**** BEGIN LOGGING AT Sun Nov 06 03:00:00 2016 Nov 06 18:34:55 hey, what's the recommended way of setting stuff in config files that ends up in /etc/ on the target? Nov 06 18:35:14 e.g. I'd like to set RuntimeWatchdogSec in /etc/systemd/system.conf Nov 06 19:05:55 zyp: You can find the recipe which installs /etc/systemd/system.conf and override that Nov 06 19:06:44 usually you'd do that in your own layer via a bbappend file Nov 06 19:48:28 yeah, I see http://git.yoctoproject.org/clean/cgit.cgi/meta-intel-edison/tree/meta-intel-edison-distro/recipes-core/systemd/systemd_%25.bbappend is doing just that by overwriting it in do_install_append Nov 06 19:48:42 I guess another approach would be applying a patch to the file in the source Nov 06 19:49:36 Hi all, I'm running into a build error when starting to compile an AstroidOS image: http://pastebin.com/HZcQp0dN Nov 06 19:49:53 it fails early in pseudo-native. Nov 06 19:50:05 I was asked to raise the question better here. Nov 06 19:50:19 build system is x86_64 Arch Linux Nov 06 19:50:46 Can somebody please have a look at the error? Nov 06 19:51:15 andyrtr, looks like code written for python2 is being executed with python3 Nov 06 19:51:43 python3 requires parantheses around the arguments for a print statement, python2 doesn't Nov 06 19:52:31 I have exported PYTHON=python2 before the build Nov 06 19:52:40 not enough? Nov 06 22:14:58 is it possible to set just a single recipie to use a different version of autoconf-native? Nov 06 22:15:16 I tried setting PREFERRED_VERSION_autoconf-native = "2.69" in the recipie but it had no effect Nov 06 22:34:14 MooGoo: it isn't possible to do that I'm afraid Nov 06 22:35:58 hm, so I guess I have to use 2.69 for everything or backport the autoconf files in the code I'm trying to configure Nov 06 22:36:32 god I wish I could use oe-core... Nov 06 22:58:24 What is your specific issue? Nov 06 23:12:34 just trying to install this one recipie....libbsd I guess, and it required autoconf >= 2.67 Nov 06 23:13:23 oe-classic uses less than that, I had to backport an autoconf version from oe-core Nov 06 23:14:25 I could just bump up to 2.69 for everything but I have some concerns about how that will affect other packages, but it'll probably be okay Nov 06 23:18:19 MooGoo, do not use oe-classic Nov 06 23:18:31 it is for reference only Nov 06 23:18:36 yes....tell that to multitech Nov 06 23:18:52 if you have an employer that insists, we have a letter stating this Nov 06 23:19:16 if you want to take this off irc, use board@openembedded.org for contact Nov 06 23:19:24 its not my employer, it's just this box we got from them all their files are setup for oe-classic Nov 06 23:19:52 unless there is some somewhat reasomable way to port to core Nov 06 23:19:57 reasonable Nov 06 23:21:33 but I'll admit most of the problems I've had with this stem back to it being based on oe-core, it also won't even build stock on anything greater than Ubuntu 12... Nov 06 23:21:40 s/core/classic Nov 06 23:21:44 yeah Nov 06 23:22:32 th elast serious commit to oe-classic was Dec 2012. florian commited some security fixes criticl to his personal needs through Sep 2014 Nov 06 23:23:00 but since gcc updates in modern distros, non of the things that use gcc to build will likely work Nov 06 23:23:31 so use vm's for old Linux, and carefully review CVE's Nov 06 23:23:55 tremendous pain making something shippable from that era oe build :( Nov 06 23:24:04 figured Nov 06 23:24:25 how difficult is it to port to core in general? Nov 06 23:24:35 depends what you are porting Nov 06 23:24:41 shouldn't be for sane stuff Nov 06 23:24:56 old kernesl could be painful due to gcc updates Nov 06 23:25:50 sure, worried about all the arch/board specific stuff included by multitech Nov 06 23:25:57 yeah Nov 06 23:26:05 that is always the hard stuff Nov 06 23:26:18 old kernels need patchin gjust to build on modern gcc Nov 07 01:13:58 MooGoo: this may be helpful: http://www.openembedded.org/wiki/Migrating_metadata_to_OE-Core Nov 07 01:14:24 mostly concentrates on recipes, and there are a lot of specifics you may not encounter **** ENDING LOGGING AT Mon Nov 07 02:59:59 2016