**** BEGIN LOGGING AT Tue Jul 02 02:59:58 2013 Jul 02 05:10:44 I'm trying to bitbake opkg-utils, but I get ERROR: Function failed: Fetcher failure for URL: 'git://git.yoctoproject.org/opkg-utils;protocol=git'. Unable to fetch URL from any source. Jul 02 05:11:42 ERROR: Logfile of failure stored in: /home/pieter/work/Connexionz/angstrom/setup-scripts/build/tmp-angstrom_v2012_12-eglibc/work/x86_64-linux/opkg-utils-native-0.1.8+gitAUTOINC+757a1664a440c60e8126443bf984e4bdf374c327-r0/temp/log.do_fetch.25207 Jul 02 05:11:43 ERROR: Task 96 (virtual:native:/home/pieter/work/Connexionz/angstrom/setup-scripts/sources/openembedded-core/meta/recipes-devtools/opkg-utils/opkg-utils_git.bb, do_fetch) failed with exit code '1' Jul 02 05:18:29 The git2_git.yoctoproject.org.opkg-utils.tar.gz file does get downloaded, but not extracted. Jul 02 05:21:10 ERROR: Fetcher failure: Fetch command failed with exit code 2, output: Jul 02 05:21:10 tar: .: Cannot change mode to rwxr-xr-x: Operation not supported Jul 02 05:21:10 tar: Exiting with failure status due to previous errors Jul 02 05:21:10 DEBUG: Python function base_do_fetch finished Jul 02 05:21:10 DEBUG: Python function do_fetch finished Jul 02 05:21:11 ERROR: Function failed: Fetcher failure for URERROR: Fetcher failure: Fetch command failed with exit code 2, output: Jul 02 05:21:16 tar: .: Cannot change mode to rwxr-xr-x: Operation not supported Jul 02 05:21:18 tar: Exiting with failure status due to previous errors Jul 02 05:21:20 DEBUG: Python function base_do_fetch finished Jul 02 05:21:22 DEBUG: Python function do_fetch finished Jul 02 05:21:24 ERROR: Function failed: Fetcher failure for URL: 'git://git.yoctoproject.org/opkg-utils;protocol=git'. Unable to fetch URL from any source.L: 'git://git.yoctoproject.org/opkg-utils;protocol=git'. Unable to fetch URL from any source. Jul 02 05:21:28 Any ideas? Jul 02 13:55:24 bluelightning: is today good day for those 2 backports? ;) Jul 02 13:55:42 JaMa: let me just send out these fixes first :) Jul 02 13:55:57 I'm still unable to actually apply them, RP will have to do that Jul 02 13:58:04 ah ok Jul 02 13:58:38 sorry for bugging you about that, it's just our fastest builder hitting it quite often :) Jul 02 14:00:40 hi there Jul 02 14:21:00 Struggling with kernel recipe. I build one based on linux-yocto-custom.bb. My recipe does 3 things. It takes kernel from SRC_URI = "git://my-internal-repo.git;branch=my-linux-3.8.y", it applies a patch by SRC_URI += "file://my-kernel.patch" and it uses SRCREV variable to point to certain commit number. Kernel config is placed in meta/cfg/kernel-cache/bsp/mymachine/my.cfg. Jul 02 14:21:33 Everything works fine, Yocto picks up proper commit, applies a patch and picks proper .config file. Jul 02 14:22:58 Now I just change SRC_URI to "git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git;branch=linux-3.8.y" and Yocto checks out proper branch, commit, it applies a patch. After that it creates totally different bzImage and ramdisk. I think it does not pick up proper .config file and have no idea why. Jul 02 14:23:53 Could you please help me a little bit withe debugging this issue? 