9.2.2 Discussion

Are those normal (expected)?

[96390.899118@0] DI: DI: tasklet schedule cost 40ms.
[96392.692127@0] DI: DI: tasklet schedule cost 44ms.
[96393.312926@0] DI: DI: tasklet schedule cost 44ms.
[96399.528806@0] DI: DI: tasklet schedule cost 44ms.
[96401.284505@0] DI: DI: tasklet schedule cost 48ms.
[96402.739209@0] DI: DI: tasklet schedule cost 36ms.
[96404.199064@0] DI: DI: tasklet schedule cost 44ms.
[96404.906386@0] DI: DI: tasklet schedule cost 32ms.
[96405.993282@0] DI: DI: tasklet schedule cost 44ms.
[96411.749825@0] DI: DI: tasklet schedule cost 44ms.
[96412.207486@0] DI: DI: tasklet schedule cost 40ms.
[96414.753739@0] DI: DI: tasklet schedule cost 40ms.
[96415.342788@0] DI: DI: tasklet schedule cost 52ms.
[96416.796595@0] DI: DI: tasklet schedule cost 44ms.
[96420.508136@0] DI: DI: tasklet schedule cost 32ms.
[96420.720732@0] DI: DI: tasklet schedule cost 52ms.
[96421.129599@0] DI: DI: tasklet schedule cost 36ms.
[96423.051205@0] DI: DI: tasklet schedule cost 40ms.
[96423.259721@0] DI: DI: tasklet schedule cost 40ms.
[96424.808609@0] DI: DI: tasklet schedule cost 52ms.
[96425.518316@0] DI: DI: tasklet schedule cost 48ms.
[96427.640084@0] DI: DI: tasklet schedule cost 40ms.
[96427.847494@0] DI: DI: tasklet schedule cost 36ms.
[96429.442870@0] DI: DI: tasklet schedule cost 52ms.
[96430.558877@0] DI: DI: tasklet schedule cost 44ms.
[96434.476886@0] DI: DI: tasklet schedule cost 44ms.
[96434.602871@0] DI: DI: tasklet schedule cost 40ms.
[96448.953017@0] DI: DI: tasklet schedule cost 48ms.
[96450.162920@0] DI: DI: tasklet schedule cost 44ms.
[96452.998233@0] DI: DI: tasklet schedule cost 40ms.
[96454.124680@0] DI: DI: tasklet schedule cost 44ms.
[96456.422880@0] DI: DI: tasklet schedule cost 48ms.

I didnt notice this in previous versions, seeing this only on NG/VIM3L… C2 with old kernel doesnt ahve those.

Hi,
thanks but that doesn’t answer the most common question that many users are going to have. Like this: “I have a version X. Do I need to update the Android FW before updating CE to 9.2.2?”

Also it is not clear, if “everything beneath failed” means that you were not able to flash older versions or that you’ve tried e.g. 907P0 but ethernet didn’t work. Or which “rather old firmware” did you actually have before?

You don’t have to know the exact version, just update to the latest android on your GTK. Simple.
My GTK (which is a newer one sent by Beelink after I sent back my previous broken GTK) worked fine with all fw I tried. 907P0 was as good as 909P0.

Is the big problem always keeping the device on the latest software? I always update to the latest firmware and have never had problems, now on 909P0 as well. It doesn’t matter from which version everything works, it’s important that it works well on the latest. After all… if necessary you can always rollback. :slight_smile:

EDIT: I have the same opinion as you, Pelican.

Actually even stranger than that, the device wakes up, plays the channel for a few seconds and then KODI is dead.

Both of my N2s updated without issue. Thanks Devs for all of your hard work and continued awesomeness.

907P0 worked, 909P0 worked but I had to use the USB burning tool 2.2.0 (I did not try 3.x). And as stated by others, I did not care for more experiments as things just worked out for me fine - YMMV.

2 of my devices are no longer starting after the update.
X96 (S905X) and X96 AIR P2 (S905X3) both hang at the splash screen and i cannot ping them. Android on the X96 Air still works, the other device has only coreelec installed.

Update: I think the problem is that starting kodi now takes 10 minutes. I changed the cpu governor to perfomance but it didn’t help.

Update 2: It was docker. I removed all docker addons and now it boots normally.

1 Like

Successfully upgraded my Wetek Hub to 9.2.2, everything looks good with one exception - both ‘Update Channel’ and ‘Available Versions’ are empty, no matter what I do. Here is the relevant (I think so) portion of the log:

2020-04-13 23:41:21.249 T:3743470464   ERROR: ## CoreELEC Addon ## updates::get_json ## ERROR: (ValueError('No JSON object could be decoded',))
2020-04-13 23:41:21.251 T:3743470464   ERROR: Traceback (most recent call last):
                                              File "/home/adamg/CoreELEC/build.CoreELEC-Amlogic.arm-9.2.2/CoreELEC-settings-34e5845c9423a4e86f0edd42ab473c3aadb17561/.install_pkg/usr/share/kodi/addons/service.coreelec.settings/resources/lib/modules/updates.py", line 476, in get_json
                                              File "/usr/lib/python2.7/json/__init__.py", line 339, in loads
                                              File "/usr/lib/python2.7/json/decoder.py", line 364, in decode
                                              File "/usr/lib/python2.7/json/decoder.py", line 382, in raw_decode
                                            ValueError: No JSON object could be decoded

Edit:
After some additional search it seems like that is desired behavior when auto updates are off - correct?
Can I put the regular upgrade URL in the custom field to see versions available on the server?

