| 00:37.47 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 01:00.54 | *** join/#oe oldtopman (~oldtopman@unaffiliated/oldtopman) |
| 01:11.46 | lerela | I've got a "QA Issue: non debug package contains .debug directory", but I can't target the precise .debug path in $FILES_{PN}-dbg as it is dynamic (it's the result of a build process)... how can I ignore this directory? |
| 01:13.28 | bluelightning | lerela: can't you wildcard it? |
| 01:14.11 | lerela | */.debug doesn't work so I probably would have to get more specific about the directory structure and that is not very future proof... |
| 01:15.14 | bluelightning | we fixed this in master and the upcoming release to automatically package any .debug directory |
| 01:16.01 | bluelightning | for now you can work around it either by adding the directory possibly using wildcards, delete it, or disable the warning |
| 01:16.44 | lerela | alright I'll do that, thank you bluelightning |
| 01:36.01 | lerela | if I remove them from the install directory with find ${D} | grep .debug | xargs rm -rf in do_install, Bitbake still complains because they are still in packages-split.. The doc says that "The do_package and do_packagedata tasks combine to analyze the files found in the D directory and split them into subsets based on available packages and files.", so I don't get how they can end up in packages-split if they're not in $D (I'v |
| 01:36.02 | lerela | e checked, the find deletes them) |
| 01:42.36 | *** join/#oe rburton (~Adium@home.burtonini.com) |
| 01:53.44 | bluelightning | lerela: something is going wrong there because any change at do_install *will* propagate to packaging |
| 02:17.30 | lerela | well, I could only make it work by ignoring the warning, the recipe looks like this: http://dpaste.com/0NG1KX8 and without the warning skip, the .debug files are not in image/ when it fails but they are in packages/ and packages-split/, which triggers the warning |
| 02:17.44 | lerela | the end of the bitbake log is here: http://dpaste.com/1JV15DY |
| 02:18.04 | lerela | I'm probably missing something |
| 02:22.17 | slapin | hi, all! |
| 02:22.30 | Crofton|work | hi |
| 02:24.06 | slapin | how I should build kernel modules now with separate kernel builddir? I created https://gist.github.com/3a5685423a1a73c48da0 but it doesn't build properly - MODPOST is not happy about missing Module.symvers |
| 02:24.59 | slapin | the complaints - https://gist.github.com/7044cc7bfb9c752e65d8 |
| 02:25.26 | slapin | what am I doing wrong? |
| 02:49.31 | bluelightning | lerela: ah, I think I figured out what might be happening |
| 02:49.48 | bluelightning | lerela: those files get split out as part of do_package |
| 02:50.10 | bluelightning | (or at least, the ones that are still there after your removals do) |
| 02:55.15 | bluelightning | you could disable debug package splitting entirely if that's what you really want |
| 02:55.42 | bluelightning | INHIBIT_PACKAGE_DEBUG_SPLIT = "1" within the recipe will do that |
| 03:06.53 | *** join/#oe hrw (~hrw@redhat/hrw) |
| 03:48.44 | *** join/#oe khem (~khem@unaffiliated/khem) |
| 04:11.11 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 04:21.04 | *** join/#oe khem (~khem@unaffiliated/khem) |
| 04:25.22 | *** join/#oe rob_w__ (~bob@ppp-93-104-103-169.dynamic.mnet-online.de) |
| 05:24.04 | *** join/#oe tbr (dm8tbr@mail.bfst.de) |
| 05:38.22 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 05:40.21 | *** join/#oe khem (~khem@unaffiliated/khem) |
| 05:40.52 | *** join/#oe onoffon (~khem@unaffiliated/khem) |
| 05:53.12 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 06:41.53 | *** join/#oe morphis (~morphis@p5480DD5E.dip0.t-ipconnect.de) |
| 07:04.12 | *** join/#oe dlan (~dennis@gentoo/developer/dlan) |
