**** BEGIN LOGGING AT Mon Jun 11 03:00:03 2018 Jun 11 03:48:48 hyperjuice was added by: hyperjuice Jun 11 05:39:36 @reivilibre, I think only adb says that Jun 11 07:39:03 @VeryOriginalUsername, Yes I think so too but I wasnt confident and that was the only reason why I could think adb access would be stopped Jun 11 09:16:54 Ramasai was added by: Ramasai Jun 11 09:18:09 Does anyone know how to port Ubuntu touch?? For kitkat Jun 11 09:19:26 @Ramasai, 5.1 or 7.1 Jun 11 09:19:34 Nothing more afaik Jun 11 09:19:46 No it's 4.4 Jun 11 09:19:56 Android Jun 11 09:20:03 Mtk chipset Jun 11 09:20:11 Fwd from Ramasai: No it's 4.4 Jun 11 09:20:12 Fwd from Ramasai: Mtk chipset Jun 11 09:20:19 No halium for you probably then Jun 11 09:20:36 Can u tell me what's halium.? Jun 11 09:21:02 Is it a platform for porting? Jun 11 09:21:03 Thing that you use to install ubuntu touch Jun 11 09:21:09 Oh Jun 11 09:26:03 @vanyasem I have pushed udev rules and manifest- … https://github.com/Halium/lxc-android/pull/14 … https://github.com/ubports/lxc-android-config/pull/10 … https://github.com/Halium/halium-devices/pull/92 Jun 11 09:35:51 also @vanyasem I am getting this error- … -bash: /dev/wcnss_wlan: Bad address … will I have to apply this patch- http://gerrit.aicp-rom.com/c/AICP/device_sony_huashan/+/32150/1/rootdir/init.board.wifi.rc ? Jun 11 09:36:12 [Edit] also @vanyasem I am getting this error- … -bash: /dev/wcnss_wlan: Bad address … will I have to apply this patch- http://gerrit.aicp-rom.com/c/AICP/device_sony_huashan/+/32150/1/rootdir/init.board.wifi.rc ? in my device tree Jun 11 09:39:08 [Edit] also @vanyasem I am getting this error- … -bash: /dev/wcnss_wlan: Bad address Jun 11 09:39:15 @rupansh you can try Jun 11 09:42:07 @z3ntu, ;_; I would prefer an alternative Jun 11 11:29:06 @rupansh, That's expected if you mean WiFi bringup Jun 11 12:51:39 @reivilibre, So what am I supposed to do? I was testing wifi Jun 11 12:54:29 @Ramasai which SoC? Jun 11 12:54:54 4.4 is possible in theory, but no easy way, especially unless you have full device Android source tree Jun 11 12:59:05 https://github.com/Halium/projectmanagement/pull/83 - could someone check that out? Jun 11 13:04:08 @rupansh, u here? Jun 11 13:12:42 @vanyasem, yeah Jun 11 13:12:46 Ivan does your arch image contain all the caf things? Jun 11 13:12:56 I'm looking for something that's around 1gig and has that Jun 11 13:13:16 @rupansh, plz pay attention to formatting in your PRs Jun 11 13:13:24 i had to manually fix it for you Jun 11 13:13:33 @vanyasem, sorry Jun 11 13:13:47 btw I think I fixed my pr ( https://github.com/Halium/docs/pull/98 ) Jun 11 13:14:04 hey hey easy i just woke up ;) Jun 11 13:14:11 I am used to python so I formatted it according to python by mistake xD Jun 11 13:14:16 Isn't it 16:14 in your time?)) Jun 11 13:14:28 @VeryOriginalUsername, correct Jun 11 13:14:32 @VeryOriginalUsername, He was woke whole night lolz Jun 11 13:16:23 @VeryOriginalUsername, i introduced a merge conflict yesterday to it Jun 11 13:16:26 and then resolved it Jun 11 13:16:32 Saw that merge, thanks Jun 11 13:16:34 @rupansh, merged all your PRs Jun 11 13:16:58 @vanyasem, Wew Jun 11 13:17:19 @rupansh, By the way is this normal? Jun 11 13:17:49 @rupansh, `nmtui` while in ss Jun 11 13:17:50 @rupansh, [Edit] `nmtui` while in ssh Jun 11 13:17:57 and try to connect to a network Jun 11 13:18:04 all right Jun 11 13:21:58 @vanyasem, Wifi isn't appearing :C Jun 11 13:22:08 only rmnet- Jun 11 13:22:14 [Edit] only rmnet0 Jun 11 13:23:27 NVM it showed up :D Jun 11 13:24:33 why have you checked "test_wifi" in your issue? Jun 11 13:25:00 it's a test that's not really supposed to work :P Jun 11 13:25:32 @vanyasem, unchecked :P . I got connected to wifi but i am getting the socket error thingy Jun 11 13:25:44 My kernel version is 3.18 though :C Jun 11 13:26:08 oh Jun 11 13:26:12 why sad face? Jun 11 13:26:23 i worked with it, it's pretty comfortable Jun 11 13:26:43 i see Jun 11 13:26:50 but i can't ping 8.8.8.8 Jun 11 13:27:14 `dmesg` ? Jun 11 13:28:41 @vanyasem, https://hastebin.com/iqecizilon.coffeescript\ Jun 11 13:31:48 Can I change hostname of my halium-rootfs system or will it break? Jun 11 13:44:45 @VeryOriginalUsername, sure you cab Jun 11 13:44:46 @VeryOriginalUsername, [Edit] sure you can Jun 11 13:54:21 should i do this? http://docs.halium.org/en/latest/porting/debug-build/wifi.html#kernel-3-10-ping-socket-permission-denied @vanyasem Jun 11 13:54:39 Its for 3.10 though :/ Jun 11 13:55:07 though the commits look like they might work nothing incompatible Jun 11 13:58:59 AdahBumbon was added by: AdahBumbon Jun 11 14:00:27 Hello Jun 11 14:01:20 @AdahBumbon, Hello 404th member :) Jun 11 14:02:47 @vanyasem, Hello ivan. Jun 11 14:03:09 I need help Jun 11 14:03:37 May someone can porting for zenfone 4 t00i Jun 11 14:11:26 @rupansh, This is often not neccesary, the echo sta thing should be enough (sorry if this already has been answered) Jun 11 14:12:44 @JBBgameich, Answered but np. I am getting socket error:permission denied right now. I am trying the fix given in the doc(Though I am on 3.18) Jun 11 14:14:11 I've done `readlink /vendor` and it points to `/system/vendor` yet files are missing Jun 11 14:14:21 Can I redo the link with `ln -s`? Jun 11 14:14:35 Is `/system/vendor` a link to `/vendor`? Jun 11 14:14:49 Yep, my device has no /vendor part Jun 11 14:14:58 I know Jun 11 14:15:08 oh ok Jun 11 14:15:21 @VeryOriginalUsername, Are you using hybris-boot? Jun 11 14:15:29 Yep Jun 11 14:15:44 And you have no vendor part, but you also have `/system/vendor > /vendor`? Jun 11 14:16:08 Correct Jun 11 14:16:20 Where are your vendor binaries? Jun 11 14:16:21 It has light blue color to it in `ls` Jun 11 14:16:37 `ls -l` will tell you the target Jun 11 14:17:11 Firwamre is in /system/etc/firmware (copied to /system/vendor/firmware but not everything appeared) and libs in /system/lib Jun 11 14:17:32 `root@nicki:/# ls -l /vendor … ` … `lrwxrwxrwx 1 root root 14 Feb 11 2016 /vendor -> /system/vendor … ` Jun 11 14:17:37 [Edit] `root@nicki:/# ls -l /vendor` … `lrwxrwxrwx 1 root root 14 Feb 11 2016 /vendor -> /system/vendor` Jun 11 14:17:58 Correct, so what is `/system/vendor` Jun 11 14:18:13 A folder in /system part Jun 11 14:18:29 Okay, I don't understand why you'd need to redo the link then? Jun 11 14:18:37 Not all files appear Jun 11 14:18:50 In fact, most dont when I `ls /vendor` Jun 11 14:18:55 @UniversalSuperBox, it should contain