**** BEGIN LOGGING AT Sat Oct 05 03:01:28 2019 Oct 05 05:47:31 stop why halium needs java? Oct 05 06:53:36 built Oct 05 06:53:42 let me copy it Oct 05 07:56:02 Target filesystem doesn't have requested /sbin/init. Oct 05 07:56:04 lol Oct 05 08:01:24 is it coming from LXC? Oct 05 08:09:13 no initrd Oct 05 08:09:34 cat: can't open '/root/var/lib/lxc/android/rootfs/fstab*': No such file or directory Oct 05 08:10:32 https://bpaste.net/show/_vON Oct 05 08:13:36 hmm it is mount problem Oct 05 08:13:40 empty Oct 05 08:23:56 @Luka That looks like something I had on Hammerhead Oct 05 08:27:41 i will look into it Oct 05 09:24:34 where must be system mounted? Oct 05 09:24:40 system.img Oct 05 09:24:57 and where rootfs.img must be mounted Oct 05 09:26:36 @Luka [and where rootfs.img must be mounted], seems to be in /root/ Oct 05 09:27:38 but sbin init is there Oct 05 09:27:44 must run as root Oct 05 09:27:50 as pid1* Oct 05 09:33:23 @Luka: I tried to apply as much as possible the patches from Halium 7.1 and Mer that they have for 8.1 builds but could be I missed something somewhere Oct 05 09:33:37 I'm not too familiar with all the low level boot sequence stuff Oct 05 09:34:09 It's likely something missing in android_system_core or something Oct 05 09:34:15 what init script must run all that? Oct 05 09:34:48 There normally is something in some of the android repos. I forgot which one now :s Oct 05 09:35:04 cp: can't create '/sbin/init': Text file busy Oct 05 09:35:18 i think it is trying to us /sbin/init Oct 05 09:35:35 instead of /root/sbin/init may be Oct 05 09:36:51 That could very well ve Oct 05 09:36:55 Be Oct 05 09:37:23 Like I said I just got it to build and applied patches where I could, but it will need work to get things running still Oct 05 09:37:43 Just all the legwork to get things build was quite significant already, that's why I tackled it Oct 05 09:38:37 no Oct 05 09:38:43 hell Oct 05 09:57:37 it is on its place Oct 05 10:17:19 https://github.com/Halium/android_system_core/blob/halium-7.1/init/init.cpp Oct 05 10:18:14 https://github.com/Herrie82/android_system_core/blob/halium-9.0/init/init.cpp Oct 05 10:18:32 cant find halium-8.1 branch Oct 05 10:24:20 sorry foun it is in -1 Oct 05 10:38:14 i think Oct 05 10:38:16 initrd: Normal boot … [ 2.919223,4] initrd: device is cedric Oct 05 10:38:18 is end Oct 05 10:38:31 due to cat: can't open '/android-system/boot/android-ramdisk.img': No such file or directory Oct 05 10:48:12 yes it is even not generated Oct 05 10:58:52 https://github.com/Halium/lxc-android/blob/master/var/lib/lxc/android/pre-start.sh#L25 Oct 05 10:59:26 @Herrie can it be a problem? Oct 05 11:17:40 @NotKit how must it build Oct 05 11:17:42 ? Oct 05 11:18:12 you need to fix scripts probably Oct 05 11:18:19 now system is mounted as root in Android Oct 05 11:18:25 so there won't be ramdisk.img to unpack Oct 05 11:18:58 so system.img should be likely mounted to /var/lib/lxc/rootfs Oct 05 11:19:33 @NotKit [so there won't be ramdisk.img to unpack], why? Oct 05 11:19:44 @NotKit [so system.img should be likely mounted to /var/lib/lxc/rootfs], may be Oct 05 11:19:50 because Android architecture changes for system as root Oct 05 11:19:56 [Edit] because Android architecture changed for system as root Oct 05 11:20:51 may be we can disable lxc for now? Oct 05 11:21:08 yes, but what you need is to get LXC running Oct 05 11:21:27 without LXC it will be pure rootfs, which doesn't depend on Halium at all Oct 05 11:22:26 yep, and no hwcomposer Oct 05 11:23:06 there will be no hwcomposer even with LXC :) though your device might still support HWC1 since it came with 7 Oct 05 11:23:52 hmmm how it creates image? Oct 05 11:24:26 so wait, you can not boot into rootfs? Oct 05 11:24:34 yep Oct 05 11:24:41 i am in initrd Oct 05 11:24:48 just remove system.img altogether and get rootfs to boot first Oct 05 11:24:54 you may need to fix halium-boot scripts Oct 05 11:25:21 initrd: Normal boot … [ 2.919223,4] initrd: device is cedric Oct 05 11:25:42 i think Normal boot is bad idea Oct 05 11:26:00 should it be abnormal or what? Oct 05 11:26:13 i think initrd: boot mode: halium Oct 05 11:26:18 is better idea Oct 