**** BEGIN LOGGING AT Sun Apr 22 03:00:05 2018 Apr 22 03:28:15 hi Apr 22 03:28:34 can i ask some semi-offtopic question? Apr 22 03:34:43 i suceeded to run ubuntu arm64 with lxde, ssh and x11 in chroot om my xperia XZ using linux deploy app . Im using XServer XSDL app as x.org/x11 server on android. now i want to run it without linux deploy app. but im getting this error when trying chroot commad... Apr 22 03:34:53 (Photo, 720x1280) https://irc.ubports.com/0c8VnqZT.png Apr 22 03:35:16 (Document) https://irc.ubports.com/8oBVW1z0.null Apr 22 03:35:17 here is my install-recovery script Apr 22 03:35:24 (Document) https://irc.ubports.com/QWfj5fdl.null Apr 22 03:35:32 and my /system/etc/linux.sh Apr 22 03:37:00 maybe mistake in last line at linux.sh, (chroot ...) i dont know what command call with chroot to initializate ubuntu. Apr 22 03:37:09 can anyone help me? Apr 22 04:33:06 I'll also leave a question here and go to bed 😅 … I tried halium-install command twice, but I always get the output "killed" when I'm unpacking rootfs to system-image Apr 22 04:33:22 Could someone help me out, please? Apr 22 05:50:17 @mayltonfnds, finally, im going to sleep too, after night shift. :D Apr 22 05:50:32 @mayltonfnds, [Edit] finally, im going to sleep too, after night shift. 😂 Apr 22 07:03:09 @mayltonfnds, please uplode the full console output to some pastebin service like paste.ubuntu.com Apr 22 08:37:21 olprt was added by: olprt Apr 22 10:16:16 I'm just going to add this : https://github.com/reivilibre/halium-scripts/blob/master/halium-install-hostbuild into the mix; it's an adapted `halium-install` script which avoids doing the packing on the target device. Apr 22 10:24:38 alternatively there is also https://github.com/JBBgameich/halium-install which looks rather professional but I've not used it (and there seem to be some more dependencies) Apr 22 11:14:56 @dohniks, https://paste.ubuntu.com/p/Cjdz6jjXbq/ Apr 22 11:16:16 Is it possible that you run out of memory on the target device? :/ Apr 22 11:16:23 mhm. not very enlightning Apr 22 11:16:42 A random 'Killed' message Apr 22 11:16:57 dmesg? Apr 22 11:17:10 from the device Apr 22 11:17:12 Once it took like 15 minutes to get that message Apr 22 11:17:21 I thought it would work Apr 22 11:17:38 yeah ... these install scripts are not fun to debug Apr 22 11:17:56 you can also try the one from JBBgameich Apr 22 11:18:00 https://github.com/JBBgameich/halium-install Apr 22 11:18:10 ok Apr 22 11:18:20 That sounds like it's taking a long time Apr 22 11:18:21 I agree, `dmesg` reveals all Apr 22 11:18:26 I adapted this: https://github.com/reivilibre/halium-scripts/blob/master/halium-install-hostbuild to do the packing on the host system Apr 22 11:18:32 or JBBgameich's looks good also Apr 22 11:18:48 there's a selection :-) Apr 22 11:18:50 or that yeah Apr 22 11:19:03 or read the script and execute the commands one by one Apr 22 11:27:12 @dohniks Have you found that getting the WiFi up with `echo 1 > /dev/wcnss_wlan` actually works on Qualcomm devices on the reference rootfs? For me, it hangs for about 60 seconds, tells me 'Bad address' and dmesg includes `subsys-pil-tz a21b000.qcom,pronto: wcnss: Failed to locate wcnss.mdt` Apr 22 11:27:56 mhm. it works on my device: nexus 7 2013 "deb" Apr 22 11:28:50 I've found that `/etc/udev/rules.d/50-firmware.rules` is zero-bytes Apr 22 11:30:04 Maybe I should re-download the rootfs and see that it isn't corrupted Apr 22 11:33:32 @lambdaperl[m], https://paste.ubuntu.com/p/zTGqVJ5dVX/ Apr 22 11:34:09 @lambdaperl did you do this: http://docs.halium.org/en/latest/porting/debug-build/udev.html?highlight=udev#add-udev-rules Apr 22 11:35:40 * lambdaperl[m] sent a long message: lambdaperl[m]_2018-04-22_11:35:40.txt Apr 22 11:37:09 lxc-info -n android Apr 22 11:37:42 * lambdaperl[m] sent a long message: lambdaperl[m]_2018-04-22_11:37:41.txt Apr 22 11:39:09 you are in ssh in halium reference right? Apr 22 11:39:15 [Edit] you are in ssh in halium reference rootfs right? Apr 22 11:39:58 @dohniks Yes, I'm in ssh, where `test_lights` and `test_vibrator` are Apr 22 11:40:15 you booted with hybris-boot? Apr 22 11:40:31 I booted with `hybris-recovery`... is that maybe the problem? Apr 22 11:40:53 dunno Apr 22 11:41:00 can you try hybris-boot? Apr 22 11:41:25 yeah, I will try that now, as I don't need to be using the hybris-recovery anymore now Apr 22 11:42:40 @mayltonfnds Not sure what to tell you, but it looks like `system.img` doesn't