1st thing I would like to do is to get to know which .config file Yocto picked up during this build. Where do I find this info? Jul 02 14:27:35 Krz_: hi Krzysztof - it's Paul :) Jul 02 14:27:59 Krz_: is your patch adding the config? Jul 02 14:28:08 yes Jul 02 14:28:19 I tested manually whole process beforehand Jul 02 14:30:39 Krz_: you can inspect the .config produced by looking in the WORKDIR for the kernel Jul 02 14:30:55 I was just wondering - maybe Yocto flow is: 1. git checkout, 2. pick up .config 3. git apply patch Jul 02 14:31:23 I would be able to fit this flow into what I see :) Jul 02 14:31:24 Krz_: patching should be well before that Jul 02 14:32:03 bluelightning: do you tag the bugs for easy, moderate, hard? I will devote my 3 months to OE and I would like to contribute as much as possible Jul 02 14:32:14 Does anyone know what the error "No recipes available for X.bbappend" means when I have written an X.bbappend to installer a newer version of X? Jul 02 14:32:57 eren: we don't at the moment, no... we have the "janitor" bugs that might be of interest Jul 02 14:33:04 bluelightning: yeah sure Jul 02 14:33:27 eren: https://wiki.yoctoproject.org/wiki/Janitors#Janitor_entries_in_Bugzilla Jul 02 14:33:52 Stygia: the bbappend version must match an existing recipe Jul 02 14:34:38 I'm interested in hamradio and stuff, as you know. Right now, I am building for qemu with aprx in it Jul 02 14:34:45 I am in the process of packaging ax25 utilities Jul 02 14:35:23 ah ok, cool, I guess they will go into meta-hamradio? Jul 02 14:37:10 bluelightning: yeah sure Jul 02 14:37:44 I've got a mini210s board which we decided to use for our applications. However, there is no support in OE right now Jul 02 14:37:53 we first decided to go with ALIX 3D3, but it is obsolete Jul 02 14:38:13 if time allows, I will probably play with mini210s but there are a lot work for this board Jul 02 14:38:16 I'm not sure if I can handle it Jul 02 14:39:15 if you can pick up a kernel config from somewhere else + whatever bootloader / userspace stuff is required it shouldn't be too difficult to get a BSP working for it Jul 02 14:39:36 bluelightning, Thank you. Jul 02 14:40:27 bluelightning: the problem is that friendlyarm ships it with 2.6.x kernel, integrates patches for user buttons, sound, leds etc Jul 02 14:40:37 I couldn't find patches for 3.8+ kernels Jul 02 14:40:49 eren: ah, that could make things trickier Jul 02 14:40:52 I checked the kernel config for the latest kernel, it doesn't have anything for Mini210s board Jul 02 14:41:20 assuming they didn't upstream support for the board? Jul 02 14:41:25 bluelightning: no Jul 02 14:41:29 I guess Jul 02 14:41:34 hmm, shame if that's the case Jul 02 14:41:41 if they had, the patches would have been in the current kernel, right? Jul 02 14:42:02 eren: right Jul 02 14:45:04 bluelightning: I can go without user buttons or leds. However, I am not sure if we can boot this board up Jul 02 14:45:10 it uses something called superboot Jul 02 14:45:18 alright, checked out - now I know for sure that Yocto is using different .config file for SRC_URI pointing to git.kernel.org Jul 02 14:45:25 I'm used to know how uboot works from beagleboard, not sure about superboot Jul 02 14:45:48 heyho Jul 02 14:46:05 eren: don't know anything about that board I'm afraid... Jul 02 14:46:26 anyone having trouble with "git ls-remote git://arago-project.org/git/projects/boot-monitor.git DEV.MCSDK-03.00.00.07 could not be run: None" ? Jul 02 14:46:40 bluelightning: okkie, no problem Jul 02 14:46:44 thanks for your time Jul 02 14:47:53 crazy_imp: 859fea1403d69f98723cefc0ba505e1ac837f13b refs/tags/DEV.MCSDK-03.00.00.07 Jul 02 14:52:36 denix: interesting, i get 64c3aae8986eb05aa9b7c326a9432810839120b1 refs/tags/DEV.MCSDK-03.00.00.07 Jul 02 14:53:16 err, wait, that was for the wrong repo :D