| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/7ba1eb0b/journalctl.log | 04:49 |
|---|---|---|
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/614fcaf3/findmnt.log | 04:49 |
| T42 | <elros34> @rodrisola there is no single line from systemd in jour journal log, just repeated kernel messages | 11:01 |
| T42 | <elros34> are you sure you have all dynamic partitions mounted? What about /dev/mapper/dynpart-product ? Where are other mount points like /cache and /vendor/firmware_mnt: https://gitlab.com/sailfish-on-courbet/android_device_xiaomi_courbet/-/blob/sfos-18.1-courbet/rootdir/etc/fstab.qcom#L49-L52. You were supposed to mount them too. | 11:06 |
| *** Ischwitch is now known as Ingvix | 11:17 | |
| T42 | <rodrisola> ok @elros34, But the partitions of the modem forward, are not mounted automatically? | 15:54 |
| T42 | <elros34> they should be but they are not if fstab is missing in build output, also you need to mount them in right order and I am not sure you do so that is why I suggested you to mount all of them manually. It's like 2 minutes work, take fstab and mount all partitions: dynamic and regular and move forward because this simple thing takes way too much time | 15:59 |
| T42 | <elros34> If you do this then you need to disable/remove systemd mount units, you can find them in /usr/liib/systemd/system/*.mount | 16:00 |
| T42 | <rodrisola> i have a question | 18:34 |
| T42 | <rodrisola> this error | 18:34 |
| T42 | <rodrisola> https://pastebin.com/EGQwJx1N | 18:34 |
| T42 | <rodrisola> I just changed, and redo the makefs_skip_entry, so that it does not create certain partitions, but I returned it and I still have problems | 18:35 |
| T42 | <elros34> from where this .fstab-decode.swp comes from? It's probably hidden file you created while editing something but no idea what for | 18:43 |
| T42 | <rodrisola> ohh ok!!! | 18:45 |
| T42 | <rodrisola> True, forget, I use the vscodium, with two plugins, and it appears when I create and delete lines, like a git with local commit and then upload it, and create the swp files. I did not see the fatigue | 18:47 |
| T42 | <rodrisola> @elros34 https://pastebin.com/JvWWRWpx | 20:37 |
| T42 | <rodrisola> Oct 27 19:49:14 mi11lite systemd[1]: selinux: Unknown class system | 20:39 |
| Thaodan | rodrisola: That's harmless, that comes from android selinux policies I think. | 20:50 |
| T42 | <rodrisola> sure, but in bootpara ... selinux = 1 and android.selinux = permisive, as indicated by the hadk-faq, I already have all the partitions started, can you help me thaodan, what can I send you | 20:52 |
| T42 | <rodrisola> update 30-12-2021 | 21:03 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/87b03ede/journalctl.log | 21:03 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/614fcaf3/findmnt.log | 21:03 |
| T42 | <elros34> where do you mount partitions if not in /usr/bin/droid/droid-hal-early-init.sh? | 21:08 |
| T42 | <elros34> oh I see way too late | 21:20 |
| T42 | <rodrisola> in the mount-android-dynpart.service | 21:20 |
| T42 | <elros34> yeah told you to use droid-hal-early-init.sh instead | 21:21 |
| T42 | <rodrisola> but how can I build, droid-hal-early-init.sh, from code, or do you tell me to edit it when it is already running? | 21:24 |
| T42 | <elros34> what build? just rename and move your script which you currently use to /usr/bin/droid/droid-hal-early-init.sh, thats all | 21:27 |
| T42 | <rodrisola> ok, I did it but it does not mount, it does not mount the partitions | 21:29 |
| T42 | <elros34> so show your changes on device and fresh journal | 21:31 |
| T42 | <rodrisola> https://www.toptal.com/developers/hastebin/gepoligude.sql | 21:35 |
| T42 | <rodrisola> finmnt https://www.toptal.com/developers/hastebin/pigapiqova.typescript | 21:35 |
| T42 | <rodrisola> not mount ssystem | 21:35 |
| T42 | <elros34> so what about journal I asked? | 21:38 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/28eaae21/journalctl.log | 21:39 |
| T42 | <elros34> did you read it before posting? Reboot device and try again and make sure it really contains systemd logs | 21:42 |
| T42 | <rodrisola> I restarted it 2 times, and that is what I get on the screen when I run journalctl --no-pager | 21:43 |
| T42 | <rodrisola> I restarted it 2 times, and that's what it shows me on the screen when I run journalctl --no-pager | 21:44 |
