| T42 | <HengYeDev> ok i got boot but no gui | 01:44 |
|---|---|---|
| T42 | <HengYeDev> i have telnet but there is no defaultuser so how do i debug gui | 01:44 |
| T42 | <vsvoid> @elros34 I have managed to downgrade droid-hal submodule and packaged parse-android-dynparts, managed to get dynamic parts to mount using systemd services. But am still having the reboot issue | 10:27 |
| T42 | <elros34> with same error? | 10:27 |
| T42 | <vsvoid> No this time I see less errors, let me send journalct log | 10:28 |
| T42 | <vsvoid> https://gist.github.com/kamstartech/eac0a72cca8b357533eafd3497fdf297 | 10:34 |
| T42 | <elros34> how excatly did you mount partitions if not by script mentioned in hadk-hot? | 10:37 |
| T42 | <vsvoid> Am using script in `hybris/droid-configs/droid-configs-device/sparse/usr/sbin/mount-android-dynpart.sh` | 10:40 |
| T42 | <vsvoid> Then have a service that runs it | 10:40 |
| T42 | <vsvoid> `hybris/droid-configs/droid-configs-device/sparse/usr/lib/systemd/system/mount-android-dynamic-part.service` | 10:41 |
| T42 | <elros34> when it runs it? clearly too late. Also I see 2 logs there, I am not sure which one is correct | 10:41 |
| T42 | <vsvoid> Where would be appropriete time for it to run? | 10:43 |
| T42 | <vsvoid> The second log is the correct one | 10:43 |
| T42 | <elros34> as early as possible now just follow instruction and put it in /init-debug. Next time try to generate clear logs so there is no need to guess. Also get rid of this audit spam | 10:46 |
| T42 | <vsvoid> Okay | 10:46 |
| T42 | <elros34> probably you need to get rid of --hwc-restart, there should be some commit about it in latest droid-config | 10:49 |
| T42 | <vsvoid> Let me look into that | 10:50 |
| T42 | <edp_17> Should I still use droidmedia 0.20211101.0 for building sfos 4.5? | 12:12 |
| mal | 4.5.0 uses 0.20220929.0 in official devices | 12:21 |
| T42 | <edp_17> Thanks mal. I'll try the older one and if doesn't work, will check out this newer one. | 12:25 |
| mal | I have 0.20211101.0 on fp2 and it works fine | 12:39 |
| T42 | <edp_17> Thanks for double-checking. | 12:42 |
| T42 | <edp_17> When running the MIC on host in Platform, I am getting "Requires [rpmlib(PayloadIsZstd)-5.4.18-1], which is not provided" warning for every package, and the build fails. I've spotted that it is using an older version of "mic 1.0.9 (Sailfish OS 4.3.0.15 (Suomenlinna))". Is this the problem? If yes, how can I update the mic? If, not, how can I resolve this? full log: https://paste.ubuntu.com/p/gFPFHQCkcb/ | 13:19 |
| mal | did you update tooling? | 13:24 |
| mal | and sdk | 13:24 |
| mal | based on that version in that message you didn't update | 13:25 |
| mal | @edp_17 try "sdk-foreach-su -ly ssu re 4.5.0.18", "sdk-foreach-su -ly zypper ref" and then "sdk-foreach-su -ly zypper dup" | 13:32 |
| T42 | <edp_17> I've removed the old ones and re-created the new. These are the command I used: https://paste.ubuntu.com/p/fdXH7GpNR6/ | 13:32 |
| mal | ok, I have used same tooling and targets for years | 13:33 |
| mal | I just never named tooling with any version in the name | 13:33 |
| mal | my tooling is just "SailfishOS-latest" | 13:35 |
| T42 | <edp_17> For me it is just easier to know whether updated or not. However, it seems there was a problem with that because it's created the tooling with the correct name "SailfishOS-4.5" but its content is an older version. | 13:37 |
| mal | I think name doesn't matter at all | 13:37 |