Jul 02 14:55:48 denix: i get the same output like someone got before http://pastebin.com/zNgHqRWr , line 2-4. and http://logs.nslu2-linux.org/livelogs/yocto/yocto.20130221.txt tells me, there had been some problems with the arago-project gits before Jul 02 14:55:59 ah, YPTM is starting? Jul 02 14:56:11 _Lucretia_: did the problem just vanished back then? Jul 02 14:57:33 YPTM: davest is on the call Jul 02 14:57:48 YPTM: eren is on the call Jul 02 14:57:59 crazy_imp: right now there are no problems with the server, but it might happen occasionally... Jul 02 14:58:59 YPTM: Mark is here... Jul 02 14:59:52 YPTM: Saul is Here Jul 02 14:59:54 YPTM: Scott Rifenbark here Jul 02 15:00:07 denix: k Jul 02 15:00:15 YPTM: Tom Z here Jul 02 15:00:22 YPTM: Corneliu joined Jul 02 15:00:37 YPTM: ross joined Jul 02 15:00:41 YPTM: Denys is here Jul 02 15:00:48 rburton: YPTM: welcome to the tech meeting, please let me know who's on the bridge. Jul 02 15:00:50 YPTM: Michael Halstead present. Jul 02 15:00:57 YPTM: belen joined Jul 02 15:01:05 YPTM: AlexG here Jul 02 15:01:10 YPTM: Paul Eggleton is on Jul 02 15:01:32 YPTM: Laurentiu Palcu joined Jul 02 15:01:42 YPTM: Mihai L here Jul 02 15:02:37 YPTM: Any opens? Jul 02 15:02:47 YPTM: Nitin is on the bridge Jul 02 15:03:13 1.5 overall test plan: https://wiki.yoctoproject.org/wiki/index.php?title=Yocto_1.5_Overall_Test_Plan Jul 02 15:11:21 1.5 overall test plan: https://wiki.yoctoproject.org/wiki/index.php?title=Yocto_1.5_Overall_Test_Plan Jul 02 15:11:51 YPTM: Beth is dialing Jul 02 15:12:14 pidge_: 73 Jul 02 15:13:24 eren: ha. haven't seen that in a while. 73s to you as well Jul 02 15:13:44 thanks! :) Jul 02 15:27:59 * davest drops YPTM for another meeting Jul 02 15:30:59 Matthew Weigel here Jul 02 15:33:45 Song_Liu: I must drop the call. Thanks for the review! :) Jul 02 15:33:46 I've a few things as well, waiting Jul 02 15:36:41 eren: you know about the meta-hamradio layer? Jul 02 15:36:55 ops Jul 02 15:36:57 just dropped Jul 02 15:36:59 1 min Jul 02 15:37:14 :( Jul 02 15:37:26 eren: I just mentioned on your behalf that you're the author of meta-hamradio Jul 02 15:37:27 ha, obviously you do know about meta-hamradio :) Jul 02 15:37:31 :) Jul 02 15:37:52 eren: one day I'll have to ping you about ax.25 stuff. Jul 02 15:38:44 ah Jul 02 15:38:46 I'm now connected Jul 02 15:38:56 seems it's finished :) Jul 02 15:39:00 yes :-) Jul 02 15:39:04 pidge_: okkie, I'm working on ax25 utilities Jul 02 15:39:20 I'll package them in the following days Jul 02 15:39:28 however, there are problems with enabling ax25 configurations in the kernel Jul 02 15:39:34 I don't know how we can handle them Jul 02 15:39:52 eren: are they still utilizing pulse for sound card based modems? Jul 02 15:40:15 pidge_: last time I used them (ax25-utils), I interfaced them with KISS modem Jul 02 15:40:16 * pidge_ had no amount of issues getting those working. Jul 02 15:40:31 so, kissattach, ax25-send and all the stuff worked 2 years ago Jul 02 15:40:33 eren: which kernel are you looking at and what kind of issues, it should be fairly straight forward (unless it's complex) Jul 02 15:40:34 I don't think they are broken Jul 02 15:40:53 sgw_: I will just need to enable configuration already existing in kernel in distribution config Jul 02 15:41:31 I will probably extend poky reference distribution, add hamradio stuff into the image, enable ax25 configuration, etc Jul 02 15:43:19 eren: soundmodem iirc was all types of broken, at least on the distro I was