| 07:07.11 | Nilesh | I'm unable to disable qtwebkit even after cherry-pic 40b7a93a3c6cf05136c606824c1cea848b75bcb6 which gave qtquick1_git_bb as https://paste.kde.org/ppifvbmv6 and my local.conf looks like https://paste.kde.org/prbxl1zzx....its still fetching qtwebkit, can someone help |
| 07:11.17 | Nilesh | I have meta-qt fido branch |
| 07:14.48 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 07:17.35 | *** join/#oe melonipoika (~jose@194.9.252.237) |
| 07:19.27 | *** join/#oe joshuagl (joshuagl@nat/intel/x-mhyleabiqxggcxud) |
| 07:44.40 | zeechs | nerdboy, Over a week of struggling with that issue, and the solution could be that simple ? haha! Thank you so much, I will try that today and give you an update. |
| 07:53.12 | *** join/#oe jbrianceau_away (uid10952@gateway/web/irccloud.com/x-tpnmyfqkkcyglptt) |
| 07:59.14 | *** join/#oe Nilesh_ (3ba0cff3@gateway/web/freenode/ip.59.160.207.243) |
| 08:00.48 | *** join/#oe apelete (~apelete@rec31-1-78-194-210-47.fbxo.proxad.net) |
| 08:05.52 | *** join/#oe ant_work (~ant__@host213-59-dynamic.50-79-r.retail.telecomitalia.it) |
| 08:09.46 | *** join/#oe mago_ (~mago@88.131.56.168) |
| 08:14.32 | Nilesh_ | I'm unable to disable qtwebkit even after cherry-pic 40b7a93a3c6cf05136c606824c1cea848b75bcb6 which gave qtquick1_git_bb as https://paste.kde.org/ppifvbmv6 and my local.conf looks like https://paste.kde.org/prbxl1zzx .its still fetching qtwebkit, can someone help |
| 08:14.51 | Nilesh_ | I've fido meta-qt |
| 08:16.48 | *** join/#oe florian (~fuchs@Maemo/community/contributor/florian) |
| 08:32.54 | *** join/#oe zeechs_ (~zeechs@ANancy-653-1-538-126.w90-6.abo.wanadoo.fr) |
| 08:34.26 | *** join/#oe edbart (ebartosh@nat/intel/x-awffdqcukszqxwzg) |
| 08:38.03 | *** join/#oe ao2 (~ao2@cl-35.trn-01.it.sixxs.net) |
| 08:44.39 | abelloni | Nilesh_: did you remove qtwebkit from the PACKAGECONFIG ? |
| 08:49.39 | Nilesh_ | abelloni: PACKAGECONFIG_remove_pn-qttools= "qtwebkit" PACKAGECONFIG_remove_pn-qtquick1= "qtwebkit" from local.conf |
| 08:51.40 | Nilesh_ | applied 40b7a93a3c6cf05136c606824c1cea848b75bcb6 commit too |
| 08:53.03 | abelloni | shouldn't that be PACKAGECONFIG_pn-qtquick1_remove ? |
| 09:00.24 | Nilesh_ | abelloni: I added that but still its fetching |
| 09:00.31 | *** join/#oe jku (jku@nat/intel/x-kpjaxfbxvlematqy) |
| 09:02.07 | *** join/#oe coolmouse (~coolmouse@113.201.102.180) |
| 09:05.38 | *** join/#oe daro (~daro@88.131.56.168) |
| 09:06.44 | mckoan | good morning |
| 09:09.02 | *** join/#oe t0mmy (~tprrt@217.114.201.133) |
| 09:13.41 | *** join/#oe jonathanmaw (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 09:20.01 | *** join/#oe coolmouse (~coolmouse@113.201.102.180) |
| 09:29.27 | *** join/#oe coolmouse (~coolmouse@113.201.102.180) |
| 09:32.12 | *** join/#oe coolmouse_ (~coolmouse@113.201.102.180) |
| 09:40.55 | *** join/#oe ecloud (quassel@nat/digia/x-nmlusyaxledehpbq) |
| 09:41.14 | *** join/#oe scot (~scot@130.164.62.160) |
| 09:47.34 | *** join/#oe yann||work (~yann@LFbn-1-1026-146.w86-247.abo.wanadoo.fr) |
| 10:11.49 | *** join/#oe belen (~Adium@134.134.139.74) |
| 11:26.09 | *** join/#oe zxc678 (zxc678@101.65.32.207) |
| 11:26.29 | *** join/#oe jonathanmaw_ (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 11:26.46 | *** join/#oe jonathanmaw (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 11:26.47 | zxc678 | hi, anyone expert can answer my question, it is about lvds configuration |
| 11:29.20 | *** join/#oe jonathanmaw (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 11:29.38 | mckoan | ~ask |
| 11:29.38 | infobot | Questions in the channel should be specific, informative, complete, concise, and on-topic. Don't ask if you can ask a question first. Don't ask if a person is there; just ask what you intended to ask them. Better questions more frequently yield better answers. We are all here voluntarily or against our will. |
| 11:31.08 | *** join/#oe spaszPL (~spasz@188.121.6.226) |
| 11:31.31 | spaszPL | Hi. Who is provider of etc/group and etc/passwd? |