05 11:26:48 https://github.com/Halium/initramfs-tools-halium/blob/halium/scripts/halium - check this script Oct 05 11:26:52 cat: can't open '/android-system/boot/android-ramdisk.img': No such file or directory … [ 2.898560,2] gzip: invalid magic … [ 2.899320,3] initrd: Normal boot Oct 05 11:27:27 elif [ -e $imagefile ] || [ -n "$_syspart" ]; then … # Regular image boot … tell_kmsg "Normal boot" Oct 05 11:27:56 @Luka [cat: can't open '/android-system/boot/android-ramdisk.img': No such file or dire …], ignore this part for the moment.initramfs tries to parse android-ramdisk.img to get fstab for Android mounts Oct 05 11:28:03 @Luka [cat: can't open '/android-system/boot/android-ramdisk.img': No such file or dire …], [Edit] ignore this part for the moment. initramfs tries to parse android-ramdisk.img to get fstab for Android mounts Oct 05 11:31:01 https://github.com/Halium/initramfs-tools-halium/blob/halium/scripts/halium#L358-L369 is not needed Oct 05 11:52:03 @NotKit so main problem is: Oct 05 11:52:05 run-init: opening console: No such device … <12>[ 3.088848,5] Target filesystem doesn't have requested /sbin/init. Oct 05 11:52:33 that's something with kernel, search chat logs here Oct 05 11:52:47 [Edit] that's something with kernel arguments, search chat logs here Oct 05 11:55:12 but may be it is problem Oct 05 11:56:35 that halium script ends ok Oct 05 11:56:36 Begin: Running /scripts/local-bottom ... done. Oct 05 11:57:12 it seems after No init found. Try passing init= bootarg. Oct 05 11:57:20 boot proces is ended Oct 05 11:58:19 @Flohack [[01-01 22:11:28.206] [0][1: init]No init found. Try passing init= bootarg.], like this Oct 05 11:58:27 sorry for pinging Oct 05 12:01:23 why dose it close connection? Oct 05 12:07:03 @NotKit is it rootfs or halium problem? Oct 05 12:07:49 @NotKit [that's something with kernel arguments, search chat logs here], ^ Oct 05 12:07:57 init can not open a console and it fails Oct 05 12:08:33 @Luka [why dose it close connection?], because of exec Oct 05 12:08:51 https://en.wikipedia.org/wiki/Exec_(system_call)#C_language_prototypes Oct 05 12:08:56 [Edit] https://en.wikipedia.org/wiki/Exec_(system_call) Oct 05 12:09:20 @NotKit [https://en.wikipedia.org/wiki/Exec_(system_call)], thanks Oct 05 12:09:45 @Luka [like this], i found that Oct 05 12:10:11 @NotKit or i must finde something else? Oct 05 12:10:31 ^ maybe try this Oct 05 12:10:32 Fwd from techfreak[m]: got it working by appending: Oct 05 12:10:32 Fwd from techfreak[m]: CONFIG_CMDLINE="console=tty0" CONFIG_CMDLINE_EXTEND=y Oct 05 12:10:32 Fwd from techfreak[m]: to my defconfig if somebody has the same issue. Oct 05 12:11:46 hmmmm sounds strange Oct 05 12:11:58 but i will Oct 05 12:12:10 show current /proc/cmdline Oct 05 12:14:12 @NotKit https://bpaste.net/show/Dr0V Oct 05 12:14:45 console=null is the issue Oct 05 12:17:12 @NotKit but BOARD_KERNEL_CMDLINE := console=ttyHSL0,115200,n8 androidboot.console=ttyHSL0 androidboot.hardware=qcom user_debug=30 msm_rtb.filter=0x237 Oct 05 12:17:28 may be delete consol from here at all Oct 05 12:19:05 @Luka [@NotKit but BOARD_KERNEL_CMDLINE := console=ttyHSL0,115200,n8 androidboot.conso …], it strangely not in /proc/cmdline at all Oct 05 12:31:28 Hannah Atkins was added by: Hannah Atkins Oct 05 13:34:37 Megan Schofield was added by: Megan Schofield Oct 05 13:49:56 Jacob Mann was added by: Jacob Mann Oct 05 13:53:21 Maybe it is the bootloader appending the kernel argument. Try to use a built in command line. Oct 05 13:55:12 @Daniel [Maybe it is the bootloader appending the kernel argument. Try to use a built in …], Which the bootloader will override still unless you patch the kernel Oct 05 13:55:27 To put internal cmdline after bootloaders Oct 05 14:45:02 uh, those 3 new joins from ~1hr ago, are all spam bots Oct 05 14:46:09 @dohbee [uh, those 3 new joins from ~1hr ago, are all spam bots], Gone Oct 05 15:08:41 harvey186 was added by: harvey186 Oct 05 15:09:02 i am in rootfs Oct 05 15:09:35 Hi, I have build everything for my natrium device. Now I stuck installing rootfs. I'm getting this error … adb: error: failed to copy 'halium-rootfs-20170630-151006.tar.gz' to '/recovery/': remote