| T42 | <edp_17> I know. What I was trying to say the "https://releases.sailfishos.org/sdk/targets/Sailfish_OS-4.5.0.18-Sailfish_SDK_Tooling-i486.tar.7z" has installed 4.3 instead of 4.5. | 13:38 |
| T42 | <edp_17> I followed what elros suggested and got this: https://paste.ubuntu.com/p/VDPYDYdH3j/ | 13:39 |
| T42 | <edp_17> This also failed, so I am not sure how to proceed. I'll try to re-download tooling again to see what version I'll get. | 13:40 |
| mal | @edp_17 Executing locally.. meaning the sdk | 13:43 |
| mal | the tooling based on that output is 4.5.0.18 | 13:43 |
| T42 | <edp_17> Okay, so the toolings are fine, just the sdk is old. Got it. | 13:44 |
| mal | try running "rpm --rebuilddb" | 13:45 |
| mal | @edp_17 wait, I think the issue is that you try to update from 4.3.0 to 4.5.0, run "ssu re 4.4.0.72" then "zypper ref" and "zypper dup" | 13:46 |
| mal | @edp_17 actually add sudo before those commands | 13:47 |
| T42 | <edp_17> Thanks mal. I've just replaced the version number to 4.4.0.72 in command elros posted and it's running. I hope this will upgrade sdk to 4.4, then I'll update it to 4.5. | 13:52 |
| T42 | <edp_17> I guess the droid-hal packages I've built today can go to bin and that was built against 4.3? Or no? | 13:53 |
| mal | what command? | 13:53 |
| mal | it should work | 13:53 |
| T42 | <edp_17> These: "sdk-foreach-su -ly ssu re 4.5.0.18", "sdk-foreach-su -ly zypper ref" and then "sdk-foreach-su -ly zypper dup" | 13:53 |
| T42 | <edp_17> I've used: "sdk-foreach-su -ly ssu re 4.4.0.72", "sdk-foreach-su -ly zypper ref" and then "sdk-foreach-su -ly zypper dup" | 13:53 |
| mal | @edp_17 you should only run the commands I gave | 13:53 |
| mal | you only want to update sdk | 13:53 |
| mal | you DO NOT want to try to downgrade tooling and target | 13:54 |
| mal | @edp_17 only run "sudo ssu re 4.4.0.72" then "sudo zypper ref" and "sudo zypper dup" | 13:54 |
| T42 | <edp_17> Okay. I got "Error message: Could not resolve host: releases.jolla.com" anyway. | 13:54 |
| mal | you need to exit sdk and re-enter before you can use network in sdk after you have updated it | 13:55 |
| mal | so once you have updated sdk to 4.4.0.72 then run "sudo ssu re 4.5.0.18" then "sudo zypper ref" and "sudo zypper dup" to update it to latest version | 13:55 |
| mal | after that exit sdk and re-enter and all should work | 13:56 |
| T42 | <edp_17> Well, I messed up because now I am getting this: https://paste.ubuntu.com/p/66hsXc83Dw/ | 13:56 |
| T42 | <edp_17> Better to re-create sdk from scratch. | 13:56 |
| mal | exiting sdk didn't help? | 13:58 |
| T42 | <edp_17> Didn't tried because re-created it sooner. Now I have 4.5 sdk and toolings. | 14:00 |
| T42 | <edp_17> The question is should I rebuild droid-hal packages with sdk 4.5? | 14:01 |
| mal | like I said version built with older version still works fine, but repackaging it just "build_packages.sh -d" so it's not that big deal to rebuild it | 14:02 |
| mal | do note that sdk, tooling or target has nothing to do with android side build i.e. hybris-hal build | 14:03 |
| T42 | <edp_17> Yeah, but I already uploaded that to OBS and rebuilt all packages there. | 14:03 |
| mal | just curious how do you upload the rpms to obs? | 14:03 |
| T42 | <edp_17> Manually, one by one on the website. Why? | 14:04 |
| mal | so you do it the hard way :) | 14:04 |
| mal | using osc makes it much easier | 14:04 |