proprietary vendor stuff? Jun 11 14:19:07 let me scroll up so i don't confuse Jun 11 14:19:09 @rupansh, The question was "Is it a folder or a symlink" Jun 11 14:19:32 @VeryOriginalUsername, So `ls /vendor` is not the same as `ls /system/vendor`? Jun 11 14:20:46 I mounted mmc of the device in telnet state and copied /system/etc/firmware to /system/vendor/firmware cause it wasn't reading it before but it's gone Jun 11 14:20:50 :thinking: Jun 11 14:21:19 (Photo, 661x434) https://irc.ubports.com/FJzN70WS.png Jun 11 14:21:27 I don't think it should stay like that.. Jun 11 14:21:49 @VeryOriginalUsername, eh Jun 11 14:21:55 Do you have a firmware zip for your device? Jun 11 14:21:59 maybe check it Jun 11 14:22:48 because I don't have firmware folder in my system/vendor Jun 11 14:22:59 @rupansh, Don't think so Jun 11 14:23:47 @VeryOriginalUsername, Your device is Moto G 2014 right? Jun 11 14:23:55 Sony Xperia M /nicki/ Jun 11 14:24:28 all right Jun 11 14:38:51 @vanyasem I am still getting socket :permission denied error even after doing this :/ - http://docs.halium.org/en/latest/porting/debug-build/wifi.html#kernel-3-10-ping-socket-permission-denied here is the dmesg log - https://hastebin.com/uselejucuz.scala should I try pushing usermod and try the usermod fix? Jun 11 14:45:03 @rupansh, add Jun 11 14:45:51 @rupansh, add … ``` … inet:x:3003:root … net_raw:x:3004:root … ``` … to ```/etc/group``` … reboot Jun 11 14:46:57 @vince1171, all right give me a min Jun 11 14:53:49 Guys give me a Nobel prize for using brain Jun 11 14:54:09 Actual /system partition isn't used and I kept modyfing it instead of mounting system.img Jun 11 14:57:09 @VeryOriginalUsername, lol Jun 11 14:57:10 tf Jun 11 14:57:27 At least I hope so, now testing Jun 11 15:02:31 @UniversalSuperBox can I use CLANG built kernel? Jun 11 15:02:51 Anything is worth trying at least once Jun 11 15:03:04 I see Jun 11 15:03:16 I guess i will test it once i test everything else :P Jun 11 15:04:12 Can I resize system.img without having to re-image it? Jun 11 15:08:41 (Photo, 661x434) https://irc.ubports.com/fd7dfTEr.png Jun 11 15:08:46 test_lights is missing a ")" Jun 11 15:14:21 M1573r71 was added by: M1573r71 Jun 11 15:14:45 @VeryOriginalUsername if that would be the case, the compilation would have failed... Jun 11 15:15:06 In my case it gets stuck on this one Jun 11 15:15:17 And this is the last message in `strace` Jun 11 15:15:23 @VeryOriginalUsername, No Jun 11 15:15:38 That's not how any of this works Jun 11 15:15:43 😛 Jun 11 15:16:07 :shrug: Jun 11 15:17:16 Also, we have no members at moment in group (if you know what I mean) Jun 11 15:23:23 No-one got my joke 😥 Jun 11 15:30:14 @vince1171, worked thanks :D Jun 11 15:33:33 "library "libgui.so" wasn't loaded and RTLD_NOLOAD prevented it" Do i need to remove RTLD_NOLOAD from my defconfig? Jun 11 15:37:17 "library "libgui.so" wasn't loaded and RTLD_NOLOAD prevented it" is usually not an error and should be ignored Jun 11 15:38:47 @krnlyng, okay Jun 11 15:38:49 smart bot Jun 11 15:39:15 LOL Jun 11 15:39:30 it's an IRC bridge, actual people are talking through it Jun 11 15:39:31 That's not a bot on other side Jun 11 15:39:39 oh lolz Jun 11 15:39:49 @krnlyng, thanks Jun 11 15:40:19 how do I know that graphics test was successful? Jun 11 15:40:29 Look at your device's display Jun 11 15:40:38 @Lyud Xaker, ok sorry, I'm a bit confused. … so nope, I used the halium script becasue when I tried with jbb's one I got `Cannot find a valid super-block for the file system.` Jun 11 15:41:32 I didn't notice any change :C … also i got … -bash: test_egl_config: command not found Jun 11 15:42:09 oh also I am getting segmentation fault Jun 11 15:43:08 and now my e4.5 is stucked: white bq boot screen for a while then black screen and again boot one... Jun 11 15:44:07 @rupansh, If your device is based on CAF, please add caf repo, (http//repo.halium.org/caf) and then upgrade Jun 11 15:44:13 after installing the halium rootfs it said `All done! Now you can use 'adb shell' to get a shell to the recovery` what does that mean? could the recovery shell be useful? Jun 11 15:47:01 @mymike00, which is your device tree? Jun 11 15:47:07 @mymike00, [Edit] where is your device tree? Jun 11 15:47:23 what is it? Jun 11 15:47:26 @mymike00, you need to change the ssh password Jun 11 15:47:35 for that you need adb shell Jun 11 15:47:58 @rupansh, thanks Jun 11 15:48:12 and if ssh is not appearing you will have to check telnet Jun 11 15:49:29 @Lyud Xaker, I only have `hybris-boot`, `system.img` and the `halium-rootfs.tar.gz` Jun 11 15:50:12 @mymike00, he means the /device/"vendor"(i.e your phone's company)/device_name Jun 11 15:50:16 @Lyud Xaker, source of /device/[vendor]/[model] Jun 11 15:51:01 https://github.com/halium-krillin/android_device_bq_krillin/blob/halium-7.1/rootdir/fstab.krillin Jun 11 15:51:07 probably the fstab is wrong, the script is wrong or smth Jun 11 15:51:41 @rupansh, I don't have a build dir, @JBBgameich built it for me... Jun 11 15:52:33 (Sticker, 512x246) https://irc.ubports.com/Kn0xHJXM.webp Jun 11 15:56:27 @UniversalSuperBox apt-get update won't work Temporary faliure resolving "xyz" is this normal? Jun 11 15:56:35 What? Jun 11 15:56:41 Do you have an internet connection? Jun 11 15:57:06 @rupansh, that reminds me that i promised to take a look at this Jun 11 15:57:13 been a bit busy Jun 11 15:57:39 @UniversalSuperBox, yeah Jun 11 15:57:59 i can ping google.com to Jun 11 15:58:02 [Edit] i can ping google.com too Jun 11 15:58:09 @vanyasem, np but fixed Jun 11 15:58:10 :P Jun 11 16:00:44 @bhushanshah, in my case that didn't help either Jun 11 16:03:50 @rupansh, thats not *really* normal Jun 11 16:04:02 @Lyud Xaker, realized Jun 11 16:04:20 have you disabled the android paranoid network? Jun 11 16:05:33 @Lyud Xaker, Nope Jun 11 16:05:39 Switched back to old kernel Jun 11 16:05:59 here is the output - https://hastebin.com/ofuvosehej.pas Jun 11 16:06:37 hello Jun 11 16:06:59 i need some help Jun 11 16:07:17 Are you working on the Moto G5 Plus Jun 11 16:07:26 yes Jun 11 16:07:29 @UniversalSuperBox, 😂 Jun 11 16:07:40 @bhushanshah, HA. Jun 11 16:07:45 What's going on? Jun 11 16:07:53 CONFIG_IKCONFIG i need to y Jun 11 16:07:57 lol Jun 11 16:08:04 how i can? Jun 11 16:08:25 editing the defconfig for example? Jun 11 16:08:36 Did you not start from the preexisting source? Jun 11 16:09:16 @rupansh, https://github.com/Halium/android_kernel_lge_bullhead/commit/3f8345978921875227cd20c09d6deff05778c923 … https://github.com/Halium/android_kernel_lge_bullhead/commit/91506c596892de2160799cf69282a7488fdd24ca … https://github.com/Halium/android_kernel_lge_bullhead/commit/0b64b0cd08b1b79eb4a26aa40651d7ff0a4fff3c … and Jun 11 16:09:16 dont forget to disable the paranoid network on the kernel config Jun 11 16:09:55 i know u build it @UniversalSuperBox Jun 11 16:10:04 @Lyud Xaker, all right Jun 11 16:10:42 I probably forgot to disable paranoid network previously :P Jun 11 16:10:43 @potter, Right, and if you use 'halium-devices potter' it'll get all set up for you Jun 11 16:10:48 @ubptgbot "dont forget to disable the paranoid network on the kernel config" can u please tell me the file name Jun 11 16:10:59 That's for someone else Jun 11 16:11:19 and how can i set value y for CONFIG_IKCONFIG Jun 11 16:11:53 please help me Jun 11 16:11:53 @potter, navigate to kernel/motorola/msm8937/arch/arm64/configs Jun 11 16:12:01 @rupansh, It's armhf Jun 11 16:12:04 and check if the defconfig exists Jun 11 16:12:08 @UniversalSuperBox, eh Jun 11 16:12:16 SD 430 and armhf? Jun 11 16:12:23 Yes. Jun 11 16:12:31 lol nice Jun 11 16:12:32 motorola haven't moved to arm64 yet Jun 11 16:12:33 Okay, that config should be enabled on my source. Have you used halium-devices to get it? Jun 11 16:13:39 kernel/motorola/msm8937/arch/arm64/configs there are many files, which one Jun 11 16:14:12 @potter, Know your defconfig by checking in Board Jun 11 16:14:29 @potter, switch to kernel/motorola/msm8937/arch/armhf/configs … instead or you can just do what @theimpulson sur said Jun 11 16:15:14 Guys. This is not fixing the real problem Jun 11 16:15:47 @UniversalSuperBox, mer kernel checker? Jun 11 16:16:02 omg, i seriously have mental issues, my fstab had "/data" instead of "/userdata" Jun 11 16:16:12 that explains why i couldnt run anything... Jun 11 16:16:18 Not really Jun 11 16:16:25 in defconfig it already seted to y, but when i am checking with mer-kernel, it saying the value is unset Jun 11 16:16:26 It doesn't explain that at all Jun 11 16:16:48 potter, did you run `./halium/devices/setup potter` Jun 11 16:17:09 @Lyud Xaker, Your fstab is supposed to have `/data`. What's not working? Jun 11 16:17:50 yes Jun 11 16:18:07 i was talking about /dev/block/bootdevice/by-name/ Jun 11 16:18:56 @UniversalSuperBox yes Jun 11 16:19:02 I managed to break my halium install Jun 11 16:19:21 I ask because that is set correctly in the kernel config. https://github.com/UniversalSuperBox/android_kernel_motorola_msm8953/blob/85e7131a3acb53c9a9e8ffb2b8bffc719ccd388b/arch/arm/configs/potter_defconfig#L115 Jun 11 16:19:46 I tried to change config_default_mmap_min_addr and now no hybris-boot will work :p Jun 11 16:20:04 I guess i'll give it a reflash, but done for now Jun 11 16:20:08 [Edit] I guess i'll give it a reflash, but I'm done for now Jun 11 16:21:38 @UniversalSuperBox It's same as Your given link Jun 11 16:21:54 Okay, so why are you trying to set it again? Jun 11 16:22:50 @UniversalSuperBox actually mer-kernel check results say's that it is unset Jun 11 16:22:58 It's fine Jun 11 16:23:25 @UniversalSuperBox ok thanks bro Jun 11 16:23:50 @rupansh, There are two commands given to enable WiFi on the Halium docs. Run them both, even if the first one looks like it didn't work. Jun 11 16:24:48 @reivilibre, eh Jun 11 16:24:49 fixed Jun 11 16:24:51 thanks Jun 11 16:25:05 @rupansh, Worked? Jun 11 16:25:43 @reivilibre, yep Jun 11 16:25:50 had to change a file Jun 11 16:26:00 What file did you change? Jun 11 16:27:54 @Lyud Xaker, eh it seems like it isn't enabled in defconf Jun 11 16:28:37 @rupansh, forgot to mention: in my case i only had to write "echo sta > /sys/module/wlan/parameters/fwpath" Jun 11 16:29:10 @Lyud Xaker, um Jun 11 16:29:13 Wifi is enabled Jun 11 16:29:13 ;_; Jun 11 16:29:27 I am unable to sudo apt-get update Jun 11 16:29:31 while pinging google works fine Jun 11 16:29:52 but i had this problem too and i think i fixed it only writing that Jun 11 16:29:53 @rupansh, I think this is because apt runs as the user `_apt` and you have to do something involving Paranoid Networikin Jun 11 16:29:53 and I pinged update servers as well they are pinging fine Jun 11 16:30:13 @reivilibre, paranoid network is disabled in defconf tho Jun 11 16:30:19 @Lyud Xaker, all right lemme try Jun 11 16:30:21 did u use the nmtu UI? Jun 11 16:30:28 I think I heard Dalton say something like 'If your kernel version is > (x), you actually have to ENABLE Paranoid Networking' but I don't remember where, hang on Jun 11 16:30:30 to connect, type password and everything? Jun 11 16:30:46 @Lyud Xaker, ye Jun 11 16:31:00 It was actually Bhushan :P Jun 11 16:31:00 Fwd from bhushanshah: Is kernel version 3.10+? Jun 11 16:31:00 Fwd from bhushanshah: In that case you actually need to keep paranoid networking enabled Jun 11 16:31:07 @reivilibre, mine's 3.18 tho Jun 11 16:31:19 @reivilibre, eh Jun 11 16:31:20 That's a bit naff, it doesn't offer you a link back to the conversation :-( Jun 11 16:31:29 it never was enabled in my kernel though Jun 11 16:31:35 ^ this maybe? Jun 11 16:31:35 Fwd from bhushanshah: In that case you actually need to keep paranoid networking enabled Jun 11 16:31:47 all right Jun 11 16:31:49 Sorry, I don't know how to share that conversation Jun 11 16:32:14 i guess i will reset back to default status and enable it Jun 11 16:32:22 (Photo, 480x427) https://irc.ubports.com/2Ix2fZjK.png search for it, if you want. Jun 11 16:33:12 (Photo, 480x427) https://irc.ubports.com/Rx0FO1aC.png search for it, if you want. *(in PlaMo channel) Jun 11 16:33:54 @reivilibre, interesting but all i need to do is probably enter ANDROID_PARANOID_NETWORK=y in defconf Jun 11 16:34:02 personally I had the same issue where only root could access the Internet and I haven't tested the above yet Jun 11 16:34:07 @rupansh, Yes, give it a try Jun 11 16:35:43 It does seem counter-intuitive for sure, but if it won't take you too long to try, then it's worth a try Jun 11 16:36:03 @reivilibre, started compiling Jun 11 16:38:01 I think a good strategy for porting Halium whilst you're still trying kernel configs is to have something to do whilst you wait for kernel compiles.. :P Jun 11 16:39:00 or maybe get it right first time, but I'm too much of a fool for that Jun 11 16:39:47 @reivilibre, xD Jun 11 16:43:53 Sed Jun 11 16:44:03 It didn't do the trick Jun 11 16:44:33 v1kky was added by: v1kky Jun 11 16:44:36 aw :C Jun 11 