exist on your host at the time that it's about to be pushed to the device Apr 22 11:42:56 To make it tricky, the script also cleans up the evidence after itself Apr 22 11:42:57 Can you maybe edit `halium-install-standalone.sh` and add `exit 1` after the line `adb push $IMAGE_DIR/system.img /data/` Apr 22 11:45:15 Sorry that line might be in `functions/core.sh` Apr 22 11:46:13 speaking of which, @JBBgameich do you see any way how the install script could be made more debug friendly. I don't see an easy solution tbh, but maybe someone has a smart idea Apr 22 11:46:37 @lambdaperl[m], I can try Apr 22 11:47:47 Maybe not particularly elegant, but it could help to have a script which is extremely linear and maybe even can be told to pause at every step ? Apr 22 11:50:05 what I did is, us the "normal" halium-install (which is pretty linear and unfancy) and run the commands one by one from command line ... Apr 22 11:50:52 (Photo, 1366x768) https://irc.ubports.com/QfT3k3pT.png Apr 22 11:50:57 Is this the line I have to edit? Apr 22 11:51:38 @dohniks, let me try this then Apr 22 11:51:58 Splitting the logic into functions might make the script code look relatively nice, but when you're debugging and not familiar with the script, it might be easier if the commands were all in sequence Apr 22 11:51:58 I found the default halium-install confusing, because it referred to the `rootfs`as `system-image` in a script where it's also handling a different image called `system.img` Apr 22 11:53:20 @mayltonfnds, Well, I can't find out what is or not a command 😂 Apr 22 11:56:23 @mayltonfnds, I hear you Apr 22 11:56:31 possibly a bit crude to recommend my own adapted script, but you could try https://raw.githubusercontent.com/reivilibre/halium-scripts/master/halium-install-hostbuild Apr 22 11:57:14 @lambdaperl[m], Any possibility is worth it a try Apr 22 12:00:24 @mayltonfnds are you able to download it and make it executable? Apr 22 12:00:35 Yes Apr 22 12:00:41 it's running already Apr 22 12:01:11 looks like i'll work Apr 22 12:01:22 [Edit] looks like it'll work Apr 22 12:02:20 Well, it almost did Apr 22 12:02:37 https://paste.ubuntu.com/p/jsHcjhk6Th/ Apr 22 12:03:41 could you issue `df -h` ? Apr 22 12:04:21 Sorry? I couldn't get Apr 22 12:04:43 Run the command `df -h` which will show how much free space you have etc Apr 22 12:04:43 I'm noob Apr 22 12:06:06 @lambdaperl[m], https://paste.ubuntu.com/p/S8PmnprWh5/ Apr 22 12:06:55 hmm, could you run `file /path/to/system.img` with the system image you told the script to use? Apr 22 12:07:28 @lambdaperl[m], sure Apr 22 12:08:22 https://paste.ubuntu.com/p/MbbSDZ86gw/ Apr 22 12:10:16 woops, I missed the bit where it looks like your device is out of space, could you do `adb shell df -h` ? Apr 22 12:11:26 https://paste.ubuntu.com/p/yt62VptYHc/ Apr 22 12:12:36 I did s backup already. I can clean system, data, if necessary Apr 22 12:12:51 [Edit] I did a backup already. I can clean system, data, if necessary Apr 22 12:13:08 ahh sorry I sent myself on a wild goose chase yet again; I need to add some more status messages to this thing, seriously :-) Apr 22 12:13:08 Can you tell me the filesize of the system.img you're trying to use? Apr 22 12:13:52 523,7MB Apr 22 12:14:56 @mayltonfnds, I think you should Apr 22 12:15:03 @dohniks, I noticed that Apr 22 12:15:26 Ahhh, you see, that's a bit of a whopper. My system.img is 146MiB and the `halium-install` script allocates 160MiB for it.. Apr 22 12:16:00 @lambdaperl[m], Maybe because of device tree? Apr 22 12:16:46 You might want to look into just *what* is quite so large in there at some point, but if you change the `halium-install-hostbuild` script , go to line 49 and change `make_ext4fs -l 160M $WORKDIR/system.img $TMPMOUNT` to `make_ext4fs -l 560M $WORKDIR/system.img $TMPMOUNT Apr 22 12:18:09 And yes, it looks like to me that you will want to do a wipe of /data using TWRP at some point Apr 22 12:19:32 I'll wipe and try again with changes Apr 22 12:20:04 OK :-) Apr 22 12:22:32 Looks like it worked Apr 22 12:22:37 😁😁 Apr 22 12:22:45 hooray :-) Apr 22 12:22:45 nice Apr 22 12:23:29 Now I have to check whether it boots Apr 22 12:25:28 phone bootloops 😂 Apr 22 12:25:50 last_kmsg Apr 22 12:26:09 or ramoops if on later kernel than 3.4 I believe it was Apr 22 12:27:02 @lambdaperl[m], I believe it's 3.18 Apr 22 12:27:06 I'm not sure Apr 22 12:27:30 @lambdaperl[m], Wait, are you talking about my computer kernel or my halium