| T42 | <edp_17> Yes, I always do it the hard way. :D I remember, tend to build all packages locally and built a zip file every time I changed something. I am getting lazy these days. :D | 14:07 |
| T42 | <edp_17> I only used osc to copy packages from a repo to another on OBS. | 14:07 |
| mal | read about "osc co" which is used to make a local copy of a obs project or package and also "osc ci" used to upload local changes to obs | 14:09 |
| mal | which means after you have a local copy, you just remove the old rpms from the local copy, add the new rpms the to folder and then "osc ar" to add/remove local files and finally "osc ci" to commit the change and upload it to obs | 14:10 |
| mal | ah, sorry, you should do "osc ar" once after removing the files and a second time after copying to new files so it will notice any changes to files which have same name | 14:11 |
| mal | like droidmedia rpms | 14:12 |
| mal | maybe I should write instructions to faq | 14:12 |
| T42 | <edp_17> Thanks, I'll do check that. A short how-to guide would be very helpful indeed. | 14:13 |
| Aboothahir[m] | mal: account created in obs for me? | 14:38 |
| mal | Aboothahir[m]: lbt handles creating obs accounts | 14:44 |
| Aboothahir[m] | lbt is not responding now | 14:44 |
| T42 | <elros34> did you send him all needed information? This is how it works | 14:47 |
| T42 | <A_T_R> Yea | 14:47 |
| T42 | <A_T_R> Hey - we're having some issues at the moment - I'll get this done ASAP. ping me or mal if it's not done later this week. Sorry for the delay. | 14:47 |
| T42 | <A_T_R> | 14:47 |
| T42 | <A_T_R> Reply from lbt | 14:47 |
| T42 | <vsvoid> @elros34 | 16:16 |
| T42 | <vsvoid> | 16:16 |
| T42 | <vsvoid> https://gist.github.com/kamstartech/7bc888a7e6bd9c7520abe5b7ccea2ba5 | 16:16 |
| T42 | <vsvoid> | 16:16 |
| T42 | <vsvoid> Apart from --hwc-restart what else am I missing out | 16:16 |
| T42 | <elros34> first read section about getting logs from hadk-hot then get full journal (see you do not have even compositor there) and logcat | 16:23 |
| HengYeDev[m] | what service is responsible for creating defaultuser? | 19:25 |
| T42 | <elros34> what do you need it for? Not sure which service but script is in /usr/lib/startup/start-autologin | 19:32 |
| HengYeDev[m] | i have no defaultuser user right now so lipstick can't start | 19:34 |
| HengYeDev[m] | i.e. passwd defaultuser: passwd: Unknown user name 'defaultuser'. | 19:34 |
| T42 | <elros34> there must be something in journal | 19:35 |
| T42 | <elros34> didn't you mask some services or made some unusual changes? | 19:36 |
| HengYeDev[m] | not really, just swapped out droid-hal-init with the one made by piggz, which seems to work fine (the container is running) | 19:37 |
| HengYeDev[m] | what should i look for in journal for which process? | 19:37 |
| T42 | <elros34> any failure for systemd services, especially *user*.service | 19:38 |
| HengYeDev[m] | we have 4 services failed... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/lpZCUibXtLsxAXzvWJEyuYNk>) | 19:40 |
| HengYeDev[m] | anything important? | 19:40 |
| T42 | <elros34> I do not think so, what about start-user-session? | 19:41 |
| HengYeDev[m] | disabled | 19:41 |
| T42 | <elros34> why? | 19:41 |
| HengYeDev[m] | Loaded: loaded (/usr/lib/systemd/system/start-user-session.service; disabled; vendor preset: enabled) | 19:42 |
| HengYeDev[m] | inactive dead | 19:42 |
| HengYeDev[m] | does that mean its disabled? | 19:42 |
| T42 | <elros34> I do not think so, what is the exit status for 2 scripts started? | 19:43 |
