**** BEGIN LOGGING AT Sun Jun 05 02:59:58 2016 Jun 05 17:50:08 getting a lot of these... Jun 05 17:50:12 ERROR: update-rc.d-0.7-r5 do_package: Function failed: package_name_hook Jun 05 17:50:25 Exception: AttributeError: 'OrderedDict' object has no attribute 'iteritems' Jun 05 17:50:42 * paulg assumes it is more python3 fallout. Jun 05 18:12:57 "As you are in python3 , use dict.items() instead of dict.iteritems(). iteritems() was removed in python3, so you can't use this method anymore." Jun 05 18:13:00 whee. Jun 05 18:28:11 package_name_hook (autogenerated) has a line like this: Jun 05 18:28:16 for (rprov_pkg, rprov) in bb.utils.explode_dep_versions2(d.getVar('RPROVIDES_' + pkg, True) or "").iteritems(): Jun 05 18:28:44 ...now to find the source of that... Jun 05 18:36:24 * paulg pushes everything off a cliff before investigating further. Jun 05 19:12:26 found it ; the iteritems was lurking in a class hiding in meta-overc Jun 06 01:26:14 I have several questions, First, I want to play with a poky rootfs over nfs, should I be using tmp/work//core-image-base/1.0-r0/rootfs? Jun 06 01:27:11 but I think I have a permission problem because in that location everything is still my user, so I get "ping: permission denied (are you root?)" Jun 06 01:32:06 Another question along the same lines, if I want to compile a project that isn't setup for cross-compiling (openswan in this case), can I use qemu for the dev system? because I don't think anything is persisted when I do "runqemu qemuarm" Jun 06 01:45:43 pthomas: no, you should add tar.bz2 to IMAGE_FSTYPES and extract the generated tarball as root. the bits in tmp/work are artifacts used by the build process only Jun 06 01:45:58 pthomas: and no, qemu is not an appropriate method for fixing it Jun 06 01:48:54 can I set IMAGE_FSTYPES from local.conf? Jun 06 02:08:22 thanks the IMAGE_FSTYPES is what I needed **** ENDING LOGGING AT Mon Jun 06 02:59:58 2016