16:45:21 In that case, I haven't a clue currently but I would be interested in the answer Jun 11 16:45:32 actually i am also getting "sudo: unable to resolve host localhost.localdomain" Jun 11 16:45:52 @rupansh, That's fine, I get that on a working Plasma Mobile on HTC10 without any visible networking issues Jun 11 16:46:16 i see Jun 11 16:46:39 Not sure what the cause of that is, probably needs a hostname to be set? But I don't think it's your issue with apt Jun 11 16:46:51 my /etc/hosts file is empty though Jun 11 16:46:56 could that be the cause Jun 11 16:47:02 That smells like it could be the cause, yes Jun 11 16:48:22 normally you'd have `127.0.0.1 localhost` or so, right? Jun 11 16:48:48 [Edit] That smells like it could be the cause, yes (of the sudo warning) Jun 11 16:49:36 @reivilibre, added it Jun 11 16:49:39 didn't fix Jun 11 16:50:04 What's the output of `hostname` (or `cat /etc/hostname` if that command doesn't exist)? Jun 11 16:50:25 localhost.localdomain Jun 11 16:50:35 also i added this in /etc/hosts … 127.0.0.1 localhost.localdomain localhost Jun 11 16:50:47 on my PC I have a entry `127.0.1.1 bbm-neon` (where bbm-neon is my hostname), I assume it's something siimlar Jun 11 16:51:14 *confused* Jun 11 16:51:24 @rupansh, Did that sort it? Jun 11 16:51:45 @reivilibre, nope Jun 11 16:52:00 though unable to resolve thingy was fixed Jun 11 16:52:14 yeah the unable to resolve thingy wasn't really an issue (more of a warning) Jun 11 16:53:08 If you want to have a greater appreciation of the issue (you'd probably rather fix it though), if you `su -s /bin/sh _apt` to switch into `_apt` user, you won't be able to ping anything in that user? Jun 11 16:54:22 actually `ping` might work because it has setuid so it gets to run as root anyway. Jun 11 16:54:40 but any other kind of network access will fail (at least in my experience with this issue) Jun 11 16:55:08 hmm Jun 11 16:55:16 i will try to ssh connect to my server maybe Jun 11 16:55:51 nice lolz ssh not found Jun 11 16:56:16 if you try and `ping halium.org` it will probably fail with a DNS resolution error Jun 11 16:56:23 @UniversalSuperBox hey r u there Jun 11 16:56:24 [Edit] if you try and `ping halium.org` it will probably fail with a DNS resolution error maybe Jun 11 16:57:01 I don't know. I'm sure I tested that networking didn't generally work for `_apt` but I don't remember how, and logic would suggest that `ping` is fine Jun 11 16:57:13 @reivilibre, Nice you predicted Jun 11 16:57:17 now i know what to fix Jun 11 16:57:31 What did you test? Jun 11 16:58:02 pinged 8.8.8.8 and got socket:permission denied Jun 11 16:58:10 and i can just apply _apt fix Jun 11 16:58:18 the way i fixed access for root Jun 11 16:58:46 Ah right, you added it to the group for android networking? Jun 11 16:59:33 @reivilibre, yeah Jun 11 16:59:39 i will add _apt Jun 11 16:59:46 though its a workaround Jun 11 16:59:52 Ok! Let me know if it works. It's a start at least Jun 11 17:02:50 On this HTC10 PlaMo installation, the `phablet` user is added to `android_net`, `android_net2`, `android_net3` groups. So maybe that's the intended way to do it. Jun 11 17:03:22 @reivilibre, didn't work ;__; Jun 11 17:03:31 @reivilibre, can u send the groups on hastebin Jun 11 17:04:09 can please someone help me with the bq e4.5 in bootloop? Jun 11 17:04:19 @rupansh, I can't guarantee it'll be any help, but I will do for you. Jun 11 17:04:38 @reivilibre, thanks :D Jun 11 17:07:41 @rupansh, https://hastebin.com/raw/uwaxureqam bear in mind it's for Plasma Mobile Jun 11 17:08:02 yeah Jun 11 17:08:14 I don't see them adding `root` or `_apt` to the networking group to Jun 11 17:08:16 [Edit] I don't see them adding `root` or `_apt` to the networking group though Jun 11 17:08:35 Linux `3.18.31-04300-g6eae5f2` Jun 11 17:09:00 @reivilibre, root= phablet in halium Jun 11 17:09:26 @rupansh, I don't think so, `phablet` is the unprivileged user that runs the GUI apps Jun 11 17:09:35 @reivilibre, hmm Jun 11 17:09:51 there is no unprivileged user in halium if i am correct Jun 11 17:10:48 Possibly not, as it's intended really only for running some tests Jun 11 17:10:53 See bullhead kernel on halium git Jun 11 17:11:18 There are 4 commits in it related to security/paranoid networking Jun 11 17:11:31 Please adapt them to your kernel Jun 11 17:11:38 @bhushanshah, Different to the ones linked in the Halium docs? Jun 11 17:11:41 @bhushanshah, its 3.18 do i need to do it? Jun 11 17:11:57 @bhushanshah, [Edit] mine's 3.18 do i need to do them? Jun 11 17:12:02 @rupansh, In anything 3.10+ you need it Jun 11 17:12:17 @bhushanshah, all right Jun 11 17:12:18 @reivilibre, What is linked from halium docs? Jun 11 17:12:18 https://docs.halium.org/en/latest/porting/debug-build/wifi.html#kernel-3-10-ping-socket-permission-denied Jun 11 17:12:39 @reivilibre, Yes those Jun 11 17:12:55 @bhushanshah, All right Jun 11 17:13:00 There are just 3 commits there, not 4 Jun 11 17:13:00 time to cherry-pick them again Jun 11 17:13:03 (Sticker, 512x461) https://irc.ubports.com/VG8Z8tUt.webp Jun 11 17:13:34 @reivilibre, Right, well iirc one is config change/sorting options Jun 11 17:13:35 @rupansh, isn't there git apply-patch or something? I'm sure I used somethingl ike that Jun 11 17:13:38 Sorry for confusion Jun 11 17:14:09 @bhushanshah, No that's fine, I was just wondering whether that's the reason I didn't get it working for myself. (You need to ENABLE Paranoid networking, as I heard you say in PlaMo room, right?) Jun 11 17:14:14 headdesk … why, ubp-5.1 on hammerhead. Why do you try unsuccessfully to load selinux Jun 11 17:14:44 @UniversalSuperBox, Wrong channel *hides* Jun 11 17:14:56 Share in my headdesk, Bhushan Jun 11 17:15:02 @reivilibre, i prefer cherry-pick Jun 11 17:15:19 @rupansh, you posted a crying face, I thought that meant you wouldn't like it :P Jun 11 17:15:48 @reivilibre, feeling a bit exhausted :P Jun 11 17:15:49 append `.patch` to a GitHub URL for a commit and you can DL it Jun 11 17:17:21 @reivilibre, Already started compiling :P Jun 11 17:24:30 @bhushanshah, still stuck on temporary faliure resolving "xyz repo" Jun 11 17:24:34 (Sticker, 512x461) https://irc.ubports.com/hHpf97DK.webp Jun 11 17:25:12 What is exact error? Jun 11 17:26:11 @bhushanshah, https://hastebin.com/uwarayilek.pas Jun 11 17:26:49 Can you ping www.google.com? Jun 11 17:27:01 @bhushanshah, yep Jun 11 17:30:43 Can you pastebin your current config? (Ideally with running zcat /proc/config.gz