kernel? Apr 22 12:27:45 Halium kernel :-) Apr 22 12:28:15 @lambdaperl[m], Oh, so yes. I think it is 3.18 Apr 22 12:29:00 Basically, when your phone is bootlooping, hold down the key to go into bootloader, get yourself in TWRP recovery Apr 22 12:29:22 @lambdaperl[m], done Apr 22 12:29:35 open up `adb shell` on the host Apr 22 12:29:58 done Apr 22 12:30:15 `cd /sys/fs/pstore/ && ls` Apr 22 12:30:36 👍 Apr 22 12:30:46 Anything there? Apr 22 12:31:02 no Apr 22 12:31:07 it's empty Apr 22 12:31:25 hmm, ok you can try `ls /proc/last_kmsg` but I'm not too hopeful about it Apr 22 12:32:30 I have to do it inside /sys/fs/pstore/ ? Apr 22 12:33:28 nah you can do that anywhere (in adb shell) Apr 22 12:33:29 just interested in seeing if that file exists Apr 22 12:33:55 @lambdaperl[m], No, it doesn't Apr 22 12:33:59 Just checked Apr 22 12:35:14 Maybe because I cleaned system/data/cache? Apr 22 12:35:26 I don't know actually Apr 22 12:35:35 No, that's OK, `/proc` isn't a real filesystem Apr 22 12:36:16 I see Apr 22 12:37:26 The first thing to check is whether you have PSTORE enabled in your kernel defconfig Apr 22 12:37:27 Key options that should be enabled (=y) include: `CONFIG_PSTORE,` `CONFIG_PSTORE_CONSOLE`, `CONFIG_PSTORE_RAM`, `CONFIG_PSTORE_RAM_ANNOTATION_APPEND` Apr 22 12:37:56 The other thing to try is `hybris-recovery` as described here: https://docs.halium.org/en/latest/porting/debug-build/early-init.html#forcing-debug-mode Apr 22 12:39:32 I see Apr 22 12:40:05 with `hybris-recovery` installed as your boot image, it will try and give you debug access as early as possible into the boot cycle Apr 22 12:47:53 CONFIG_PSTORE=y … CONFIG_PSTORE_CONSOLE=y … CONFIG_PSTORE_PMSG=y … CONFIG_PSTORE_RAM=y Apr 22 12:47:58 those lines are enabled Apr 22 12:48:51 Hmm, well give `hybris-recovery` a shot Apr 22 12:49:33 👍 Apr 22 13:00:24 I just flashed hybris-recovery. Now phone doesn't boot but stuck on splash screen. I'm going to debug it now (try) Apr 22 13:02:25 thats ok. thats expected Apr 22 13:02:31 recovery doesnt have a ui Apr 22 13:02:36 get telnet access Apr 22 13:02:51 and post output of mount Apr 22 13:08:58 https://paste.ubuntu.com/p/P4Jff5tR7Y/ Apr 22 13:09:27 halium documentation says to first run cat diagnosis.log Apr 22 13:09:51 sure doesnt' hurt :) Apr 22 13:10:22 ah ok, no such file, means there was no problem diagnosed Apr 22 13:10:29 mount? Apr 22 13:10:36 Now I have to just type "mount"? Apr 22 13:10:41 jup Apr 22 13:11:28 https://paste.ubuntu.com/p/DxmrC5tZHF/ Apr 22 13:13:07 echo "mount_stowaways" >/init-ctl/stdin Apr 22 13:13:21 it was suggested a little bit up Apr 22 13:13:47 I think the mount output surprising but lets try mounting anyway Apr 22 13:13:52 Sorry Apr 22 13:13:59 ? Apr 22 13:14:07 no need to apologize for anything Apr 22 13:14:22 I just wanted to see what is mounted after booting Apr 22 13:14:24 now we DO the mount Apr 22 13:14:27 then we check again Apr 22 13:16:14 Nothing happened Apr 22 13:16:32 well, who knows Apr 22 13:16:37 what is the output of mount now? Apr 22 13:17:46 https://paste.ubuntu.com/p/9Vt3DKR8CH/ Apr 22 13:18:42 mhm ... that last line doesn't make any sense to me. now we have it four times ... makes ... uhm .... four times as little sense Apr 22 13:18:50 ls -l /data Apr 22 13:20:02 no output Apr 22 13:20:09 mount /data Apr 22 13:20:11 then again Apr 22 13:20:15 ls -l /data Apr 22 13:20:42 mount: can't find data in /etc/fstab Apr 22 13:20:53 ha Apr 22 13:21:38 http://docs.halium.org/en/latest/porting/build-sources.html?highlight=fstab#include-your-device-in-fixup-mountpoints Apr 22 13:21:41 did you do that? Apr 22 13:21:52 yes Apr 22 13:22:08 can you post your fixup file? Apr 22 13:22:25 and Apr 22 13:22:30 cat /etc/fstab Apr 22 13:22:32 on the device Apr 22 13:22:51 I think something went wrong there Apr 22 13:27:57 where fix-mountpoints is located btw? Apr 22 13:28:30 ;) I see what you did there Apr 22 13:31:11 (Photo, 1366x768) https://irc.ubports.com/AjMjQYyP.png Apr 22 13:32:09 Maybe because of that "x2") line being backwards ? Apr 22 13:33:42 x2 is your device? Apr 22 13:33:53 looks ok syntactically Apr 22 13:34:04 what does the /etc/fstab look like? Apr 22 13:34:08 yes Apr 22 13:35:46 oh wait there is sda sde and sdf is that true? Apr 22 13:36:06 that would mean three different "hard disks"/storage devices Apr 22 13:36:23 maybe post also from device: Apr 22 13:36:29 fdisk -l /dev/sda Apr 22 13:36:34 fdisk -l /dev/sde Apr 22 13:36:39 fdisk -l /dev/sdf Apr 22 13:37:13 on telnet server, right? Apr 22 13:37:17 yes Apr 22 13:38:29 Any ideas on how to dig into an issue where, after leaving hybris-recovery, the USB status goes to 'Mer Debug: done debug, disabling storage' but SSH doesn't come up? Apr 22 13:38:42 https://paste.ubuntu.com/p/2zRV2Z7GTN/ Apr 22 13:39:38 @lambdaperl last_kmsg (or that other newer thing) Apr 22 13:40:26 Nothing from that, there is no bootloop or anything, I have to manually repower it Apr 22 13:40:46 @mayltonfnds intersting. you DO have three (at least) devices. and sda9 is /userdata Apr 22 13:41:12 so please cat /etc/fstab Apr 22 13:41:44 I might see if I can compile last_kmsg in, as it sounds like it doesn't matter whether or not the kernel panics, it would still give you output, right? Apr 22 13:41:46 @lambdaperl and you have a newer kernel? no last_kmsg? Apr 22 13:42:10 I do always get it regardless of panic Apr 22 13:42:16 @dohniks, https://paste.ubuntu.com/p/pvjx7YKhrB/ Apr 22 13:42:17 kernel version is 3.18.49-reivilibre-002-g7f3649c-dirty Apr 22 13:42:27 the only trick is to NOT turn it OFF but actually REBOOT Apr 22 13:42:42 no last_kmsg, I'll see if I can compile that in Apr 22 13:42:43 yeah, I guess that means you have that newer thing taht only talks if it panics :( Apr 22 13:42:53 not sure how hard that is Apr 22 13:43:08 maybe rather try the lxc disabling? Apr 22 13:43:20 does holding down the power button reboot or turn off, I wonder Apr 22 13:43:34 it reboots on my device Apr 22 13:43:59 @mayltonfnds no data in that fstab .... mhm Apr 22 13:44:35 😞 Apr 22 13:44:47 try Apr 22 13:44:56 mount /dev/sda9 /data Apr 22 13:45:48 mount: mounting /dev/sda9 on /data failed: Invalid argument Apr 22 13:46:12 oh Apr 22 13:46:15 dmesg Apr 22 13:46:24 that can be caused by a wrong filesystem type Apr 22 13:46:56 @dohniks, I mean this: http://docs.halium.org/en/latest/porting/debug-build/early-init.html?highlight=mask#the-device-reboots-after-leaving-hybris-recovery Apr 22 13:47:07 @dohniks, That gave me a really big output lol Apr 22 13:47:12 yes Apr 22 13:47:18 maybe only look at the end Apr 22 13:47:29 we're looking for hints why the mount failed Apr 22 13:48:09 and indeed as lambdaperl said could be wrong filesystem Apr 22 13:48:16 <3>[ 2880.967335] EXT3-fs (sda9): error: couldn't mount because of unsupported optional features (40) … <3>[ 2880.967601] EXT2-fs (sda9): error: couldn't mount because of unsupported optional features (44) … <4>[ 2880.970666] EXT4-fs warning (device sda9): ext4_enable_quotas:5326: Failed to enable quota tracking (type=0, err=-3) Apr 22 13:48:16 . Please run e2fsck to fix. … <3>[ 2880.970869] EXT4-fs (sda9): mount failed Apr 22 13:48:17 @dohniks with LXC disabled & masked, it still doesn't give me any luck Apr 22 13:48:44 https://paste.ubuntu.com/p/9FgN92vmzh/ Apr 22 13:48:55 we may have found the problem Apr 22 13:49:26 mhm @lambdaperl and you are positive you don't get ssh? you did the ip link thing right? Apr 22 13:49:51 Yes, no SSH, but I gtg now, will see you later Apr 22 13:50:02 ok Apr 22 13:50:34 @mayltonfnds you said you did a wipe/clean/whateveritscalled Apr 22 13:50:45 I did Apr 22 13:50:49 so then I'd assume a new ext4 fs Apr 22 13:50:58 so how can it be ... messed up Apr 22 13:51:18 No idea 😂😂 Apr 22 13:51:22 but i guess we can do as it says Apr 22 13:51:56 (thinks whether there is any risk to this ... ) Apr 22 13:52:04 (I guess not, you do have back ups) Apr 22 13:52:33 e2fsck /dev/sda9 Apr 22 13:53:53 Do you think an f2fs twrp could cause that problem? Apr 22 13:53:59 You need to do "wipe data" in recovery Apr 22 13:54:07 A friend of mine has just given me that hint Apr 22 13:54:17 @UniversalSuperBox, I did Apr 22 13:54:18 Not just go to "wipe" and select "data" and swipe Apr 22 13:54:57 There's a special option that removes encryption and actually destroys all your data Apr 22 13:56:09 I see Apr 22 13:56:41 @dohniks, # e2fsck /dev/sda9 … /bin/sh: e2fsck: not found Apr 22 13:56:50 your backups are not ON the device right? Apr 22 13:57:01 Sure not 😂😂 Apr 22 13:57:06 are on my computer Apr 22 13:57:08 okok, just checking Apr 22 13:57:18 Be careful Apr 22 13:57:32 what dalton said, do the wipe in twrp again Apr 22 13:57:35 Twrp backups don't back up the media folder where all your files are kept Apr 22 13:57:57 So copy all of those files that you stored on your device. Pictures and stuff Apr 22 13:59:16 I'm trying to boot plasma mobile off of halium-boot. has