trying to get working. I backreved all the way to soundmodem using OSS trying to get the thing working. no dice. Jul 02 15:43:32 although it could have entirely been pebkac Jul 02 15:43:46 eren: can you give me some idea of which configs, and I can maybe give you an example. Jul 02 15:43:54 * sgw_ has to drive kids, brb Jul 02 15:44:42 pidge_: soundmodems are mostly problematic. I don't know if there is a decent application that is capable of using ALSA and decoding decently Jul 02 15:45:48 however, KISS TNC's are doing the job well, and I managed to get aprx (digipetar and i-gate) daemon to work with ax25 utitilies Jul 02 15:46:06 of course, it was 2 years ago :) Jul 02 15:46:17 eren: That's kind of the conclusion I came to. Ohhh... need. new. hardware. Jul 02 15:46:42 pidge_: I also need soundmodem working, I will look at it again Jul 02 15:47:02 at that time, I already had TNCx, still I have Jul 02 15:47:19 eren: Oh good, someone else to take up my banner of frustration :) Jul 02 15:47:32 pidge_: heheh :) Jul 02 15:47:51 I would be very like to have 2 APRS station running with TNC and regular sound card Jul 02 15:48:20 ok, I will let you know about the status of ax25 utitilies Jul 02 15:48:34 these days, I will be working on libax25, ax25-utils, ax25-apps Jul 02 15:48:55 I'm building for qemu, and I will build for beagleboard as well to see if they are really working Jul 02 15:48:58 eren: It sure would. You're going to make me have to get my antenna back up (minor wind storm collapse) Jul 02 15:49:22 pidge_: what's your QTH? Jul 02 15:49:51 I will be operating from DA0UP until September 30 Jul 02 15:50:12 we may try your antenna setup, if you want Jul 02 15:50:36 eren: portland, or Jul 02 15:50:47 KF&ESD Jul 02 15:50:56 ergh /&/7 Jul 02 15:52:47 my ability to hit EU is limited by location. furthest I've been able to hit in this location going east is NYC. West, India. I need to get my dipole up further. Jul 02 15:54:48 pidge_: ah, seems hard, but it's doable. I worked K1UI on 20 meters today Jul 02 15:55:16 we have dipole with rotor, flex 5000a and acom 2000a linear amplifier Jul 02 15:55:29 the power that I used was around 800W Jul 02 15:58:11 oh, I'm much simplier than that. long wire folded sloper cut to 40m that is maybe 40 feet up with a 100w icom 706mkII and a 2m jpole. for dxing it's an R390a that dims the neighborhood when it's on. Jul 02 16:00:43 pidge_: ah I see Jul 02 16:01:29 well I remmeber working USA with 100W on mobile station on 40m, however it was just in the grey line Jul 02 16:01:36 propagation was really good for 15 minutes Jul 02 16:01:50 then disappeared Jul 02 16:03:10 pidge_: it was nice to talk to you. I need to go out Jul 02 16:03:21 see you later! DE TA1AET Jul 02 16:03:31 s/later/tomorrow Jul 02 16:04:19 73s Jul 02 16:04:24 eren ^^ Jul 02 16:12:54 * kergoth sighs Jul 02 18:37:20 anyone know offhand if postinsts are run for sdk sysroots? Jul 02 18:37:25 I'm guessing no Jul 02 18:47:20 Hi I am trying to build fsl-image-minimal and receiving this error - Summary: 1 task failed: /home/lahar/QorIQ-SDK-V1.4-2013/meta-fsl-ppc/recipes-devtools/qemu/qemu_fslgit.bb, do_install Jul 02 18:47:35 Can somebody please help? Jul 02 18:51:56 contact freescale, i'ds ay Jul 02 19:11:40 kergoth: ok thanks! Jul 02 19:20:08 apparently postinsts are *not* run for the sdk, that's a problem.. Jul 02 20:03:27 Working on a package. Wanted to get the executable started running as a daemon. Found that the tar ball included an rpm