couldn't create file: Is a directory Oct 05 15:14:12 the standalone script showing this error: … ./halium-install-standalone.sh: line 56: syntax error near unexpected token `(' Oct 05 15:17:16 Seems to be some issue with the standalone script. You could try cloning the full git repository instead while i fix it. Oct 05 15:17:33 @JBBgameich [Seems to be some issue with the standalone script. You could try cloning the ful …], OK, will try Oct 05 15:21:07 @JBBgameich [Seems to be some issue with the standalone script. You could try cloning the ful …], OK, thx, it's running now Oct 05 15:30:12 how to crate udev rules? Oct 05 15:30:26 cat: '/var/lib/lxc/android/rootfs/ueventd*.rc': No such file or directory Oct 05 15:30:38 ther is no ueventd.rc Oct 05 15:30:44 only init.rc Oct 05 15:33:22 root@ubuntu-phablet:/var/lib/lxc/android/rootfs# /system/bin/logcat … CANNOT LINK EXECUTABLE "/system/bin/logcat": "/android/system/lib/liblog.so" is 32-bit instead of 64-bit … libc: CANNOT LINK EXECUTABLE "/system/bin/logcat": "/android/system/lib/liblog.so" is 32-bit instead of 64-bit … libc: Fatal signal 6 (SIGABRT), code -6 in tid 1320 ( Oct 05 15:33:22 logcat), pid 1320 (logcat) Oct 05 15:35:59 init: lxc-android-config main process (1039) terminated with status 1 … [ 1653.858799] init: lxc-android-config post-start process (1040) terminated with status 1 … [ 1678.850980] init: lxc-android-config main process (1137) terminated with status 1 … [ 1707.858781] init: lxc-android-config post-start process (1138) terminated with status 1 Oct 05 15:37:03 @NotKit thanks i got to rootfs, any idea what happens, and why it is trying to finde 64-bit in lib instead of 64? Oct 05 15:45:15 @Luka [root@ubuntu-phablet:/var/lib/lxc/android/rootfs# /system/bin/logcat … CANNOT LINK …], `sudo env LD_LIBRARY_PATH=/system/lib64:/vendor/lib64 /system/bin/logcat` Oct 05 15:50:24 @fredldotme [sudo env LD_LIBRARY_PATH=/system/lib64:/vendor/lib64 /system/bin/logcat], logcat read failure Oct 05 15:53:36 will it work without lxc? Oct 05 15:54:06 probably not, logcat will most likely require logcatd to be running which is part of the android image Oct 05 15:54:14 Is your server down ?? https://ci.ubports.com/job/xenial-rootfs-armhf/ Oct 05 15:55:29 lxc-android-config main process (784) terminated with status 1 Oct 05 15:56:22 Will lxc work without udev? Oct 05 15:56:40 it should, as ueventd is run inside the container Oct 05 15:57:27 @harvey186 [OK, will try], It should be fixed / workarounded now. Oct 05 15:59:00 @NotKit [it should, as ueventd is run inside the container], https://bpaste.net/show/V_BB Oct 05 16:00:45 faileds to start Oct 05 16:02:20 but @NotKit it is not all init: lxc-android-config post-start process (789) terminated with status 1 … in dmesg Oct 05 16:07:48 any ideas how to debug? Oct 05 16:10:06 first disable LXC container autostart I think Oct 05 16:10:53 then check LXC script if it makes sense for Android 8 and try to start container manually in foreground with verbose output Oct 05 16:11:20 @NotKit [first disable LXC container autostart I think], is it in halium script? Oct 05 16:11:42 no, in rootfs Oct 05 16:13:42 no... check upstart scripts Oct 05 16:16:26 any other scripts? Oct 05 16:16:53 I meant upstart service files Oct 05 16:17:00 there a service for LXC container Oct 05 16:56:27 @NotKit [there a service for LXC container], cant finde it Oct 05 16:56:38 phablet@ubuntu-phablet:/$ initctl list | grep lxc … lxc-android-boot stop/waiting … lxc-android-config stop/waiting … lxc stop/waiting … lxc-instance stop/waiting … lxc-net stop/waiting Oct 05 17:03:27 @Luka [cant finde it], i am really too stupide Oct 05 17:16:50 @NotKit like, upstart sends an event to the lxc-android-config job which uses lxc-start to initialize the preconfigured "android" lxc container config. ? Oct 05 17:17:09 isnt that info outdated? Oct 05 17:20:09 usr/lib/lxc-android-config/ Oct 05 17:22:52 @NotKit i found lxc log Oct 05 17:23:02 @Luka [@NotKit like, upstart sends an event to the lxc-android-config job which uses l …], I don't remember with upstart, but something does that it Oct 05 17:30:33 so what am i serching for? Oct 05 