| HengYeDev[m] | there are no logs. its inactive(dead) | 19:43 |
| HengYeDev[m] | persistent journalctl is on | 19:43 |
| T42 | <elros34> systemctl status start-user-session | 19:44 |
| HengYeDev[m] | # systemctl status start-user-session... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/xDnBryZGAWDySbFMuNXQPkXP>) | 19:44 |
| HengYeDev[m] | then all empty | 19:44 |
| T42 | <elros34> I am not sure why you don't have more info, how about you just post full journal, maybe even from first boot | 19:45 |
| HengYeDev[m] | this is big | 19:49 |
| HengYeDev[m] | https://pastebin.com/zp2r8WX8 | 19:49 |
| T42 | <elros34> what telnet port is that? | 19:51 |
| T42 | <elros34> did you reboot before getting this logs? | 19:51 |
| HengYeDev[m] | 2323 | 19:52 |
| HengYeDev[m] | no i didnt reboot | 19:52 |
| T42 | <elros34> so read hadk-hot about generating proper logs, this one doesn't even have systemd initialization | 19:52 |
| HengYeDev[m] | ok i now have https://pastebin.com/eBJPzKjv | 20:07 |
| T42 | <elros34> what is the output of findmnt? Do you use linux terminal or putty to telnet? | 20:09 |
| HengYeDev[m] | i am in linux terminal and i tee it into a file | 20:10 |
| HengYeDev[m] | here i have findmnt https://pastebin.com/cb8nYUAj | 20:11 |
| T42 | <elros34> I am asking because of broken formating, whey do you put it to file? Do you have these options https://github.com/mer-hybris/mer-kernel-check/blob/master/mer_verify_kernel_config#L330-L333 | 20:11 |
| HengYeDev[m] | its worked around with lxc config by notkit | 20:13 |
| HengYeDev[m] | i tee to file since i have no ssh | 20:13 |
| HengYeDev[m] | well i have it but no defaultuser to login with | 20:14 |
| HengYeDev[m] | hrmm, the lxc config wont work the firejail tho | 20:14 |
| HengYeDev[m] | im operating without kernel source, using patches from https://gitlab.com/ubports/porting/community-ports/android10/unihertz-titan/unihertz-titan | 20:15 |
| T42 | <elros34> I do not get how using telnet instead ssh change anything and you need to tee something. It's just hard to read these logs: [0;1;39m and I wonder what you are doing | 20:16 |
| T42 | <elros34> about kernel configs just check whether they are enabled in /proc/config.gz | 20:17 |
| HengYeDev[m] | they are not enabled, but we cannot change them for this device (see ubports) | 20:18 |
| HengYeDev[m] | ssh gives the error of System is booting up. Unprivileged users are not permitted to log in yet. Please come back later. For technical details, see pam_nologin(8)." | 20:18 |
| HengYeDev[m] | when login with root and even after bootup. | 20:18 |
| T42 | <elros34> I am not sure but these configs might be needed for systemd | 20:20 |
| HengYeDev[m] | they arent' in https://github.com/systemd/systemd/blob/main/README | 20:24 |
| T42 | <elros34> I do not see user session in your logs, not to mention lipstick but you said that lipstick can't be started | 20:28 |
| T42 | <elros34> ah you didn't use correct command so it just 10s of logs, reboot and use journalctl --no-tail -f | 20:30 |
| HengYeDev[m] | new paste: https://pastebin.com/dhU4xCu6 | 23:03 |
| HengYeDev[m] | the issue is that even start-user-session isnt started so no defaultuser = no lipstick | 23:03 |
| T42 | <elros34> this is wast of time if you do not even read logs you post, see there is literally 2 lines from systemd in that log. How do you want to debug why user service is not started with just some kernel noise | 23:06 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!