from ssh) Jun 11 17:31:01 @bhushanshah, sure Jun 11 17:32:00 @bhushanshah, https://hastebin.com/ubahesegob.vbs Jun 11 17:33:11 Okay Jun 11 17:33:13 Hm Jun 11 17:33:27 I'll try to help you debug this tomorrow Jun 11 17:33:32 Today going off Jun 11 17:58:02 hey anybody there Jun 11 17:58:58 i need some help Jun 11 17:59:39 Don't ask to ask, just ask and anyone who is here may help Jun 11 18:03:19 @bhushanshah, kk Jun 11 18:03:43 @UniversalSuperBox, Is there a guide for porting Plasma Mobile? Jun 11 18:04:26 @rupansh, Have you ported Halium already? Jun 11 18:04:45 @ilyaishere, not stable Jun 11 18:05:01 Your port? Jun 11 18:05:09 Yes Jun 11 18:05:27 Do you have test_hwcomposer working? Jun 11 18:05:48 @ilyaishere, I haven't tested it yet Jun 11 18:06:11 hey i need some help when i am building hybris-boot my pc getting restart Jun 11 18:06:13 Well, then you'll have to finish porting Halium before you can run Plasma Mobile Jun 11 18:06:33 can you tell me how to test it? Jun 11 18:06:43 Have you read the docs? Jun 11 18:07:19 @ilyaishere please help me Jun 11 18:07:46 @ilyaishere, Yes but I don't see a "testing hwcomposer" or a similar doc in debugging seciton Jun 11 18:08:53 @rupansh, There sure is, it's one of the main tests Jun 11 18:09:25 http://docs.halium.org/en/latest/porting/debug-build/graphics.html Jun 11 18:09:34 @ilyaishere, Oh! you meant this Jun 11 18:09:45 Ok, I've done my job at pointing you to the docs, so you can move on ☺ Jun 11 18:10:13 well I am trying to test it but I am unable to do so due to apt-get update not working Jun 11 18:10:25 @ilyaishere, 🙈 Jun 11 18:10:25 @potter, Well, I don't think your rebooting PC is related to Halium, sorry... Jun 11 18:10:34 when i am building hybris-boot my pc getting restart Jun 11 18:10:41 I need to get the caf specific stuff Jun 11 18:10:53 @rupansh, It shouldn't really be working Jun 11 18:11:17 Why would you need to use it? Jun 11 18:11:25 My device is CAF Jun 11 18:11:36 @rupansh, Just install the plasma rootfs Jun 11 18:11:40 Caf plasma rootfs Jun 11 18:11:41 Well, there are many CAF devices Jun 11 18:11:52 @ilyaishere it's going on every time when try to building on anoter laptop Jun 11 18:11:57 potter, it sounds like your computer is overheating Jun 11 18:12:06 Maybe you need a better cooling solution Jun 11 18:12:32 Or CPU governor :) Jun 11 18:12:43 @UniversalSuperBox ok i will try it Jun 11 18:12:47 @rupansh, Don't mess with the caf repo Jun 11 18:12:50 Go for plasma caf Jun 11 18:12:59 Also, is your device 64 bit? Jun 11 18:13:07 @vanyasem, Yeah Jun 11 18:13:15 Have you applied libcompat patch? Jun 11 18:13:26 To the kernel Jun 11 18:13:35 um Jun 11 18:13:37 let me check Jun 11 18:13:59 nope Jun 11 18:14:03 Do it Jun 11 18:14:31 @vanyasem, can you link it? Jun 11 18:15:35 Fwd from vanyasem: https://github.com/Halium/android_kernel_oneplus_msm8996/pull/1 Jun 11 18:17:05 @vanyasem, all right so I will pick the change and build hybris-boot, download pm rootfs and use halium-install with pm argument? Jun 11 18:17:24 @rupansh, Yep. Be sure to download CAF rootfs Jun 11 18:17:39 all right thanks :D Jun 11 19:30:22 @vanyasem It isn't booting into gui. Here is the dmesg log- https://hastebin.com/xahiqaketu.log Jun 11 19:30:32 @rupansh, have you made the udev riles Jun 11 19:30:34 @rupansh, [Edit] have you made the udev rules Jun 11 19:30:49 @vanyasem, Nope ;_; Jun 11 19:31:03 well you must have them for GUI to work Jun 11 19:31:07 Will I have to make them the same way i did with reference rootfs? Jun 11 19:31:18 exactly the same way Jun 11 19:31:34 all right :D Jun 11 19:33:24 Yay Jun 11 19:33:28 I booted into GUI Jun 11 19:33:35 :D Jun 11 19:35:37 Its quite buggy though Jun 11 19:35:53 apps aren't working Jun 11 19:40:42 @vanyasem, so I can play chess , use filemanager,use konsole,play mahajong. Chromium works for 10-20 seconds . Everything else isn't working :/ Jun 11 19:40:48 OKAY Jun 11 19:40:54 Video + Photos Jun 11 19:41:00 to @plasmamobile Jun 11 19:41:12 all righty Jun 11 19:59:50 @vanyasem what was the country that the stickers were sent from to you? Jun 11 20:00:33 FUCK again missed that with OT chat Jun 11 22:13:07 Hey everyone Jun 11 22:13:22 I finally got Halium boot on Pixel XL (1st gen) Jun 11 22:13:35 Now stuck at udev & lxc-android configuration Jun 11 22:14:01 (Photo, 1280x295) https://irc.ubports.com/FMFEqe0K.png lxc@android looks for some init.rc but cannot find them Jun 11 22:14:26 I suppose these are in out/target/product/[codename]/root Jun 11 22:14:44 So, just copy all of init*.rc to the device, right? Jun 11 22:15:15 no Jun 11 22:15:26 probably your system.img wasn't mounted Jun 11 22:15:43 `ls /var/lib/lxc/android/rootfs` Jun 11 22:16:07 `dev socket` Jun 11 22:16:09 @vanyasem, either it's missing, or corrupt, or there is some other issue with it Jun 11 22:16:15 @Sergobot, yeah, system.img wasn't mounted Jun 11 22:16:23 that's why it can't find these files - they aren't there Jun 11 22:16:31 `file /data/system.img` Jun 11 22:17:29 interesting, it's there and it's in valid format Jun 11 22:17:40 `mount` output Jun 11 22:18:20 (Photo, 1280x642) https://irc.ubports.com/DoMJKhG3.png Jun 11 22:18:45 prefer copying output to pastebin (paste.ubuntu.com for example), it's easier to read that way Jun 11 22:18:54 sure Jun 11 22:19:29 could you try to manually mount /data/system.img somewhere ? Jun 11 22:19:33 and see if it fails Jun 11 22:19:54 also, have you disabled any of the services before? (system.mount or android-mount.service) Jun 11 22:19:55 yep Jun 11 22:20:01 not yet Jun 11 22:20:01 @Sergobot, yep as in "it fails"? Jun 11 22:20:08 @Sergobot, ?? Jun 11 22:20:18 will try in a couple minutes :) Jun 11 22:20:25 oh okay Jun 11 22:20:59 @vanyasem, no Jun 11 22:21:08 good, don't do it Jun 11 22:22:13 Hmm Jun 11 22:22:38 rebooted into hybris-boot and now system.img appears in `mount` Jun 11 22:22:52 what have you booted before? :) Jun 11 22:22:59 `halium-boot` Jun 11 22:23:14 well as funny as it sounds Jun 11 22:23:23 halium-boot doesn't support halium or plasma mobile Jun 11 22:23:33 it's meant to be used with Ubuntu Touch for now (UBports) Jun 11 22:23:50 you should use hybris-boot for halium reference and plasma mobile rootfs Jun 11 22:23:57 okay Jun 11 22:24:36 there's still nothing but `dev socket` in /var/lib/lxc/android/rootfs/ Jun 11 22:24:42 [Edit] there's still nothing but `dev socket` in `/var/lib/lxc/android/rootfs/` Jun 11 22:25:13 well, let's investigate it Jun 11 22:25:20 are you sure that your system.img isn't empty? Jun 11 22:25:24 yes Jun 11 22:25:38 `systemctl --failed` Jun 11 22:26:21 https://paste.kde.org/pbb45hzen Jun 11 22:26:39 okay, it reports mounting as successful Jun 11 22:27:13 may i ask you, which script have you used to install the rootfs? Jun 11 22:27:20 `halium-install` Jun 11 22:27:24 JBB's one Jun 11 22:27:40 okay, so that's not an issue Jun 11 22:28:06 `dmesg` could give us some clues Jun 11 22:29:46 https://paste.kde.org/pctlyt3pr Jun 11 22:30:48 lines 1622, 1633, 1634 Jun 11 22:30:53 it succeeds <.