| 11:31.52 | spaszPL | base-passwd.ipk in my case doesn't have this files inside. |
| 11:32.41 | *** join/#oe ldnunes (~ldnunes_@179.105.251.106) |
| 11:34.24 | zxc678 | ok,thanks, my question is i use wandboard solo board to light my lcd panel, i modify the dtsi file to get the ldb driver installed when os run it display the information, i also modify the bootargs env, set video=ldb, but still display nothing |
| 11:34.48 | zxc678 | so i think to know what is error with my configuration for lvds |
| 11:40.58 | jku | spaszPL: it is base-passwd but the files get created in a preinst script (see the recipe: it's a bit of a hack) |
| 11:45.38 | *** join/#oe jonathanmaw (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 11:49.24 | *** join/#oe berton (~fabio@179.105.251.106) |
| 11:53.42 | *** join/#oe jonathanmaw_ (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 12:00.05 | *** join/#oe adelcast (~adelcast@130.164.62.82) |
| 12:16.34 | *** join/#oe ant_work (~ant__@host213-59-dynamic.50-79-r.retail.telecomitalia.it) |
| 12:18.26 | zxc678 | Hi, I put my question here: https://community.freescale.com/message/620472#620472 because more information describe my question, so pls expert go there to answer it, thanks. |
| 12:47.16 | *** join/#oe edbart (ebartosh@nat/intel/x-jetpduvhxhhywhjo) |
| 13:08.06 | *** join/#oe tasslehoff (~Tasslehof@82.147.55.166) |
| 13:27.05 | *** join/#oe morphis (~morphis@p5480DD5E.dip0.t-ipconnect.de) |
| 13:42.31 | *** join/#oe pidge (~pidge@2a02:8084:ac1:d80:1006:1a7:fed2:8989) |
| 13:55.27 | *** join/#oe joshuagl (joshuagl@nat/intel/x-flmetvbkxabfyqco) |
| 14:10.57 | *** join/#oe paulg_ (~paulg@198-84-239-75.cpe.teksavvy.com) |
| 14:11.04 | *** join/#oe belen (~Adium@134.134.139.74) |
| 14:31.34 | *** join/#oe galak (~galak@104-57-189-103.lightspeed.austtx.sbcglobal.net) |
| 14:35.44 | *** join/#oe Aethenelle (~Aethenell@166.170.220.66) |
| 14:52.42 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 14:55.40 | *** join/#oe ftonello (~felipe@81.145.202.106) |
| 14:59.45 | *** join/#oe madisox (~madison@12.30.244.5) |
| 14:59.50 | *** join/#oe belen1 (Adium@nat/intel/x-onjzxpnxuaqapujd) |
| 15:00.24 | *** join/#oe thaytan (~thaytan@199.7.70.115.static.exetel.com.au) |
| 15:06.26 | *** part/#oe madisox (~madison@12.30.244.5) |
| 15:07.23 | *** join/#oe vdehors (~vdehors@193.56.60.161) |
| 15:13.18 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 15:24.14 | ant_work | JaMa: please pull in -next #116753 . Somehow was flagged as superseded/archived |
| 15:29.51 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 15:30.40 | *** join/#oe flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 15:35.47 | JaMa | ant_work: done |
| 15:36.39 | ant_work | thanks, it was the important one ;) |
| 15:40.18 | *** join/#oe rob_w (~rob@unaffiliated/rob-w/x-1112029) |
| 16:01.13 | *** join/#oe berton (~fabio@179.105.251.106) |
| 16:02.04 | *** join/#oe joshuagl (joshuagl@nat/intel/x-hpjjoroeszoxjxnu) |
| 16:15.37 | *** join/#oe balkamos_ (balkamos@unaffiliated/motley) |
| 16:15.46 | *** join/#oe radhus__ (~radhus@sevh.radhuset.org) |
| 16:16.52 | *** join/#oe darknighte (~darknight@pdpc/supporter/professional/darknighte) |
| 16:18.44 | *** join/#oe Aethenelle (~Aethenell@199.15.128.75) |
| 16:23.27 | *** join/#oe jackmitch|home (~Thunderbi@cpc2-slam8-2-0-cust880.2-4.cable.virginm.net) |
| 16:28.10 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 16:34.31 | *** join/#oe vdehors (~vdehors@193.56.60.161) |
| 16:38.22 | *** join/#oe jonathanmaw (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 17:00.36 | *** join/#oe jonathanmaw (~jonathanm@82-70-136-246.dsl.in-addr.zen.co.uk) |
| 17:09.56 | *** join/#oe rburton (~Adium@home.burtonini.com) |
| 17:19.26 | *** join/#oe edbart (~ebartosh@192.198.151.43) |
| 17:30.10 | *** join/#oe flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 17:33.18 | *** join/#oe chouimat (~mathieu@206-248-139-164.dsl.teksavvy.com) |