| T42 | <elros34> then enable persisten logging (hadk-faq), reboot and use also -b0 argument | 21:45 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/75940419/journalctl.log | 21:46 |
| T42 | <elros34> so you did not add -b0... | 21:51 |
| T42 | <elros34> partitions seems to be mounted, do you have system_ext directory in /system/? | 21:55 |
| T42 | <elros34> try also strace -e file -f /sbin/droid-hal-init | 21:56 |
| Thaodan | You can see most of the partition layout already in the root folder in out/target/product/$device/root | 21:59 |
| Thaodan | if you have a real system ext etc. | 21:59 |
| Thaodan | Not all partitions are necessary required to be mounted e.g. product is mostly unused or system_ext too. | 22:01 |
| Thaodan | @rodrisola: You have still the copies of the mount units from droid-hal e.g. system.mount | 22:04 |
| Thaodan | This is the only thing bugging me right now: system-2.mount: Cannot add dependency job, ignoring: Unit system-2.mount is not loaded properly: Invalid argument. I would cat the mount unit that fails (systemctl cat $unit) and see what's wrong. | 22:05 |
| Thaodan | Either get mounting of those dynamic partitions working or package droid-system like for sony ports. | 22:05 |
| T42 | <rodrisola> thaodan, system-2.mount , I no longer use them, I will remove them in the next build, they no longer point to anything. and if I had them mounted, but elros34 tells me that I should start them in droid-hal-early-init | 22:13 |
| T42 | <elros34> you can remove them right now on device but this is not really relevant, droid-hal-init fails with no such file or directory so clearly you are missing some files, strace might help | 22:15 |
| T42 | <elros34> what about strace? | 22:22 |
| T42 | <rodrisola> / # strace -e file -f /sbin/droid-hal-init | 22:23 |
| T42 | <rodrisola> execve("/sbin/droid-hal-init", ["/sbin/droid-hal-init"], 0x7fe4edfae8 /* 7 vars */) = -1 ENOENT (No such file or directory) | 22:23 |
| T42 | <rodrisola> strace: exec: No such file or directory | 22:23 |
| T42 | <rodrisola> +++ exited with 1 +++ | 22:23 |
| T42 | <elros34> ok so try without -e file | 22:24 |
| T42 | <rodrisola> / # strace -f /sbin/droid-hal-init | 22:25 |
| T42 | <rodrisola> execve("/sbin/droid-hal-init", ["/sbin/droid-hal-init"], 0x7fdb688f98 /* 7 vars */) = -1 ENOENT (No such file or directory) | 22:25 |
| T42 | <rodrisola> strace: exec: No such file or directory | 22:25 |
| T42 | <rodrisola> +++ exited with 1 +++ | 22:26 |
| T42 | <elros34> hm something is missing on android side | 22:28 |
| T42 | <elros34> do you have up to date droid-config uploaded? | 22:29 |
| T42 | <rodrisola> The system is not mounted, since droid-hal-early does not mount them | 22:29 |
| Thaodan | elros34: See the mce fail, his system partition is still not mounted. | 22:30 |
| Thaodan | eh | 22:30 |
| Thaodan | & | 22:30 |
| Thaodan | ^ was about to write that | 22:30 |
| Thaodan | either dynmount working or go droid-system way | 22:31 |
| T42 | <elros34> how? I can see in logs that at least /system_root is mounted | 22:31 |
| T42 | <elros34> and again please use -b0 argument because your journal is confusing | 22:35 |
| T42 | <rodrisola> after exec the phone is reboot in fastboot | 22:36 |
| T42 | <rodrisola> https://pastebin.com/ | 22:36 |
| T42 | <elros34> in that case journal from previous boot will be in -b-1 | 22:38 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/b422da88/strace.log | 22:39 |
| T42 | <elros34> so droid-hal-init works now, what was wrong? | 22:40 |
| T42 | <elros34> this line should be before "%include droid-configs-device/droid-configs.inc": https://gitlab.com/sailfish-on-courbet/droid-config-courbet/-/blob/new2/rpm/droid-config-courbet.spec#L25 | 22:41 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/75940419/journalctl.log | 22:50 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/a4f3b0c3/journal.log | 22:52 |
| T42 | <rodrisola> https://irc.thaodan.de/.imgstore/3866f3e8/journalctl_b_1.log | 22:52 |
| T42 | <rodrisola> Warning[12/30 22:54:03] : file /etc/ofono/ril_subscription.conf conflicts between attempted installs of droid-config-courbet-1-202112302245.aarch64 and ofono-configs-mer-1.23+git35-1.10.1.jolla.aarch64 | 22:59 |
| T42 | <rodrisola> only chage position of the line https://gitlab.com/sailfish-on-courbet/droid-config-courbet/-/blob/new2/rpm/droid-config-courbet.spec#L25 | 22:59 |