1 Like

I updated my VIM3 from a without a hitch from the CE settings menu. Afterwards I removed use_rgb_to_yuv=1 from my config.ini and now I experience no more “green screen” issue, even with that setting disabled. Great, thanks for all the hard work guys!

One question though: should I update my VIM3 with the latest Android firmware in order to get the latest U-BOOT from them? Their latest firmware is from 2020-03-19 and I think I installed the one from 2019-11-22 before I replaced Android with CE on the eMMC. Or should I never worry about this when I run CE from eMMC?

I had same problems with a bootloop when my USB HD’s are plugged into the bottom port near the power connector. I spent several hours learning how to pull UART logs and think I have one (below). I captured 3 boot events. It seems like it hangs at a slightly different point during the boot process. I only get a boot loop when the HD is plugged into that one port. I tried a bluetooth adapter in that same slot, and the device booted normally. HD_bootloop.txt (103.3 KB)

Had an issue today. We had a delivery I paused the program and when I came back about 10 minutes later started to play the program and the video ran at erratic speeds. The sound and picture were out of sync it caught up then ran behind again every few seconds. I have not seen it on short pauses and it may have been a one off I just thought I’d mention it if someone else has seen similar.

I was also struggling with Kodi taking 10 mins to start. Thanks to corona insomnia, I’ve read your post earlier than I would normally. It was indeed. With all 3 of my docker addons disabled kodi started normally. I suspected the image updater cause that caused a hang when I tried to reenable it. Had to pull the plug for the nth time recently and on restart the addon updated. Rebooted to check and kodi starts within 30 secs.

So, I am a happy bunny again with 9.2.2. The release wasn’t the train wreck I had started to think of it as (sorry for these thoughts) and the community has been on fine form. I know that the docker image updater won’t have fixed itself so thanks to the unknown people who sorted it.

Great update for me! Thank you!
X96 Max+ (S905X3) CE on SD card, H96 Pro+ (S912) CE on EMMC and X96 mini (S905W) CE on EMMC - updated without any problems via SSH.
PS: Boot splash screen looks awesome!!! Much better then red one.

Thank you for the new release! It is working well for me.

I have an x96 (s905x), I can’t install in emmc (missing the “ceemmc” command), how did you install in emmc?

installtointernal
I had to repeat it more than once. But this is a known issue.
ceemmc is for newer devices.

1 Like

OOO! Thank you very much, I read about this team somewhere, but I forgot.

The pause after resume seems not fixed on my minix neo u9-h, I think it’s shorter then before but still there.
Same symptoms: if you resume within 5 seconds then there is no issue, when you pause for more then 5 seconds the image is paused for about 3/4 seconds, sound is not affected.

Hi I found this in ~/.kodi/temp/kodi.log at startup, in both releases (NG and 1GEN)

2020-04-14 19:59:12.806 T:3805266816 ERROR: ## CoreELEC Addon ## updates::get_json ## ERROR: (ValueError(‘No JSON object could be decoded’,))
2020-04-14 19:59:12.809 T:3805266816 ERROR: Traceback (most recent call last):
File “/home/adamg/CoreELEC/build.CoreELEC-Amlogic-ng.arm-9.2.2/CoreELEC-settings-34e5845c9423a4e86f0edd42ab473c3aadb17561/.install_pkg/usr/share/kodi/addons/service.coreelec.settings/resources/lib/modules/updates.py”, line 476, in get_json
File “/usr/lib/python2.7/json/init.py”, line 339, in loads
File “/usr/lib/python2.7/json/decoder.py”, line 364, in decode
File “/usr/lib/python2.7/json/decoder.py”, line 382, in raw_decode
ValueError: No JSON object could be decoded

And this in 1GEN
2020-04-15 12:15:45.001 T:3855602560 ERROR: ## CoreELEC Addon ## hardware::load_values ## ERROR: (IOError(2, ‘No such file or directory’))
2020-04-15 12:15:45.002 T:3855602560 ERROR: Traceback (most recent call last):
File “/home/adamg/CoreELEC/build.CoreELEC-Amlogic.arm-9.2.2/CoreELEC-settings-34e5845c9423a4e86f0edd42ab473c3aadb17561/.install_pkg/usr/share/kodi/addons/service.coreelec
.settings/resources/lib/modules/hardware.py”, line 440, in load_values
File “/home/adamg/CoreELEC/build.CoreELEC-Amlogic.arm-9.2.2/CoreELEC-settings-34e5845c9423a4e86f0edd42ab473c3aadb17561/.install_pkg/usr/share/kodi/addons/service.coreelec
.settings/oe.py”, line 358, in get_config_ini
IOError: [Errno 2] No such file or directory: ‘/flash/config.ini’

BR N