| 17:33.25 | chouimat | hi |
| 17:34.05 | *** join/#oe vdehors (~vdehors@193.56.60.161) |
| 17:36.10 | chouimat | is there a recipe I can use as a base for a meta-sdk that will run on the target? |
| 17:37.50 | lerela | to do what? you could check out Texas Instruments sdk |
| 17:38.00 | lerela | in meta-arago |
| 17:38.08 | kergoth | chouimat: for on-target development? |
| 17:38.10 | kergoth | there's packagegroup-core-sdk and packagegroup-self-hosted |
| 17:38.21 | kergoth | the latter being the bits needed to run bitbake on target, iirc |
| 17:38.52 | chouimat | kergoth: hi, it's more like some developer test tools with need to be able to put on the target without having to flash a developer image |
| 17:38.53 | kergoth | if you're looking to ship an sdk for development on a host, not target, that's be populate_sdk or populate_sdk_ext, depending on your needs |
| 17:39.39 | kergoth | you might want to try a populate_sdk_ext, which ships the layers+bitbake+etc and use devtool to build & package & deploy |
| 17:39.47 | chouimat | kergoth: another question, is there a way to add sanity checks? |
| 17:39.59 | kergoth | what sort of sanity checks? |
| 17:40.55 | chouimat | well we have 2 recipes and each time we do changes to them and forget to update the internal version number, the resulting image cannot be flashed since the version number are the same |
| 17:45.09 | kergoth | you can add packaging sanity checks via insane.bbclass, or you can add any arbitrary function to the rootfs/image hooks. i.e. ROOTFS_POSTPROCESS_COMMAND, etc |
| 17:46.34 | chouimat | kergoth: I will take a look at those. as for the devtools, I just want the content of the package-group-devtools in the tarball, anyway I will go look at populate_sdk |
| 17:59.44 | *** join/#oe kristoffer (~kristoffe@ua-83-227-162-207.cust.bredbandsbolaget.se) |
| 18:23.54 | zeechs | nerdboy, thank you for the advice you gave me yesturday (to add kernel-modules). It solved my problem on one board (Topic Miami/Florida) with a usb sound card. I finally got the snd_usb_audio module available. It didn't work on the ZyBo board though, but the problem must be somewhere else. I'm trying to get help on the Diligent forum right now. |
| 18:24.29 | zeechs | as for the odroid, I didn't test but i'm pretty sure it would solve the problem too |
| 18:25.16 | *** join/#oe anarsoul|2 (~kvirc@S0106848dc7ec0367.vs.shawcable.net) |
| 18:36.30 | *** join/#oe caiortp (~inatel@131.221.240.204) |
| 18:56.53 | *** join/#oe jackmitch|home (~Thunderbi@cpc2-slam8-2-0-cust880.2-4.cable.virginm.net) |
| 19:45.46 | *** join/#oe bluelightning (~paul@2406:e007:5383:1:5e51:4fff:febb:401d) |
| 19:45.46 | *** join/#oe bluelightning (~paul@pdpc/supporter/professional/bluelightning) |
| 19:47.36 | *** join/#oe flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 19:48.00 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 19:50.47 | Crofton|work | notes the meta-maker goof balls have a privte copy of python-requests |
| 20:19.34 | Crofton|work | any changes in the way wic creates ext4 file systems |
| 20:20.01 | Crofton|work | I'm having a problem that the card works, but I put it in my fedora box, it will not mont the rootfs afterwards |
| 20:20.20 | Crofton|work | feels like earlier cards I amde do not show this issue |
| 20:35.34 | kergoth | I've seen a case where the ext4 rootfs in the wic image mounted successfully via loopback, but says it needs to be checked when the target tries to mount it. haven't nailed that down yet |
| 20:40.36 | *** join/#oe challinan (~chris@173-10-226-189-BusName-WestFlorida.hfc.comcastbusiness.net) |
| 20:41.39 | Crofton|work | in this case after mounting in pc I see |
| 20:41.53 | Crofton|work | the cat sit down in front of me |
| 20:42.37 | Crofton|work | argues with the cat |