anybody done that? Apr 22 13:59:31 Do I need to replace the lxc pre-start stuff? Apr 22 13:59:35 Got all the changes in #40? Apr 22 13:59:44 lemme see Apr 22 14:00:41 nope I don't Apr 22 14:00:57 working on it Apr 22 14:21:18 I don't have /android/vendor and /android/system ... so ... I just create them I guess Apr 22 14:27:12 That means your system image isn't getting mounted at all Apr 22 14:28:32 uh, no I meant in twrp chroot Apr 22 14:29:01 I do basically everythign in #40 in a twrp chroot right? Apr 22 14:29:19 Only the folder `/android` needs to be there Apr 22 14:29:55 "We'll also need to make some links to locations in /android" Apr 22 14:29:57 I think I used `ln -sf` to force linking so it'll happen whether or not the target exists Apr 22 14:35:34 it booted to ssh, no UI, lxc says it's running, I can't see system.img being mounted anywhere Apr 22 14:37:25 Output of `dmesg` then? Apr 22 14:43:19 http://paste.ubuntu.com/p/3sKw8yxpw4/ Apr 22 14:46:10 `init-script: mounting stowaways DATA_PARTITION="/dev/mmcblk0p30", data_subdir=""` Apr 22 14:46:16 I, uh, don't think you're using halium-boot Apr 22 14:46:56 pfff ... entirely possible Apr 22 14:47:08 I'm banging my head against the wall for so long Apr 22 14:47:15 can't see the forest anymoore Apr 22 14:47:27 lemme flash it and reboot Apr 22 14:52:56 now the screen went black Apr 22 14:52:59 I think it was still stuck on bootlogo before Apr 22 14:53:08 but no ui still Apr 22 14:53:10 http://paste.ubuntu.com/p/G2yM7TgPMX/ Apr 22 14:53:16 test_hwc segfaults Apr 22 14:53:24 Sounds like it's trying to start Apr 22 14:53:25 test_lights works Apr 22 14:53:37 yeah I think so Apr 22 14:54:29 Is `/vendor` a valid link? Apr 22 14:56:06 ls -ld /vendor /android/vendor … ls: cannot access '/android/vendor': No such file or directory … lrwxrwxrwx 1 root root 15 Apr 22 14:21 /vendor -> android/vendor/ Apr 22 14:56:44 I've got a PR for that. https://github.com/Halium/initramfs-tools-halium/pull/13 Apr 22 14:56:56 For now you can do Apr 22 14:57:36 ```cd /android … ln -sf system/vendor vendor``` Apr 22 14:58:35 ok. done. rebooting Apr 22 15:01:28 a lot more stuff is mounted now, screen still black test_hwc still segfaults Apr 22 15:02:44 Why does it segfault Apr 22 15:02:48 What happens Apr 22 15:04:25 http://paste.ubuntu.com/p/x7r94Q384Z/ Apr 22 15:04:44 dunno it doesn't say anything else Apr 22 15:05:31 Anything in logcat when it fails? Apr 22 15:05:47 lemme check Apr 22 15:05:49 It's getting far enough to turn on and off the display, at least Apr 22 15:06:05 in dmesg there are these binder transaction failed lines Apr 22 15:06:11 haven't seen them earlier Apr 22 15:07:16 https://paste.ubuntu.com/p/RKvkcN69QP/ Apr 22 15:07:31 wats that Apr 22 15:07:46 the logcat output during test_hwcomposer segfault Apr 22 15:07:54 Nothing before or after it? Apr 22 15:08:31 Well I started /system/bin/logcat Apr 22 15:08:34 wait a second Apr 22 15:08:38 hit enter a couple times Apr 22 15:08:42 go to other terminal Apr 22 15:08:46 run tst_hwc Apr 22 15:08:49 come back Apr 22 15:08:54 that's all there is Apr 22 15:09:26 there was stuff BEFORE, but not sure ... do you want that? Apr 22 15:09:33 Well how about a backtrace then Apr 22 15:09:59 ok lemme think whether I remember :) Apr 22 15:10:29 `EGL_PLATFORM=hwcomposer gdb test_hwcomposer` Apr 22 15:10:31 `run` Apr 22 15:10:33 `bt full` Apr 22 15:11:54 https://paste.ubuntu.com/p/qzgMDCRRR5/ Apr 22 15:12:29 wait Apr 22 15:12:31 WHAT Apr 22 15:13:07 Segmentation fault at `strlen()`? Apr 22 15:13:24 sounds kinda fundamental ... Apr 22 15:13:45 But... that's what happened on the... 5X Apr 22 15:14:07 so is halium-boot/ubports-boot at fault? Apr 22 15:14:24 It could very well be... but how? Apr 22 15:14:31 It's Nexus 7 2013 ? Apr 22 15:14:35 yes Apr 22 15:14:37 Yeah, kernel 3.4 Apr 22 15:14:37 and PlaMo rootfs? Apr 22 15:14:41 yes Apr 22 15:15:00 IDK... you're halium-boot dev Apr 22 15:15:01 :P Apr 22 15:15:54 I could try halium rootfs Apr 22 15:16:09 but I've no clue whether that makes any sense Apr 22 15:18:16 But Apr 22 15:18:18 but how Apr 22 15:20:17 Can you `strace` it? Apr 22 15:20:26 strace of test_hwc please Apr 22 15:20:27 ^5 Apr 22 15:20:49 @bhushanshah, ☜(゚ヮ゚☜) Apr 22 15:22:03 https://paste.ubuntu.com/p/gTV3wbV242/ Apr 22 15:23:50 So it tries to print a string Apr 22 15:23:55 And segfaults Apr 22 15:25:06 is it that linker thing again ... mind you, I'm just talking out of my ... aehem ... Apr 22 