• 1
  • 2
  • 3(current)
  • 4
  • 5
  • 15
Bug Kernal killing xbmc task on 256 board
#31
(2012-12-12, 07:16)lowridin_guy Wrote: Here is another log from when XBMC crashed tonight while streaming a higher quality SD movie....256MB Pi with OE Beta 4.

http://pastebin.com/Z8F6GcFH

Useful log. Looks like a problem in 1channel.

Code:
22:22:33 T:2791306336   ERROR: EXCEPTION Thrown (PythonToCppException) : -->Python callback/script returned the following error<--
                                             - NOTE: IGNORING THIS CAN LEAD TO MEMORY LEAKS!
                                            Error Type: <type 'exceptions.UnboundLocalError'>
                                            Error Contents: local variable 'sql_update' referenced before assignment
                                            Traceback (most recent call last):
                                              File "/storage/.xbmc/addons/plugin.video.1channel/playback.py", line 136, in onPlayBackEnded
                                                self.onPlayBackStopped()
                                              File "/storage/.xbmc/addons/plugin.video.1channel/playback.py", line 109, in onPlayBackStopped
                                                self.ChangeWatched(self.imdbnum, self.video_type, self.title, self.season, self.episode, self.year, watched=7)
                                              File "/storage/.xbmc/addons/plugin.video.1channel/playback.py", line 152, in ChangeWatched
                                                metaget.change_watched(video_type, name, imdb_id, season=season, episode=episode, year=year, watched=watched)
                                              File "/storage/.xbmc/addons/script.module.metahandler/lib/metahandler/metahandlers.py", line 1751, in change_watched
                                                self._update_watched(imdb_id, type, watched, name=name, season=season, episode=episode, tvdb_id=tvdb_id)
                                              File "/storage/.xbmc/addons/script.module.metahandler/lib/metahandler/metahandlers.py", line 1793, in _update_watched
                                                print 'SQL UPDATE: %s' % sql_update
                                            UnboundLocalError: local variable 'sql_update' referenced before assignment
                                            -->End of Python script error report<--
22:22:33 T:2791306336  NOTICE: 1Channel:

repeated several times.
Reply
#32
Continuing on with my test, I have found this version of open elec to be stable OpenELEC Version: -r11930, Have been running this version for 24 hours, no reboots at all, am about to start upgrading untill I find a version that causes the reboots while watching hd media
to speed my testing of these versions up, i ma running 3 pi`s and have made it up to version r12029 with no reboots watching HD media, all with audio passthrough on
Reply
#33
HI Popcornmix,

You are correct that these script errors were in 1Channel but further down in the log you will see that I connected to a .mp4 file in a different plugin and that is where it actually locked up after about 20 mins into the video.

Thanks,
Reply
#34
(2012-12-12, 14:18)lowridin_guy Wrote: HI Popcornmix,

You are correct that these script errors were in 1Channel but further down in the log you will see that I connected to a .mp4 file in a different plugin and that is where it actually locked up after about 20 mins into the video.

Thanks,
Okay. It may or may not be related to the 1channel python errors earlier. Can you connect to the mp4 site after a reboot, and see if it still fails?
Reply
#35
I can check tonight and try to produce a log. I do know that it has happened to me before without using 1Channel prior so I personally don't think the issues are related.
Reply
#36
Just an update, i found it to be introduced between openelec version r12238 and r12320, thestateofme is offline at the min, so i cant get the versions between those dates at the moment, as soon as i do i will post back here
Reply
#37
Ok guys, I have found the version where this bug first came in, every time i ran the test, it was with the same movie, the same xbmc settings, all versions of OE worked up untill r12280, thats when they first swapped to kernel 3.6.3, now i cant tell if its something to do with the kernel or changes to xbmc on tat date, anymore testing required let me know, As i said at the start this also affected RaspBMC,
Reply
#38
(2012-12-13, 13:09)jas0npc Wrote: Ok guys, I have found the version where this bug first came in, every time i ran the test, it was with the same movie, the same xbmc settings, all versions of OE worked up untill r12280, thats when they first swapped to kernel 3.6.3, now i cant tell if its something to do with the kernel or changes to xbmc on tat date, anymore testing required let me know, As i said at the start this also affected RaspBMC,

Thanks for testing. It should be possible to start with a latest Raspbmc install and run
Code:
sudo rpi-update 3851111657474038659_4d5a97d9e02e899c63090
(remove the _ from the hash - this forum converts the hash into a link without it)
I think that will work and get you back to 3.2 kernel and firmware. I'd be interested if that fixes things.
Reply
#39
Yep no probs, Will do that now and get back to you
Reply
#40
Great job jas0npc. Must have been quite tedious. I hope you guys can find out what happend. (:
Reply
#41
Done that , but i get sudo: rpi-update: command not found
before i tried to downgrade , i tried again the fresh install and that also caused xbmc to restart
Reply
#42
Did you use raspbmc? Maybe you got locked into a openelec mind cycle with all that OE testing. Wink
Know it sounds stupid but don't take it as an insult. Just trying to throw out some ideas.
Or maybe Sam removed the command so it's only present in raspbian.
Reply
#43
Lol yes did a fresh install this morning after finding where in OE the bug made its first apperanceSmile
then tried to watch the same movie, in freshly installed and update raspbmc, xbmc restarted.
then ssh pi raspberry and then copied and pasted
Code:
sudo rpi-update 38511116574740386594d5a97d9e02e899c63090
hit enter Smile
so stuck at the min

well the command i was given earlier minus the _
Reply
#44
(2012-12-13, 15:02)jas0npc Wrote: Lol yes did a fresh install this morning after finding where in OE the bug made its first apperanceSmile
then tried to watch the same movie, in freshly installed and update raspbmc, xbmc restarted.
then ssh pi raspberry and then copied and pasted
Code:
sudo rpi-update 38511116574740386594d5a97d9e02e899c63090
hit enter Smile
so stuck at the min

well the command i was given earlier minus the _

Try (from here: https://github.com/Hexxeh/rpi-update)
sudo apt-get install -y git-core ca-certificates
sudo wget http://goo.gl/1BOfJ -O /usr/bin/rpi-update && sudo chmod +x /usr/bin/rpi-update

then the rpi-update <hash> command.
Reply
#45
(2012-12-13, 14:19)tuxen Wrote: Great job jas0npc. Must have been quite tedious. I hope you guys can find out what happend. (:

It was, but he end redsult will benifit all 256 boards:_

Ok this is my latest error Smile

Code:
pi@raspbmc:~$ sudo rpi-update 38511116574740386594d5a97d9e02e899c63090
Raspberry Pi firmware updater by Hexxeh, enhanced by AndrewS
Performing self-update
ARM/GPU split is now defined in /boot/config.txt using the gpu_mem option!
Downloading specific firmware revision (this will take a few minutes)
Using HardFP libraries
fatal: Not a git repository: '//root/.rpi-firmware/.git'
pi@raspbmc:~$
Reply
  • 1
  • 2
  • 3(current)
  • 4
  • 5
  • 15

Logout Mark Read Team Forum Stats Members Help
Kernal killing xbmc task on 256 board0