| 20:43.23 | Crofton|work | [ 1.458125] JBD2: Unrecognised features on journal |
| 20:43.23 | Crofton|work | [ 1.462842] EXT4-fs (mmcblk0p2): error loading journal |
| 20:43.49 | Crofton|work | the funny thing is card worked until I put it in desktop to make a copy |
| 20:45.39 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 20:48.04 | kergoth | strange |
| 20:48.24 | Crofton|work | indeed |
| 20:48.59 | kergoth | Hmm, wonder what the status is of efi with qemu |
| 20:49.05 | kergoth | will have to check into that |
| 20:51.47 | Crofton | and for more, insert card in desktop it mounts, but e2fsck does: |
| 20:51.48 | Crofton | [balister@mini ~]$ sudo e2fsck /dev/mmcblk0p2 |
| 20:51.48 | Crofton | e2fsck 1.42.13 (17-May-2015) |
| 20:51.48 | Crofton | /dev/mmcblk0p2 has unsupported feature(s): metadata_csum |
| 20:51.48 | Crofton | e2fsck: Get a newer version of e2fsck! |
| 21:13.56 | Crofton|work | I bet it is related to old kernels |
| 21:16.14 | *** join/#oe khem` (~khem@unaffiliated/khem) |
| 21:16.59 | *** join/#oe jackmitch|home (~Thunderbi@cpc2-slam8-2-0-cust880.2-4.cable.virginm.net) |
| 21:31.11 | JaMa | Crofton|work: there was some patch on ML to disable experimental features in latest e2fsprogs upgrade |
| 21:32.09 | JaMa | "e2fsprogs: do not enable non-stable features by default" |
| 21:32.31 | JaMa | Revert "mke2fs: enable the metadata_csum and 64bit features default" |
| 21:36.55 | Crofton|work | ah |
| 21:37.10 | Crofton|work | but not in master yet? |
| 21:38.22 | lerela | is there any clean way to prevent OE from forcing OpenSSH env variables in image.class in case of a read-only filesystem? |
| 21:39.08 | lerela | I don't think there's a lot of options other than completely overriding read_only_rootfs_hook but who knows |
| 21:40.13 | Crofton|work | buys jama a cart ride |
| 21:40.46 | *** join/#oe yann||work (~yann@nan92-1-81-57-214-146.fbx.proxad.net) |
| 21:41.39 | *** join/#oe Aethenelle (~Aethenell@199.15.128.78) |
| 21:41.59 | rburton | anyone know much about elfutils? specifically does eu-objdump "not work on mips, ppc, arm" still |
| 21:42.23 | rburton | we started deleting it in 2013 |
| 21:42.35 | Crofton|work | what's the script to get a patch from pw? |
| 21:43.13 | *** join/#oe galak (~galak@104-57-189-103.lightspeed.austtx.sbcglobal.net) |
| 21:46.58 | bluelightning | lerela: you have read-only-rootfs in (EXTRA_)IMAGE_FEATURES? |
| 21:47.12 | bluelightning | Crofton|work: pwclient? |
| 21:47.40 | Crofton|work | pe-am from meta-oe |
| 21:47.47 | Crofton|work | where is pwclient? |
| 21:48.02 | Crofton|work | bitbake path? |
| 21:49.37 | bluelightning | I don't think we supply it, I think it's distributed with patchwork |
| 21:49.48 | bluelightning | btw, some of my colleagues are still working on patchwork upgrade / patch testing stuff, I hope to have something to share soon |
| 21:51.24 | Crofton|work | I need a patch out of patchwork |
| 21:53.32 | Crofton|work | http://patchwork.openembedded.org/patch/116903/ |
| 21:53.42 | Crofton|work | hopefully this resolves a nast y issue I am having |
| 22:01.59 | bluelightning | you can just download the mbox |
| 22:02.15 | bluelightning | then git am it |
| 22:02.42 | Crofton|work | yeah, I read the old script :) |
| 22:08.07 | Crofton|work | hmm, I need to copy a 1 GB fioe onto my then sync to dropbox on fast wifi |
| 22:12.32 | lerela | bluelightning, I do, so the read_only_rootfs_hook is called and some variables are hardcoded in /etc/default/ssl in it |
| 22:13.23 | lerela | by the way the INHIBIT trick worked bluelightning, thank you |
| 22:35.29 | bluelightning | lerela: great |
| 22:35.49 | bluelightning | lerela: can you explain what the issue is with those settings being changed? |
| 22:38.17 | *** join/#oe DJW|Home (~djwillis@cpc2-trow6-2-0-cust204.aztw.cable.virginm.net) |
