**** BEGIN LOGGING AT Mon Apr 13 03:00:00 2015 Apr 13 05:50:03 <_2_Bird> Well Apr 13 07:51:19 koen, I was wondering. We have a new device and we are using angstrom/yocto1.7. systemd claims it is managing /etc/resolve.conf.... but that file should contain only one entry, namely 127.0.0.1 to play along with connman's way of dishing up a nameserver though dhcp... Apr 13 07:55:51 pompomJuice: yes? Apr 13 07:58:34 Any ideas how to switch it back like it was? Apr 13 07:58:45 ad am I correct? Apr 13 07:59:19 The best way I can think of to fix this is to include my own file in base-files... would this be the correct way to do it? Apr 13 07:59:57 Or, how can I disable this systemd managed resolve.conf Apr 13 08:00:10 And I wanted to mention it since it is a distro issue right? Apr 13 08:01:17 I don't get the problem Apr 13 08:02:01 As far as I understand... connman is supposed to be a one stop shop for all our network connectivity management Apr 13 08:02:20 systemd is not supposed to be managing anything Apr 13 08:02:41 if you're using connman, yes Apr 13 08:02:51 I thought connman is an angstrom distro default Apr 13 08:03:02 gm bluelightning Apr 13 08:03:04 no, it's an image default Apr 13 08:03:16 e.g. 'angstrom-image' has no connman, systemd does everything Apr 13 08:03:18 morning woglinde, koen, all Apr 13 08:03:21 hey bluelightning Apr 13 08:03:23 oh Apr 13 08:04:15 my mistake then Apr 13 08:04:15 hey, we have the old crowd online, gm :) Apr 13 08:07:21 good morning all Apr 13 08:09:43 ant_work except us Apr 13 08:10:23 hi mckoan Apr 13 08:35:24 bluelightning: I'm sorry I'm plaguing you whith extra-issues Apr 13 08:35:39 but what did happen to the license code? Apr 13 08:35:40 http://pastebin.com/Yy6BbmTH Apr 13 08:35:56 sigh Apr 13 08:36:36 I have to leave in 1 minute, can you please file a bug? Apr 13 08:37:01 bbl Apr 13 08:37:05 sure, bye Apr 13 08:37:16 thanks Apr 13 08:37:31 * ant_work before will rebuild from scratch just in case... Apr 13 11:15:25 kergoth: you ever released that bitbake env tool? or is it just still a gist proof-of-concept Apr 13 12:00:39 bitbake 1.24 wont pickup connman-tests, but bitbake 1.22 does. connman-tests is provided like so: PACKAGES =+ "${PN}-tools ${PN}-tests ${PN}-client" Apr 13 12:01:36 Anyone else seen this with this PACKAGES += "foo" method in 1.7? Apr 13 12:19:46 does anyone know what layers/meta-fsl-arm/classes/fsl-kernel-localversion.bbclass is good for, I thought there is already a KERNEL_LOCALVERSION variable in the kernel.bbclass in core? Apr 13 12:53:02 mago_: it got some more work after the gist in the form of a subcommand based tool, see https://github.com/kergoth/bb — but that hasn't gotten much love recently either Apr 13 13:25:32 what is the best way to debug why a package cannot be found by bitbake? Apr 13 13:29:11 pompomJuice: cannot be found in what context? Apr 13 13:29:19 what's the actual message you are receiving? Apr 13 13:29:55 Nothing PROVIDES 'connman-tests. Close matches: Apr 13 13:30:23 The recipies are exaxtly the same as in 1.6 Apr 13 13:30:27 and in 1.6 that works Apr 13 13:30:48 for some reason bitbake wont find the ${PN}-tests packages Apr 13 13:31:19 connman-tests is a package (a runtime target) Apr 13 13:31:35 and you appear to be attempting to treat it as a build-time target Apr 13 13:31:58 what exactly are you building when you get that error? Apr 13 13:31:59 Yea that used to work Apr 13 13:32:19 I am building it like in the past: bitbake connman-tests Apr 13 13:32:20 I'm fairly sure that it didn't Apr 13 13:32:25 no Apr 13 13:32:26 mmm Apr 13 13:32:30 bitbake connman Apr 13 13:32:45 the connman recipe builds the connman-tests package Apr 13 13:32:47 let me check Apr 13 13:33:42 yea only if you have that package as a dependency somewhere else, for example if you have included it inside your packages Apr 13 13:34:09 oh my Apr 13 13:34:16 seems I am mistaken Apr 13 13:34:22 but how Apr 13 13:35:00 you are correct Apr 13 13:35:04 it never worked Apr 13 13:35:10 somehow I remember it working Apr 13 13:35:33 if you point to it via RDEPENDS or IMAGE_INSTALL, yes, that works - because there runtime targets are expected Apr 13 13:35:48 testing... Apr 13 13:42:24 thanks bluelightning Apr 13 13:42:32 np Apr 13 14:21:26 how should that works? Apr 13 14:21:30 zos Apr 13 14:21:32 ups Apr 13 14:35:32 when I use FAT USB sticks on YP generated systems, when I insert them on a Win7/8 system it claims to repair unexistent corruption. Do you know why? Apr 13 14:35:58 apatr I should avoid to use Win :-D Apr 13 14:36:07 s/apatr/apart Apr 13 14:36:53 I mean, does that happen to me only? Apr 13 14:51:14 mckoan: are you creating an fs on the disk directly and not using a partition table? Apr 13 14:51:32 windows will expect a partition table with a single fat partition Apr 13 14:52:01 kergoth: not sure, IIRC I created the USB disk partition with Win Apr 13 14:53:01 seems unlikely, but *shrug* Apr 13 14:53:05 dunno Apr 13 14:53:45 kergoth: np, thx anyway Apr 13 14:56:47 * mckoan wonders if he's the only one to have seen this problem Apr 13 15:09:56 mckoan: are you unmounting the sticks? Apr 13 15:13:26 XorA: doh! catch it! you're right Apr 13 15:14:19 hah Apr 13 15:14:20 bmaptool #ftw Apr 13 15:14:35 it will not let you write a raw image to a mounted device Apr 13 15:14:48 http://gnuradio.org/redmine/projects/gnuradio/wiki/Copy_an_image_file_to_the_SD_card Apr 13 15:14:54 is a work in progress Apr 13 15:15:44 mckoan: windows since vista I think clears the dirty flag on the fat when its finished writing files and there are none open, which is why you can pull them out of windows live Apr 13 15:19:47 Crofton|work: huh, nice, never heard of bmaptool, looks sweet Apr 13 15:20:02 Crofton|work: one of those - why'd it take everyone so long to do that, things.. Apr 13 15:20:03 yeah, much faster, especially when you have a sparse file Apr 13 15:20:14 likely the best thing to come from Tizen Apr 13 15:20:15 obvious in retrospect Apr 13 15:20:19 * kergoth nods Apr 13 16:31:40 Someone please sed -i -e s/defice/device/g $(git grep -l defice) Apr 13 16:41:37 i think it is an unfortunate mistake that gcc's EXTRA_OECONF_INITIAL does not contain at least --with-arch* and --with-tune* and maybe --with-abi* or am i missing the spot where these are set? Apr 13 16:48:29 any libvirt gurus around? **** ENDING LOGGING AT Tue Apr 14 02:59:58 2015