15:26:05 No Apr 22 15:26:12 we use n linker Apr 22 15:26:13 open("/usr/lib/arm-linux-gnueabihf/libhybris/linker/n.so", O_RDONLY|O_CLOEXEC) = 3 Apr 22 15:26:32 ah Apr 22 15:26:48 so it tries to print to logcat and fails Apr 22 15:26:57 if I understan this corretly Apr 22 15:27:35 ahah Apr 22 15:28:06 ah wait Apr 22 15:28:09 it was android side fix Apr 22 15:28:31 @UniversalSuperBox how do you bind mount sockets in container? Apr 22 15:29:23 https://gist.github.com/UniversalSuperBox/9421e4f1350e8959d2205cb3ad14a944#file-pre-start-sh-L19 Apr 22 15:30:34 hm Apr 22 15:30:37 I am out of idea Apr 22 15:30:41 (s) Apr 22 15:31:00 IIRC @mariogrip had fixed strlen crash earlier no? Apr 22 15:31:01 Should we be able to echo into `/dev/socket/logdw` Apr 22 15:31:06 no Apr 22 15:31:10 that's socket Apr 22 15:31:11 oh Apr 22 15:31:14 so yoou need to connect Apr 22 15:31:55 But what it's trying to write... I think that comes out in logcat? Apr 22 15:32:12 https://paste.ubuntu.com/p/RKvkcN69QP/ nevermind Apr 22 15:32:16 wait Apr 22 15:32:28 It does come out Apr 22 15:33:10 I think it's symlink mess Apr 22 15:33:28 maybe not though Apr 22 15:33:42 I don't understand Apr 22 15:33:49 symlink mess? Apr 22 15:34:22 nvm me Apr 22 15:55:47 Guys, my internet ran away Apr 22 15:56:02 😂😂 Apr 22 15:56:04 I'm back with mobile data Apr 22 15:56:36 Well, I did format dada of my device and that fixed the problem with data partition Apr 22 15:56:47 oh cool Apr 22 15:56:52 what did you do different this time? Apr 22 15:57:38 I completely formated my storage Apr 22 15:57:38 not just wiped Apr 22 15:57:49 ah Apr 22 15:58:39 so, where are you at now? Apr 22 15:59:22 I did all that stuff related to telnet Apr 22 15:59:44 I did this: Apr 22 15:59:45 mkdir /a … mount /data/rootfs.img /a … chroot /a /bin/bash … . /etc/environment … passwd Apr 22 16:00:10 and this> exit … umount /a … rmdir /a … sync Apr 22 16:00:33 now I'm stuck at this Apr 22 16:00:34 You may now reboot your device into Halium. … Once your device is booted, you should see that you have a new network interface again. Assign it an IP of 10.15.19.100. Once finished, you should be able to do the following to log in: … ssh root@10.15.19.82 Apr 22 16:01:37 no response from ssh? Apr 22 16:02:00 I actually have not sure what to do Apr 22 16:02:14 I rebooted phone, but stucks at the same screen Apr 22 16:02:20 the splash screen Apr 22 16:02:38 It doesn't boot Apr 22 16:02:39 that's ok... we'll get to the pixels later :) Apr 22 16:03:07 What should I do? Apr 22 16:03:41 ping 10.15.19.100 Apr 22 16:04:07 uhm I mean Apr 22 16:04:15 ping 10.15.19.82 Apr 22 16:04:58 But I have to set up 10.15.19.100 first I suppose Apr 22 16:05:02 yes Apr 22 16:05:14 same as you did before for telnet Apr 22 16:05:19 just different numbers Apr 22 16:05:28 ok Apr 22 16:06:03 so .100 ending is my ip and .82 ending is my route? Apr 22 16:06:26 yeah Apr 22 16:06:30 100 is your laptop Apr 22 16:06:33 82 the device Apr 22 16:06:40 you want the route to the device Apr 22 16:06:57 Got it Apr 22 16:10:41 https://paste.ubuntu.com/p/nZKJZfYQ4P/ Apr 22 16:11:29 ip link ; ip route Apr 22 16:11:29 seems that previous ip is interfering on connection Apr 22 16:12:51 https://paste.ubuntu.com/p/KkTZ925jh9/ Apr 22 16:12:59 I messed up things, didn't I? Apr 22 16:16:01 I think I'll format again 😂😂 Apr 22 16:16:11 hehe Apr 22 16:17:06 the laptop this time, right? just make sure to take a backup of your laptop and store it on the .... uhh ... oh wait no :) Apr 22 16:17:11 It doesn't hurt Apr 22 16:17:28 I don't think it would help Apr 22 16:17:35 it's something with the ip stuff Apr 22 16:17:41 @dohniks, Why the laptop 😂😂 Apr 22 16:17:47 just kidding Apr 22 16:18:21 😂😂 Apr 22 16:18:26 DONT FORMAT YOUR LAPTOP! Apr 22 16:18:45 That was not a option 😂👍 Apr 22 16:18:55 [Edit] That was not an option 😂👍 Apr 22 16:19:29 ip address show dev USBNETWORK Apr 22 16:19:57 I was just doing that lol Apr 22 16:20:10 👌 Apr 22 16:20:15 https://paste.ubuntu.com/p/tWp2KD98Y7/ Apr 22 16:20:31 ah thats wrong Apr 22 16:20:42 192.168.2.38 Apr 22 16:20:45 is not what you want Apr 22 16:21:10 I see Apr 22 16:21:19 but I'm sure I used the right one Apr 22 16:21:41 sudo ip address add 10.15.19.100 dev USB Apr 22 16:22:30 maybe the network management software of your laptop trying to be helpful and assigning an ip Apr 22 16:22:43 RTNETLINK answers: File exists Apr 22 16:22:44 see ? Apr 22 16:23:49 mhm Apr 22 16:24:03 @mayltonfnds, if