Posts by mglae

service.mariadb 11.0.0.1 including mysql_upgrade should have been already installed.

Don't run any commands in the background in autostop.sh. Remove ( and )& from your script.

Background execution is used in autostart.sh to not delay the boot process. If used in shut down they are killed after a short time.

For logging enable "Persistent Logging" in the "System" tab of the LE Settings Addon. After reboot use journalctl -b -1 -u kodi-autostart.

Can confirm the issue. Changes to the samba configuration are only applied after reboot.

A work around via ssh is:

systemctl stop smbd
/usr/lib/samba/samba-config
systemctl start smbd

To avoid auto detect you can add KODI_GL_INTERFACE=GLX to /storage/.config/kodi.conf.

There is at least a report of LE11beta1 Generic-legacy starting on Nvidia 9500gt: https://github.com/xbmc/xbmc/issu…ment-1422405042

The used (auto generated) config file is /run/samba/smb.conf.

It is assembled from /storage/.config./samba.conf if existing or else /etc/samba/smb.conf and combined with the settings.

sunlollyking Are new *-FAILED.zip files generated in /storage/logfiles? If yes please post the latest.

You must not rename .kodi_FAILED yourself. The system will boot directly into safe mode again.

Just reboot and safe mode will be disabled.

The Wiki may be wrong. Try renaming kodi.FAILED yourself.

For the records: this is an exceptional case. Kodi is crashing so early that nothing was updated and LE10 is still working. Downgrade is not supported.

Kodi can ever stopped with systemctl stop kodi and no reboot will occur. You only have to type reboot when finished.

First please read https://wiki.libreelec.tv/configuration/safe-mode (especially https://wiki.libreelec.tv/configuration/…gging-or-fixing)

From crashlog:

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x0000000000d2732b in CVideoDatabase::GetUseAllExternalAudioForVideo(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
[Current thread is 1 (Thread 0x7fcf567dc6c0 (LWP 1378))]


2023-04-03 21:18:53.158 T:1312    error <general>: SQL: [MyVideos121.db] SQLite error SQLITE_NOTADB (file is not a database)
                                                   Query: PRAGMA cache_size=4096
                                                    (file is not a database)
2023-04-03 21:18:53.158 T:1312    error <general>: Connect failed with 'SQL: [MyVideos121.db] SQLite error SQLITE_NOTADB (file is not a database)
                                                   Query: PRAGMA cache_size=4096
                                                    (file is not a database)'
2023-04-03 21:18:53.161 T:1312     info <general>: Old database found - updating from version 120 to 121
2023-04-03 21:18:53.161 T:1312    error <general>: SQL: Can't copy database. (26)
2023-04-03 21:18:53.161 T:1312    error <general>: Unable to copy old database MyVideos120 to new version MyVideos121

Start with:

rm /storage/.kodi_FAILED/userdata/Database/MyVideos121.db

If Safe Mode remains do:

rm /storage/.kodi_FAILED/userdata/Database/MyVideos*

No.

Unfortunately there are other reports too that Intel GEN 3 GPUs using Mesa's Gallium i915 driver are crashing.

VLouis You are right. The old Kodi issue of not updating automatic time zone changes in combination of now saving EPG into a database cause the wrong data to be shown even after a Kodi restart.

Der Olli in addition to restarting Kodi you have to reload EPG via Clear Data.

Quote from labzilla

I'm assuming it's supposed to be the /storage/.config directory, the dot is missing.

Yes.

Quote from labzilla

I'm also assuming that anything in that directory that has execute permission will be executed.

Any *.power file is executed.

Is working on my LE11 Generic-legacy machine and logged to the journal:

Boot[1303]: ### running sleep script 20-custom-sleep.sh (pre)... ###
Boot[1307]: ### running custom sleep script /storage/.config/sleep.d/myscript.power (pre)... ###

Sorry, the sample seem to have been deleted.

Create the directory mkdir /storage/.config/sleep.d and the script:

#!/bin/sh
case "$1" in
   pre)
     # <do something on suspend>
     ;;
   post)
     # <do something on resume>
     ;;
esac

Got it

You can try to set up overlayfs on top of /storage.

Correct timezone value Is IMO not required (and we need to patch connman to fix it)

See ./storage/config/sleep.d/99-sample.power.x and create your 01-wake.power script