >= 7 and use > overrides rather than the current 'old style' rules which define all > tasks? I was toying with doing this anyway (and have it working > locally) - it does cut out some of the content from d/rules and makes > it a bit more *magic* > > Thoughts? I'm happy to raise a pull request for this. Yes, please! In the meantime, does build-indep sound like a valid workaround? (This hiccup has been delaying some simple fixes in 0.55.1, but unless there is a quick fix I think we'll just release it without the java bindings for now to get those fixes out to people.) Thanks- sage -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