| 22:44.25 | lerela | bluelightning, I do not want my ssh keys to be in the volatile /var/run/ssh, I used a .bbappend to add a different SYSCONFDIR in /etc/default but this is getting overridden by this hook (and the key paths are set in stone in sshd_config_readonly by the openSSH recipe itself) |
| 22:44.54 | lerela | but since everything is hardcoded I might as well mount /var/run/ssh myself in an other folder from my own init script, I guess |
| 22:45.20 | bluelightning | lerela: ok, so perhaps our code in the hook there needs to be a bit smarter |
| 22:45.53 | bluelightning | as you can appreciate all it's trying to do is ensure sshd can start with the rootfs read-only |
| 22:46.23 | lerela | that is true but what that means if I understand correctly is that keys will be regenerated at each boot |
| 22:46.47 | lerela | it is a problem to have a host with constantly changing keys |
| 22:47.07 | *** join/#oe BohemianHacks (boho@gateway/shell/insomnia247/x-mzgajawzdosfpcdl) |
| 22:47.41 | *** join/#oe BohemianHacks (~boho@unaffiliated/bohemianhacks) |
| 22:49.23 | BohemianHacks | Howdy, I'm trying to port linux to a random android tablet with an omap3 process. I have a kernel that is reported to work but I need to find/generate a rootfs. As long as the processor is the same I should be able to use a precompiled rootfs and add the modules I need, correct? |
| 22:49.37 | lerela | I'm not sure there could be a better standard behavior with a ro rootfs of course but perhaps it should be easier to configure this differently so that keys can be stored to permanent storage |
| 23:01.52 | bluelightning | lerela: fundamentally if you want the keys to not to change then you need to have some permanent storage which we don't have by default |
| 23:02.11 | bluelightning | lerela: I agree some way of easily configuring that would be helpful though |
| 23:03.33 | bluelightning | BohemianHacks: in theory yes but you'll need to ensure the versions match up (kernel version and possibly even compiler version, I'm not sure) |
| 23:04.23 | bluelightning | BohemianHacks: our system isn't really designed to be used like that, the assumption is that you build everything (possibly using caches to reduce unnecessary rebuilding though) |
| 23:06.14 | BohemianHacks | bluelightning, well the hardware is very similar to a beaglbone so im hoping to use that |
| 23:06.43 | BohemianHacks | using the oe tools to build a kernel and rootfs, but im working in baby steps for now as the tablet is dated and lacks any kind of useful documentation |
| 23:07.51 | *** join/#oe yann||work (~yann@nan92-1-81-57-214-146.fbx.proxad.net) |
| 23:08.09 | BohemianHacks | bluelightning: Im used to using buildroot or a regular linux kernel makefile. Is it reasonable to import a kernel config into your build system or do I need to generate my own machine profile? |
| 23:08.41 | lerela | bluelightning, I do have permanent storage for sure but I don't want my whole rootfs to be r/w just for those keys to be created once, but yes I probably can hack something |
| 23:09.10 | bluelightning | lerela: if you come up with something generic we'd really appreciate a patch to add it |
| 23:09.32 | bluelightning | BohemianHacks: you can certainly override the kernel config fairly easily |
| 23:10.07 | BohemianHacks | cool beans |
| 23:16.12 | BohemianHacks | bluelightning: also a random note, making sure toolchain versions match drives me insane. I've had builds fail by a mismatch a minor version before after downloading a several GB toolchain |
| 23:16.22 | BohemianHacks | I have satellite internet so it was... painful to say the least |
| 23:22.29 | *** join/#oe Aethenelle (~Aethenell@166.170.220.66) |
| 23:24.05 | lerela | bluelightning, there is currently a check to test if there is an existing key, it's easy to add a test to check if /etc/default/ssh is already in the image; if it is it means that someone explicitly added it, so the hook could just skip the variable overriding in that case |