< Jun 11 22:31:36 reboot to twrp and try to mount it from there, see if you can see the files inside the image Jun 11 22:31:44 something is really out of it's way here Jun 11 22:32:54 `mount: losetup failed 1` Jun 11 22:33:11 hm, that could be an issue in your twrp build Jun 11 22:33:24 does mounting rootfs.img also fail with the same error message? Jun 11 22:33:41 yeah Jun 11 22:33:52 okay, so mount is broken in your twrp build Jun 11 22:34:06 make a note of that on the docs section about your device once you're done Jun 11 22:34:26 sure Jun 11 22:35:12 though i think it worked some time ago... Jun 11 22:35:25 any way Jun 11 22:35:37 any new ideas? Jun 11 22:35:54 well, you could try to remount system.img once booted into the rootfs Jun 11 22:36:15 to see if it has any issues (i.e. the image appears empty or mounting fails for some reason) Jun 11 22:37:12 i mean it mounts correctly, why would it still be empty <.< Jun 11 22:38:28 no, it mounts fine Jun 11 22:38:38 and has files inside? Jun 11 22:38:40 and there are all the files in `/system` Jun 11 22:38:49 no, system is a partition Jun 11 22:38:53 system is a partition Jun 11 22:39:02 it's a totally different thing Jun 11 22:39:07 it's where your android lies Jun 11 22:39:47 does it also mount into `/var/lib/lxc/android/rootfs/system`? Jun 11 22:39:53 it gets mounted in 2 places Jun 11 22:40:02 https://phabricator.mynameisivan.ru/P17$66 Jun 11 22:40:07 take a look at for a live example Jun 11 22:40:14 [Edit] take a look at this paste for a live example Jun 11 22:40:37 [Edit] does it also get mounted into `/var/lib/lxc/android/rootfs/system`? Jun 11 22:40:43 [Edit] it should get mounted in 2 places Jun 11 22:40:48 no Jun 11 22:41:02 okay, so that's the issue Jun 11 22:41:22 please don't mount it there yourself. and never restart lxc manually Jun 11 22:41:35 lxc@android should never be started/stopped/restarted manually Jun 11 22:41:48 just informing you in case you wanted to do that Jun 11 22:42:09 I've been in this chat for a long time, seen that many times :) Jun 11 22:42:14 okay, just making sure Jun 11 22:42:58 how does system.img get mounted? Jun 11 22:43:14 via a service Jun 11 22:43:20 `system.mount` Jun 11 22:43:28 `systemctl status system.mount` Jun 11 22:43:43 @UniversalSuperBox, right, in `local-fs.target.wants` Jun 11 22:44:20 @UniversalSuperBox, it's fine Jun 11 22:44:26 https://paste.kde.org/p0k2cb0zd Jun 11 22:44:30 @UniversalSuperBox, it doesn't get reported as failed, and dmesg also shows the mount as successful Jun 11 22:44:34 that's the weird part Jun 11 22:44:58 Is lxc@android not trying to start? Jun 11 22:45:18 @UniversalSuperBox, it is, and it's failing as system.img was not mounted in the correct place Jun 11 22:45:26 so it doesn't find the init.rc files Jun 11 22:46:02 The service mounts only in /system … https://paste.kde.org/p4yecstlq Jun 11 22:46:03 but! it gets mounted in /system and according to Sergey it's doing fine Jun 11 22:46:23 I guess I should put another mount entry there Jun 11 22:46:47 well no Jun 11 22:46:54 oh rly Jun 11 22:46:55 why? Jun 11 22:47:26 it's not the way it's supposed to work Jun 11 22:47:42 ah, ok then Jun 11 22:47:59 it's not broken in halium, the service is fine Jun 11 22:48:19 there's something wrong with your build, and i'm trying to come up with ideas Jun 11 22:49:49 @vanyasem, i assume that something fails in https://github.com/Halium/lxc-android/blob/master/var/lib/lxc/android/pre-start.sh Jun 11 22:50:13 that's the script that bind mounts system.img Jun 11 22:50:30 @vanyasem, [Edit] i assume that something fails in https://github.com/Halium/lxc-android/blob/master/var/lib/lxc/android/pre-start.sh#L15 Jun 11 22:53:39 @Sergobot, and well it gets executed Jun 11 22:53:43 because socket and dev are properly created Jun 11 22:53:48 [Edit] because `socket` and `dev` are properly created Jun 11 22:54:22 try to run `mount -n -o bind,recurse /system `/var/lib/lxc/android/rootfs/system`` Jun 11 22:54:36 [Edit] try to run`mount -n -o bind,recurse /system /var/lib/lxc/android/rootfs/system` Jun 11 22:54:46 [Edit] try to run`mount -n -o bind,recurse /system /var/lib/lxc/android/rootfs/system` Jun 11 22:54:49 [Edit] try to run `mount -n -o bind,recurse /system /var/lib/lxc/android/rootfs/system` Jun 11 22:54:56 and see if it fails Jun 11 22:55:09 worked fine Jun 11 22:55:26 it got mounted into `/var/lib/lxc/android/rootfs/system`, right? Jun 11 22:55:38 exactly Jun 11 22:55:53 ooookay Jun 11 22:58:07 i was wrong, that's not the script that mounts system.img, but a script that bind mounts other android partitions there Jun 11 22:58:32 out of ideas for now Jun 11 22:58:59 which script mounts it then? Jun 11 22:59:52 or maybe that's the script, it's 2 AM in Moscow, it's hard for me to think Jun 11 23:00:48 GMT +3 ayyy Jun 11 23:04:06 where does $LXC_ROOTFS_PATH come from? Jun 11 23:04:16 @vanyasem, in this script Jun 11 23:04:32 [Edit] where does `$LXC_ROOTFS_PATH` come from? Jun 11 23:04:33 i assume that line 29 is the issue Jun 11 23:04:53 do you have `/system/boot/android-ramdisk.img`? Jun 11 23:05:32 makes sense Jun 11 23:05:40 no Jun 11 23:05:52 okay, one step forward Jun 11 23:05:58 there's no `/system/boot/` at all Jun 11 23:06:52 Wait Jun 11 23:06:53 Hold up Jun 11 23:06:55 Pixel XL? Jun 11 23:07:23 Yep Jun 11 23:07:33 A/B partitioning Jun 11 23:07:38 oooooh Jun 11 23:07:40 There is no ramdisk in `/system/boot` Jun 11 23:07:53 @vanyasem here are the files- … system.img- https://drive.google.com/file/d/1P5cyxBr7tz2cvXffzcX1XLZIoYVblKcw/ … hybris-boot.img … https://drive.google.com/file/d/1R_0xBK5j4aDCe3eI53iUutHVthSRXyxj Jun 11 23:09:10 What was the complete contents of `/system`, @Sergobot ? Jun 11 23:10:13 https://paste.kde.org/p7w2tzhdl Jun 11 23:10:20 I believe (I have not had the pleasure of toying with an A/B device) that the system partition is root