you look at it again you will see 10.15.19.100 just below Apr 22 16:24:16 ah yeah Apr 22 16:24:48 I think the problem is that scope global dynamic stuff Apr 22 16:25:33 https://paste.ubuntu.com/p/CwCr7cvtm9/ Apr 22 16:25:33 this is what it looks like for me Apr 22 16:25:41 just added 101 as well ... Apr 22 16:26:27 I'll clear that mess Apr 22 16:26:29 wait a few minutes 😂 Apr 22 16:26:42 [Edit] I'll clean that mess Apr 22 16:28:26 just to clear all Apr 22 16:28:48 What should I do after telnet 192.168.2.15 Apr 22 16:28:50 ? Apr 22 16:29:45 uhm. wait. do you still get the telnet interface? Apr 22 16:29:59 if it works, then it SHOULDNT give you telnet anymore Apr 22 16:30:07 but instead it would continue booting Apr 22 16:30:14 and reach a later state Apr 22 16:30:17 a state with ssh Apr 22 16:30:42 are you using hybris-recovery.img? Apr 22 16:30:45 I just cleaned install rootfs and system.img Apr 22 16:30:58 now i have to go to telnet part, ok? Apr 22 16:31:06 well, you don't HAVE to go there Apr 22 16:31:08 @dohniks, Yes Apr 22 16:31:13 ah ok Apr 22 16:31:21 flash the other one now Apr 22 16:31:26 hybris-boot.img Apr 22 16:31:37 after clean install? no tellnet? Apr 22 16:31:38 the recovery will ALWAYS give you telnet Apr 22 16:31:47 it's for trouble shooting Apr 22 16:31:59 but we shot all your mount trouble Apr 22 16:32:02 so now on to hybris-boot Apr 22 16:32:11 and if it goes well, there will be NO telnet Apr 22 16:33:29 got it Apr 22 16:34:34 well, I installed hybris-boot Apr 22 16:34:48 and it's stuck in splash screen Apr 22 16:35:09 patience, as I said, the pixels come later Apr 22 16:35:12 next up: ssh Apr 22 16:35:46 it gave me some options to use ssh Apr 22 16:36:40 so you're logged in? Apr 22 16:36:58 using that commands? Apr 22 16:37:19 the ip address blabla stuff Apr 22 16:37:45 I think I'll have do redo my root passwd before Apr 22 16:38:01 @mayltonfnds, this thing Apr 22 16:38:05 jup Apr 22 16:38:42 @UniversalSuperBox @bhushanshah if it is "just" the logcat stuff that is acting up. could I disable that somehow? Apr 22 16:40:28 bash: groups: command not found Apr 22 16:40:38 it didn't happen before Apr 22 16:40:49 That's just because the environment isn't sourced. Dont' worry about it. Apr 22 16:44:23 @dohniks, Wonder what would happen if you just removed the socket Apr 22 16:45:49 @UniversalSuperBox, comment out lines 20 and 21? Apr 22 16:45:56 Well taht might do Apr 22 16:46:06 I don't think that would do any good Apr 22 16:47:11 https://paste.ubuntu.com/p/kRzDPXHdJn/ Apr 22 16:49:36 Which script are you using? If you use the one from halium-scripts, you could try github.com/jbbgameich/halium-install. Apr 22 16:49:37 ip route Apr 22 16:50:42 @JBBgameich that was directed to @mayltonfnds right? Apr 22 16:51:12 https://paste.ubuntu.com/p/CfHbWZ49nf/ Apr 22 16:51:35 the route is not there Apr 22 16:52:09 sudo ip route add 10.15.19.82 dev enxdeded8098677 Apr 22 16:52:10 @JBBgameich, That script wasn't working because of my system.img was bigger than 160mb Apr 22 16:52:11 again? Apr 22 16:52:39 ok Apr 22 16:52:40 [Edit] do the ip route add ... again? Apr 22 16:52:52 then check whether it actually gets set Apr 22 16:53:18 that's weird. Now it worked, but I'm sure I did the same Apr 22 16:53:47 well, maybe some network manager who interfers Apr 22 16:54:22 Now I have to run ssh root@10.15.19.82 Apr 22 16:54:36 right? Apr 22 16:54:36 yes Apr 22 16:55:34 @mayltonfnds, in that repository there's also a connect.py script that does all the ip blabla for you Apr 22 16:56:11 sh root@10.15.19.82 … ssh: connect to host 10.15.19.82 port 22: Connection timed out Apr 22 16:56:12 (Sticker, 512x232) https://irc.ubports.com/pDu70eB9.webp Apr 22 16:56:38 route still there? Apr 22 16:57:08 Actually no Apr 22 16:57:09 wtf Apr 22 16:57:17 be faster :) Apr 22 16:57:42 What rootfs are you running? Apr 22 16:58:04 @UniversalSuperBox, latest one i think Apr 22 16:58:14 But which one... who made it? Apr 22 16:58:18 I downloaded from halium documentation page Apr 22 16:58:22 okay Apr 22 16:58:22 halium reference rootfs I'm sure Apr 22 16:58:35 bshah I suppose Apr 22 16:59:02 ssh root@10.15.19.82ssh: connect to host 10.15.19.82 port 22: No route to host Apr 22 16:59:14 C'mooooon Apr 22 16:59:57 did you say you are on a mac? Apr 22 17:00:12 Now I'm on linux mint =) Apr 22 17:00:58 ah, don't know that much better eithter :) can you click through your network management settings try to make it NOT