17:33:24 @NotKit log is spamed with https://bpaste.net/show/IlZb Oct 05 17:35:03 LXC fails because current startup script in config expects to extract ramdisk with init Oct 05 17:39:05 @NotKit [LXC fails because current startup script in config expects to extract ramdisk wi …], but ramdisk is extracted by halium script in initrd isnt it? Oct 05 17:40:13 ouch sorry Oct 05 17:40:26 no, /var/lib/lxc/android does this as well Oct 05 17:40:46 @NotKit [LXC fails because current startup script in config expects to extract ramdisk wi …], but where is init now, if there is no ramdisk? Oct 05 17:40:59 finaly understood Oct 05 17:41:15 Android does not, like other Linux, jump out of its initrd into a root filesystem on disk. Instead the initrd is actually the rootfs. The Android build that gets created from phablet.ubuntu.com takes care that the android initrd is shipped in /system/boot/android-ramdisk.img from where it gets copied to /boot at install time and on upgrades. Oct 05 17:41:25 is now outdated Oct 05 17:41:36 begins from android 8 Oct 05 17:41:41 what is the structure of your system.img? Does Android mount it to /? Oct 05 17:44:46 with Android 8, there are system as root and system as system devices Oct 05 17:45:13 so probably your builds ramdisk, but it was not copied by build scripts Oct 05 17:46:28 @NotKit [with Android 8, there are system as root and system as system devices], efidroid dev saied that it is right, begins from android 10 it is sytem as root Oct 05 17:46:43 @NotKit [so probably your builds ramdisk, but it was not copied by build scripts], but where it is? Oct 05 17:47:30 cant finde it Oct 05 17:48:37 not in out directory Oct 05 17:49:10 @NotKit [with Android 8, there are system as root and system as system devices], can i somehow check my device withot instaling android? Oct 05 17:49:33 just check system.img contents Oct 05 17:49:43 you probably have ramdisk.img in out Oct 05 17:52:35 only recovery-ramdisk Oct 05 17:53:50 sorry Oct 05 17:53:58 i have ramdisk.img Oct 05 17:54:11 where must i place it? Oct 05 17:55:20 @NotKit [you probably have ramdisk.img in out], can i copy it to /android-system/boot/android-ramdisk.img? Oct 05 17:56:00 there is no such dir Oct 05 17:56:06 when i am in rootfs Oct 05 17:58:04 @Luka [can i copy it to /android-system/boot/android-ramdisk.img?], maybe, it's up to you to try Oct 05 18:03:43 @NotKit [maybe, it's up to you to try], that dir is not avalible in rootfs Oct 05 18:03:52 where it mus be? Oct 05 18:04:02 in system.img root? Oct 05 18:26:32 check the path that /var/lib/lxc/android preinit script tries to access Oct 05 18:44:52 If anyone has the device tree or boot.img for the fairphone 3 it would be much appreciated, thanks. Oct 05 18:46:22 it has too new Android, so I doubt Halium has anything to do with that device yet Oct 05 18:56:37 @Flohack [Which the bootloader will override still unless you patch the kernel], There is kernel config to override the bootloader: CMDLINE_OVERRIDE Oct 05 19:01:16 @Daniel [There is kernel config to override the bootloader: CMDLINE_OVERRIDE], It does not work that way but you can try ^^ - because then it overrides the whole cmdline and your device will probably bootloop Oct 05 19:01:25 There is a ton of needed params in the bootloader cmdlien Oct 05 19:04:00 you need to append your custom cmdline to bootloader cmdline, and make selective overrides Oct 05 19:04:08 these are inside device tree if im not wrong Oct 05 19:04:13 i mean cmdline Oct 05 20:57:59 Does anyone knows where I can download a UbuntuTouch-fsroot image ? The link on halium website doesn't work Oct 05 21:04:13 have you read http://docs.ubports.com/en/latest/porting/introduction.html? Oct 05 21:04:49 it has a link to https://ci.ubports.com/job/xenial-rootfs-armhf/ Oct 05 22:04:58 @NotKit [have you read http://docs.ubports.com/en/latest/porting/introduction.html?], Yes, but i stuck on download link Oct 05 22:06:42 @NotKit [it has a link to https://ci.ubports.com/job/xenial-rootfs-armhf/], This link wasn't working some hours ago. Thx, now I can go further **** BEGIN LOGGING AT Sun Oct 06 01:47:31 2019 **** ENDING LOGGING AT Sun Oct 06 02:59:57 2019