Jun 11 23:10:29 yyyyyyyyyep Jun 11 23:10:40 That's system.img Jun 11 23:10:47 mounted in `/system` Jun 11 23:10:53 Right Jun 11 23:11:08 Which is the system partition... at least for Android. Jun 11 23:11:22 And in your case it's `/` for Android Jun 11 23:12:01 is `android-ramdisk.img` in `out/target/product/${CODENAME}` Jun 11 23:12:07 in your build folder? Jun 11 23:12:39 `system/boot/android-ramdisk.img` Jun 11 23:12:40 it's there, yes Jun 11 23:12:54 i guess we could make use of that :P Jun 11 23:13:20 i've never had a device with A/B partitions, but it would be fun to try Jun 11 23:13:41 remount system.img as rw on your device and try to shove that ramdisk in there :P Jun 11 23:13:58 I have an idea of how to implement this in halium-boot Jun 11 23:14:21 @UniversalSuperBox, do you maybe have an idea of how to implement configfs as well? Jun 11 23:14:24 * vanyasem runs Jun 11 23:14:28 Nah, that's a you problem Jun 11 23:14:38 that was rude. Jun 11 23:14:44 So was that. :P Jun 11 23:14:53 fair 'nuff Jun 11 23:15:22 @vanyasem, @Sergobot really interested in how that goes Jun 11 23:15:44 10 mins Jun 11 23:15:48 or less Jun 11 23:15:52 No, that won't do Jun 11 23:16:02 @UniversalSuperBox, you're dull :P Jun 11 23:16:09 i was geniunely interested Jun 11 23:16:23 Because then you'll have `/system/system` Jun 11 23:21:13 okay so then @Sergobot you could try to modify you current ramdisk's init files so that they won't have an issue that Dalton described above Jun 11 23:21:32 or you could wait for someone to implement A/B partitioning support in Halium Jun 11 23:21:51 nobody has tried that yet, so well.. you could be the one Jun 11 23:22:14 heh Jun 11 23:22:27 i'll try to run it at all Jun 11 23:22:49 @vanyasem, there's no space left in system.img to copy that file Jun 11 23:23:08 @Sergobot, well you could expand the image Jun 11 23:23:15 [Edit] I'm better to get it to run at all Jun 11 23:23:26 anyways dalton said that won't work, i trust him Jun 11 23:23:54 ok then Jun 11 23:24:02 @vanyasem, you probably need to open that ramdisk and correct init files inside it Jun 11 23:24:08 @vanyasem, let me find the command Jun 11 23:25:10 `resize2fs -f /data/system.img 4G` Jun 11 23:25:19 replace 4G with the appropriate size for your image Jun 11 23:25:48 200M :P Jun 11 23:26:08 oh okay, it's 480 mb for me Jun 11 23:26:25 same Jun 11 23:26:41 Oh lol Jun 11 23:26:49 There's no resize2fs Jun 11 23:26:56 in reference rootfs Jun 11 23:27:01 run it in twrp Jun 11 23:27:15 (if it's not broken as mount :P) Jun 11 23:27:23 [Edit] (if it's not broken just as mount :P) Jun 11 23:27:31 we'll hope for the best Jun 11 23:27:49 you can always push your own busybox from the out folder and use it Jun 11 23:30:40 weird stuff... Jun 11 23:31:10 `mount` … https://paste.kde.org/pcxhn5blu Jun 11 23:31:51 sda35 is userdata Jun 11 23:34:00 The thing is, I haven't done anything than `mkdir /system/boot` and resizing Jun 11 23:41:39 Ah, now I got it Jun 11 23:42:32 The script succeeds probably Jun 11 23:42:57 As it mounts the partitions Jun 11 23:43:04 @Sergobot, in correct locations? Jun 11 23:43:16 not all of the Jun 11 23:43:16 [Edit] not all of them Jun 11 23:43:21 which are incorrect? Jun 11 23:43:46 give as much info as possible, that's the first time i see an A/B device :P Jun 11 23:43:50 Ok, so Jun 11 23:44:19 Fwd from Sergobot: There're two so-called "slots" that have separate system, boot, modem, and others Jun 11 23:44:19 Fwd from Sergobot: I have briefly researched how that dual-slot thing works. Device boots the pre-selected slot (dual-boot can be enabled through manually setting the bootable slot using TWRP or fastboot), so there's only one set of partitions mounted at a time. I think, it should be possible to find out the slot during haliu Jun 11 23:44:19 m installation. Not sure how exactly to do it but I will research more tomorrow. Jun 11 23:44:20 (Photo, 1280x78) https://irc.ubports.com/50TwqZDg.png Jun 11 23:44:51 (4 months ago, in this chat) Jun 11 23:46:16 @Sergobot, okay, so that's something Jun 11 23:46:44 i am afraid i can't help you any further Jun 11 23:46:48 My idea is Jun 11 23:46:58 Push the image to /data/android-rootfs Jun 11 23:47:14 Mount that at /android Jun 11 23:47:14 And go from there Jun 11 23:47:16 [Edit] Push the image to /data/android-rootfs.img Jun 11 23:47:47 `android-rootfs.img`? Jun 11 23:47:53 I'll work on that, but first eat Jun 11 23:48:00 @Sergobot, Not yet implemented. :) Jun 11 23:48:14 fun Jun 11 23:50:29 @Sergobot, so, the `system.img` gets mounted into `/system/` and all the files are there … but `/system/vendor/`is empty Jun 11 23:50:34 @Sergobot, [Edit] so, the `system.img` gets mounted into `/system/` and all the files are there … but `/system/vendor/` is empty Jun 11 23:51:19 In your case, yes. But not normally. Jun 11 23:51:20 So here's what's going on: Jun 11 23:51:29 On a regular device, the `/system` partition is just that: a partition that gets mounted at `/system` Jun 11 23:51:53 When the device boots, it mounts that ASAP Jun 11 23:51:53 Then, it grabs the `android-initramfs.img` and extracts that into `/` Jun 11 23:52:24 That changed with A/B partitioning. They skipped that last step Jun 11 23:52:53 Now, `/system` (or `system.img` in our case) is root. So `/system` gets mounted at `/` Jun 11 23:53:23 Makes a lot of sense, really. Don't know why they didn't do it in the first place. But what do I know Jun 11 23:53:49 Anyway, I'll hack up a prototype. brb. Jun 11 23:55:32 sure, thanks in advance Jun 12 00:17:01 Alright, building it all now Jun 12 00:19:37 nice Jun 12 00:20:10 @Sergobot, While we wait, can you put `BOARD_USE_LOCAL_INITRD := true` in your BoardConfig.mk? Jun 12 00:20:24 sure Jun 12 00:21:44 Boy I sure do love DigitalOcean Jun 12 00:21:52 I only need one good core... and they got me. Jun 12 00:22:07 ah, they're awesome Jun 12 00:22:13 thanks to github i have some credits ^^ Jun 12 00:22:54 @UniversalSuperBox, rebuilding systemimage now Jun 12 00:23:14 Oh no, that's alright Jun 12 00:23:31 No need without the actual initramfs Jun 12 00:27:09 (Document) https://irc.ubports.com/U4NPMCzM.gz Jun 12 00:27:13 It now occurs to me that your device is arm64 Jun 12 00:27:19 Ah well Jun 12 00:27:49 :D Jun 12 00:28:06 Put that in your device folder at `initramfs.gz` and rebuild Jun 12 00:28:11 Hmm Jun 12 00:28:17 Maybe do `make clean halium-boot` first Jun 12 00:28:38 Then rebuild halium-boot and we should be