interfere with the usb network Apr 22 17:04:01 it off Apr 22 17:04:01 (Photo, 1366x768) https://irc.ubports.com/8M2bWPFp.png Apr 22 17:06:19 @bhushanshah, fwiw, commenting out 20 and 21, breaks logcat but it doesn't fix test hwc, same segfault at strlen() Apr 22 17:07:42 Something interesting hapened Apr 22 17:07:49 (Photo, 1366x768) https://irc.ubports.com/V1Cxg9aF.png Apr 22 17:08:34 @UniversalSuperBox should I file a bug somewhere, just so we have the reproduction and the logs all in one place? Apr 22 17:08:35 couldn't connect on terminal, but network manager says it's connected Apr 22 17:08:43 I guess? Apr 22 17:09:12 @mayltonfnds the ip route still there? Apr 22 17:09:36 it has changed Apr 22 17:09:52 I thing you are right about network manager Apr 22 17:10:42 [Edit] I think you are right about network manager Apr 22 17:12:32 I just confirmed Apr 22 17:12:58 But how can I stop network manager to insterfer? Apr 22 17:13:03 [Edit] But how can I stop network manager to interfer? Apr 22 17:13:51 not sure Apr 22 17:14:02 systemclt stop NetworkManager Apr 22 17:14:02 ? Apr 22 17:14:05 just guessing Apr 22 17:14:21 in case you lose internet connection, just reboot :) Apr 22 17:14:28 command not found Apr 22 17:14:53 t before the l Apr 22 17:15:40 systemctl stop NetworkManager Apr 22 17:17:27 @dohniks, actually, I havent' really understood your answer to this question Apr 22 17:17:38 you said you did something in the netowrk mgr ui Apr 22 17:17:48 what is the effect on Apr 22 17:17:48 ip route Apr 22 17:22:35 @dohniks, I mean that I think network manager changed my route Apr 22 17:23:18 without network manager I get "Connection refused" Apr 22 17:24:30 ah and that is different right Apr 22 17:24:45 @mayltonfnds, this is what you had before Apr 22 17:26:20 maybe try this disable&mask lxc Apr 22 17:26:21 http://docs.halium.org/en/latest/porting/debug-build/early-init.html?highlight=mask#the-device-reboots-after-leaving-hybris-recovery Apr 22 17:34:03 I don't think that's the reason Apr 22 17:34:39 and I can't get hybrid-recovery shell anymoe once i installed hybris-boot Apr 22 17:37:08 well, I think connection refused means you are past the network problem. in principle you get an answer back from the device. only now the answer is no Apr 22 17:37:37 I don't KNOW that disabling lxc container would help but it's a try Apr 22 17:37:54 you don't necc need hybris boot, you can also do it in a twrp chroot Apr 22 17:38:29 our you re flash recovery, disable&mask, flash boot, ???, profit! Apr 22 17:52:40 Yes, you're right Apr 22 18:13:53 @dohniks, no luck with it Apr 22 18:14:24 @dohniks, I also tried it Apr 22 18:15:49 I tried desabling lxc with hybris-recovery and then installed hybris-boot, but it bootlooped Apr 22 18:58:57 @UniversalSuperBox @bhushanshah total shot in the dark: does the reference in line 205 ot this patch provide any hint? Apr 22 19:00:52 @mayltonfnds I don't see how you could get ssh connection refused with lxc, but a bootloop without lxc. I think something else must be different as well Apr 22 19:01:18 if it bootloops now, maybe you can get a ramoops Apr 22 19:01:25 @dohniks, ...what patch? Apr 22 19:01:42 bruahaha, forgot to post the link ... Apr 22 19:01:49 https://code.launchpad.net/~kzapalowicz/bluez/new-upstream-version-5.42/+merge/306854 Apr 22 19:02:20 I should go to sleep Apr 22 19:04:55 another one https://issues.asterisk.org/jira/browse/ASTERISK-26314 Apr 22 19:05:07 are we for some reason passing a null pointer to strlen Apr 22 19:09:10 @dohniks, does #5 0xb6ac173e in ?? () mean, I have no debug symbols for test_hwc? do you know how I can recompile it with? Apr 22 19:11:02 Is there a libhybris-dbg package or similar? Apr 22 19:11:22 apt install libhybris-dbgsym libhybris-test-dbgsym Apr 22 19:11:26 just saw it now Apr 22 19:11:33 retrying gdb Apr 22 19:11:55 #5 0xb6ac173e in ?? () Apr 22 19:11:57 same Apr 22 19:41:50 Omer035 was added by: Omer035 Apr 22 20:55:37 @dohniks, I got to connect ssh Apr 22 20:56:02 👍 Apr 22 20:56:08 Using my Linux in text mode Apr 22 20:57:06 What was causing the trouble was the network manager Apr 22 20:57:50 Thank you so much for all your help Apr 22 20:58:04 glad it worked! Apr 22 20:58:07 Actually there are a lot of things to do Apr 22 20:58:17 But I'll rest a bit Apr 22 20:58:48 I didn't even have lunch 😅 Apr 22 20:59:32 bedtime for me. see you around! Apr 22 20:59:48 Good night dude **** ENDING LOGGING AT Mon Apr 23 03:00:12 2018