| Thaodan | whats the contents of system root? | 23:06 |
| Thaodan | I think system_root should be / | 23:07 |
| Thaodan | that's why the contents of / besides /system /vendor /odm should go to droid-hal | 23:08 |
| T42 | <elros34> he managed to get droid-hal-init working but without explanation how | 23:10 |
| Thaodan | His device has this set probably https://source.android.com/devices/bootloader/partitions/system-as-root | 23:11 |
| Thaodan | It looks like this in PlatformConfig.mk: https://github.com/sonyxperiadev/device-sony-kumano/blob/master/PlatformConfig.mk#L52 | 23:11 |
| Thaodan | When system as root you need to either just package the /system folder in the partiton or if you use that partion image directly mount it under /system_root and symlink /system_root/system -> /system | 23:13 |
| T42 | <elros34> so similar to what he is doing: https://www.toptal.com/developers/hastebin/gepoligude.sql. Except he bind mount, I hope he do not bind mount it on symlink instead directory | 23:14 |
| Thaodan | I have a workaround in my sailfish/hadk build scripts for that case too: https://gitlab.com/Thaodan/hadk_tools/-/blob/master/templates/mer.droid.syspart.hadk#L78 | 23:14 |
| Thaodan | elros34: Bind mount is probably even better | 23:15 |
| T42 | <rodrisola> Not , I mount in system directory (re @elros34: so similar to what h...) | 23:20 |
| Thaodan | He mean this mount -o bind /system_root/system /system | 23:21 |
| Thaodan | Try to look that droid-hal contains all the files in /system_root except those mentioned above | 23:22 |
| Thaodan | e.g. like this https://paste.opensuse.org/63057293 | 23:25 |
| Thaodan | detrius then contains the rest e.g. symlinks | 23:26 |
| Thaodan | https://paste.opensuse.org/76768384 | 23:26 |
| T42 | <rodrisola> / # findmnt | 23:27 |
| T42 | <rodrisola> TARGET SOURCE FSTYPE OPTIONS | 23:27 |
| T42 | <rodrisola> / /dev/sda34[/.stowaways/sailfishos] | 23:27 |
| T42 | <rodrisola> | ext4 rw,noatime,data=ordered | 23:27 |
| T42 | <rodrisola> |-/data /dev/sda34 ext4 rw,noatime,data=ordered | 23:27 |
| T42 | <rodrisola> |-/dev devtmpfs devtmpf rw,nosuid,noatime,size=26754 | 23:27 |
| T42 | <rodrisola> | |-/dev/pts devpts devpts rw,noatime,gid=5,mode=620,pt | 23:27 |
| T42 | <rodrisola> | |-/dev/shm tmpfs tmpfs rw,nosuid,nodev,noexec | 23:27 |
| T42 | <rodrisola> | `-/dev/binderfs binder binder rw,noatime,max=1048576,stats | 23:27 |
| T42 | <rodrisola> |-/sys sysfs sysfs rw,noatime | 23:28 |
| T42 | <rodrisola> | |-/sys/fs/selinux selinuxfs selinux rw,nosuid,noexec,noatime | 23:28 |
| T42 | <rodrisola> | |-/sys/fs/cgroup tmpfs tmpfs ro,nosuid,nodev,noexec,mode= | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/unified cgroup2 cgroup2 rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/systemd cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/cpu,cpuacct cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/memory cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/cpuset cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/schedtune cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | |-/sys/fs/cgroup/freezer cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | | `-/sys/fs/cgroup/blkio cgroup cgroup rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | |-/sys/fs/pstore pstore pstore rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | |-/sys/fs/bpf bpf bpf rw,nosuid,nodev,noexec,noati | 23:28 |
| T42 | <rodrisola> | |-/sys/kernel/debug debugfs debugfs rw,noatime | 23:28 |
| T42 | <rodrisola> | |-/sys/kernel/config configfs configf rw,noatime | 23:28 |
| T42 | <rodrisola> | `-/sys/fs/fuse/connections fusectl fusectl rw,noatime | 23:28 |
| T42 | <rodrisola> |-/proc proc proc rw,noatime | 23:28 |
| T42 | <rodrisola> |-/config none configf rw,noatime | 23:28 |
| T42 | <rodrisola> |-/run tmpfs tmpfs rw,nosuid,nodev,mode=755 | 23:28 |
| T42 | <rodrisola> | `-/run/user/100000 tmpfs tmpfs rw,nosuid,nodev,noatime,size | 23:28 |
| T42 | <rodrisola> |-/tmp tmpfs tmpfs rw,nosuid,nodev,noatime | 23:28 |
| T42 | <rodrisola> |-/linkerconfig /dev/sda34[/.stowaways/sailfishos/linkerconfig/bootstrap] | 23:28 |
| T42 | <rodrisola> https://pastebin.com/LTvi03mu | 23:29 |
| T42 | <rodrisola> i have this error | 23:41 |
| T42 | <rodrisola> https://pastebin.com/nthXrN2b | 23:41 |
| T42 | <elros34> you can find solution in hadk-hot | 23:43 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!