off to the races Jun 12 00:29:00 last time halium-boot didn't work for me Jun 12 00:29:17 @UniversalSuperBox, when has halium-boot started to support Halium reference rootfs/ Jun 12 00:29:18 @UniversalSuperBox, [Edit] when has halium-boot started to support Halium reference rootfs? Jun 12 00:29:33 🤔 Jun 12 00:29:45 It doesn't? Jun 12 00:29:50 no Jun 12 00:29:57 and it doesn't support plasma as well Jun 12 00:30:09 i told you that like 3 months ago Jun 12 00:30:11 when you just wrote it Jun 12 00:30:25 Well plasma I really can't fix since that's an integration thing Jun 12 00:30:42 @vanyasem, i assume something could've changed, but i sincerely doubt it Jun 12 00:31:17 Does "Installation" in this document not work? https://github.com/Halium/projectmanagement/issues/40 Jun 12 00:31:48 @UniversalSuperBox, I haven't tried that. @Sergobot ? Jun 12 00:31:57 @UniversalSuperBox, [Edit] I haven't tried that (or forgot) @Sergobot ? Jun 12 00:32:12 I was thinking to but didn't Jun 12 00:34:50 @UniversalSuperBox, just to clarify … i put it into my device dir and rename … then rebuild halium-boot (or hybris-boot?) and systemimage Jun 12 00:35:15 right? Jun 12 00:35:24 It'll only work with halium-boot Jun 12 00:35:29 But that'll be the only thing that needs to be built Jun 12 00:39:26 ok, booted Jun 12 00:44:31 `systemctl edit lxc\@android.service` … when running this, the editor is blank Jun 12 00:44:47 is it a problem? Jun 12 00:45:20 Yes. `edit` creates a second file which is overlaid on the first one Jun 12 00:45:20 Little-known feature of systemd, very useful. Jun 12 00:48:07 lxc@android still fails, now with a different error Jun 12 00:48:25 https://paste.kde.org/pcz0z3pzz Jun 12 00:49:59 You'll need to reboot first Jun 12 00:50:13 Well, I did reboot Jun 12 00:50:37 After editing the file? systemd is asking for a `daemon-reload` which alerts me that it hasn't Jun 12 00:50:51 yeah, after editing Jun 12 00:51:27 Maybe give it the `systemctl daemon-reload` it wants and reboot Jun 12 00:51:59 done; rebooting Jun 12 00:53:01 same thing Jun 12 00:53:20 including that message about `systemctl daemon-reload` Jun 12 00:53:41 Now why would it be doing that Jun 12 00:55:59 Anyway, what's in `/android` now? Jun 12 00:57:13 `data misc none system vendor` Jun 12 00:57:39 actual partitions are mounted in system and vendor Jun 12 00:58:06 facepalm Jun 12 00:58:18 `mv /userdata/system.img /userdata/android-rootfs.img` Jun 12 00:58:25 Forgot I just made that change, sorry Jun 12 00:58:50 no problem Jun 12 00:59:45 by the way Jun 12 00:59:55 wrong vendor partition got mounted there :/ Jun 12 01:00:13 vendor_a, while i'm running halium in slot b Jun 12 01:00:37 There is no handling for slot changes Jun 12 01:00:42 how does halium-boot determine what to mount and where? Jun 12 01:00:45 i know Jun 12 01:00:55 Whichever fstab it finds first, I suspect Jun 12 01:01:11 there's only one entry in fstab Jun 12 01:01:23 it's just `.../by-name/vendor` Jun 12 01:01:41 That'll do it Jun 12 01:02:01 Can you do `cat /proc/cmdline`? Jun 12 01:02:13 I'm honestly not sure how init is notified of the running slot Jun 12 01:02:31 there's a cmdline flag, yeah Jun 12 01:02:51 Which one? Jun 12 01:03:01 lemme boot into halium Jun 12 01:03:42 https://paste.kde.org/pcriyqvvq Jun 12 01:03:52 yeesh Jun 12 01:03:56 Noisy Jun 12 01:03:57 `androidboot.slot_suffix=_b` Jun 12 01:04:21 Oh, that's an easy one then. I don't think I'll be doing it tonight though. Jun 12 01:04:32 So `..by-name/vendor` is slot A? Jun 12 01:04:42 And there's a `..by-name/vendor_a` and `_b`? Jun 12 01:04:49 @UniversalSuperBox, this way Jun 12 01:05:04 Okay, and there's one of those for... which partitions? Jun 12 01:05:11 on android `../by-name/vendor is just a symlink` Jun 12 01:05:19 [Edit] on usual android `../by-name/vendor` is just a symlink Jun 12 01:05:28 [Edit] on android `../by-name/vendor` is just a symlink Jun 12 01:05:39 (Photo, 1280x78) https://irc.ubports.com/0C2o3XoN.png Jun 12 01:05:44 ^ all the partitions Jun 12 01:06:14 Okay, so I'd need to check for all the partitions with a slot suffix then Jun 12 01:06:32 Looks variable per-device Jun 12 01:08:17 @UniversalSuperBox, i moved it but now neither system nor vendor get mounted Jun 12 01:08:38 So there's nothing there at all? Jun 12 01:08:46 yeah Jun 12 01:09:00 Can you just give me the full `dmesg` output? Jun 12 01:09:19 only data is mounted Jun 12 01:09:21 one sec Jun 12 01:10:24 https://paste.kde.org/pclo682te Jun 12 01:10:44 lol Jun 12 01:11:21 Oh, duh. I se. Jun 12 01:11:23 [Edit] Oh, duh. I see. Jun 12 01:15:38 Rebuilding Jun 12 01:17:13 ah, i haven't told you about how system gets mounted on A/B Jun 12 01:17:16 `root=/dev/dm-0 dm="system none ro,0 1 android-verity /dev/sda34"` Jun 12 01:17:27 (nothing to worry about now, though) Jun 12 01:17:38 We don't touch the system partition at all Jun 12 01:18:00 yeah, but bootloader doesn't care Jun 12 01:18:17 @Sergobot, so it gives that as part of cmdline Jun 12 01:18:42 `root=` is the only thing I'd be concerned about, but we've already seen that `halium-boot` ignores it Jun 12 01:18:43 and kernel mounts system partition as linear device-mapper target Jun 12 01:19:00 on some other device? Jun 12 01:19:09 i use a dirty hack to ignore it :/ Jun 12 01:19:24 Okay, what? Jun 12 01:20:44 (Photo, 683x102) https://irc.ubports.com/wAK058TK.png in obsolete_checksetup() in init/main.c Jun 12 01:21:09 The bootloader mounts that? Jun 12 01:21:18 no lol Jun 12 01:21:22 kernel does Jun 12 01:21:38 but bootloader says to mount it at early stage Jun 12 01:26:03 (Document) https://irc.ubports.com/jS0kRNCk.gz Jun 12 01:26:21 There, all better Jun 12 01:26:27 And it's arm64 this time! Jun 12 01:26:47 yay! Jun 12 01:30:57 no difference, hmm Jun 12 01:31:22 Bah. I need to leave soon, but how about `dmesg` before I go? Jun 12 01:31:37 sure Jun 12 01:32:29 https://paste.kde.org/pkws6d3qf Jun 12 01:36:41 Thanks for all this, @UniversalSuperBox Jun 12 01:36:47 Gotta go now too Jun 12 01:36:50 Hmm... Did you do `make clean halium-boot` first? Jun 12 01:36:52 Ah, okay Jun 12 01:37:03 No problem, we'll continue some other time Jun 12 01:37:14 yep Jun 12 01:38:00 "initrd: mounting android system image ro" is the old message from before my latest change. I think you didn't get a new initramfs, cleaning should fix that Jun 12 01:45:28 I'm expecting "mounting /tmpmnt/android-rootfs.img ro" **** ENDING LOGGING AT Tue Jun 12 03:00:04 2018