directory with an initialization script. Decided to use that, with modifications. Found out that the script has a call the the daemon utility which isn't on core-image-sato, the machine I'm using. What's the appropriate thing to do next? Jul 02 20:06:59 libslack.org/daemon Jul 02 20:10:41 mulhern: start-stop-daemon is what your looking for, look at cron Jul 02 20:11:10 mulhern: meta/recipes-extended/cronie/cronie Jul 02 20:13:19 sgw_: OK, I see it. Jul 02 21:01:23 woo hoo, successfully did an ls-remote on one of our https repositories using the git from buildtools-tarball on my centos5 vm Jul 02 21:25:41 I create a directory called /var/log/snort in my ${WORKDIR}/image subdirectory. It gets to my package* directories but never makes it to the root filesystem. However, /usr/lib/snort_dynamicrules which is created in the same way does make it to the root filesystem. What am I doing wrong? Jul 02 21:29:19 mulhern: /var is a volatile directory so is created at boot time, you could create it in your init script or use the top level /var/log dir Jul 02 22:52:35 woot. can now do builds on my centos 5.9 VM with buildtools-tarball. expect a pile of patches for oe-core :) Jul 02 22:53:19 mulhern: you can add it to the volatiles processing and keep your current path(s) if you want... Jul 02 22:55:09 without pulling in the ro_rootfs patches, you would need to create a file for /etc/default/volatiles Jul 02 22:55:58 what's the status of ro rootfs in master, still lots pending? Jul 02 22:56:11 afaik, yes Jul 02 22:56:50 i ported the gist of those patches to my local oe-classic build but that's it for me Jul 02 22:57:17 and my eyes are getting to old to read the darker irc colors... Jul 02 22:59:10 kergoth: depends what you want in the rootfs Jul 02 22:59:44 kergoth: it works for core-image-minimal at the very least Jul 02 22:59:59 heh, true, every piece of softwar that writes somewhere has to be tested and potentially fixed piecemeal Jul 02 23:00:05 right Jul 02 23:00:11 and no postinstalls allowed either Jul 02 23:00:27 nothing at runtime anyway. which means pregenerated ssh host keys, blah blah Jul 02 23:00:31 right Jul 02 23:06:04 i didn't take it quite that far with oe-classic Jul 02 23:06:32 it still needs to run the configure script on first boot then sets everything ro at the end of the boot Jul 02 23:06:52 so it still allows for postinst, etc Jul 02 23:08:22 once it boots up the first time it works as expected, albeit with some /etc files symlinked to a writable location Jul 02 23:08:50 pretty close to full ro-rootfs but not quite... Jul 02 23:10:42 mr_science: cool, sounds like it was good progress Jul 02 23:16:34 meh Jul 02 23:16:58 not sure how to ship a bunch of units if i have to pregenerate keys Jul 02 23:18:24 kergoth: those patches don't translate well to classic, so it didn't quite work out as i originally intended Jul 02 23:18:56 should be much better with current oe Jul 02 23:19:21 yeah, keys are problematic, but as long a sthe host has some sort of writable storage you're okay, or if you don't mind regenerating the keys on startup. *shrug* Jul 02 23:21:51 yup, that's my current work-around, ie, let dropbear make a key on first boot before things get remounted ro, as well as symlinking a few things like network/interfaces Jul 02 23:22:55 let's just say i have some extensive rootfs/image post-process functions in the main image recipes... **** BEGIN LOGGING AT Wed Jul 03 01:02:05 2013 **** ENDING LOGGING AT Wed Jul 03 02:59:57 2013