There have been a couple of reports from users about graphical glitches and random freezes. We have been looking in to the problem and believe that this could be caused by an excess amount of heat generated by the device under certain conditions. We’ve worked on implementing some new thermal management changes. The device should now generate less heat and when it does get a little toasty, should handle things more gracefully.
We would appreciate feedback from affected users and hope this improves things. We would also like to know:
Your climate
Is your device in an AV cupboard?
Which skin are you using?
Do you have the RSS ticker enabled?
I understand some of you aren’t able to keep your device online for long enough to update the software. Fortunately it seems that without Kodi running, stability is not an issue. I would recommend:
Seattle, so its chillier than a good portion of places. The house is kept at about 65F (18C).
No.
OSMC default skin
I didn’t turn it on, so I don’t think so.
Still getting a crash, but it wasn’t the same graphical crash as last time. Just a frown crash, followed by a freeze. Unplug and replug went back to the reboot loop. http://paste.osmc.io/cegocevavi
Thanks for replying and testing so quickly. I can’t see an obvious error for the sad-face loop, but if I had to make a guess (is this a fresh install?), it would be that Kodi’s userdata has become slightly corrupt from a hard freeze and power cycle.
You can try sudo systemctl stop mediacenter, then rm -rf /home/osmc/.kodi && sudo reboot.
Good to know you are not in a warm climate, so we can rule out ambient temps.
Following -> was coming on to report just this problem when I saw this thread:
Your climate: 74F
Is your device in an AV cupboard? Equipment Closet, conditioned to 74F. Does not get above 80*F.
Which skin are you using? OSMC
Do you have the RSS ticker enabled? No
Random freezing, navigating menus, updating, idling, watching dvd.iso via smb, listening to pandora… really anything random. Sometimes right away, sometimes after a day of use.
The freezing results in a frozen pic, whatever audio frequency (if any) was being put out, then the screen begins to grow fuzzy, then it displays a color trip of colors before going completely black.
Seems to be occurring more and more frequently.
I have tried to capture logs, but once it is frozen a hard boot and sometimes a clean install appears to be only solution.
Reboot loop still occurs after a freeze, even with the new install in place.
One of the things I find interesting is that this particular symptom leaves the unit unusable. Wouldn’t their be more reports of failures if this were wide spread? Did something happen with a certain production batch?
let me know if you want me to test anything further, or need more logs.
20-25C ambient temperature most days as it is in a air conditioned room.
Device is located on bench top
Predominately confluence (glitch still present in osmc skin)
No
I followed your instructions precisely. It did not go into the sad face loop first as per usual however still had the graphical glitch and froze after five minutes usage. Reboots resulted in sad face loops. Unfortunately there has be little observable change in the problems with my device. Once it has cooled down, would you like me to capture additional logs?
@sam_nazarko. Just followed this again, exactly. This time did nothing post patch/upgrade instructions other than browse around default menus (not actually selecting anything). No mapping, no add-ons, just pure stock osmc on vero2. After a couple of minutes the screen froze, everything became fuzzy, pixels flaring all over, and despite just navigating around default osmc screen loud screeching sounds across audio.
Took it to mother in laws house, connected to her Sony (we have only Samsung TVs here), her cables, on top of the tv bench (full air exposure), etc… Same exact thing.
Now this happens consistently within a short time period. Anything else you want me to try and grab before I box it up for a month or so?
On the other devices when a log was needed the device was responsive, or at least could be accessed via SSH.
How the heck do I provide a log when the only way to obtain it is to hard boot the device once the problem occurs?
Sam, would you like me to crack the unit and try adding heatsinks to anything? I do not know if this is an issue or not, but willing to try if you want me to
Well if you upload logs it will also upload kodi.old.log which depending how “hard” the crash was will include some info on what might have gone wrong.
But it would definetly help to enable debug logging first (you can disable the onscreen part of debugging).
No, I don’t advise opening the device and trying to add a heatsink. It’s unlikely to be that, particularly if just playing content.
I would suggest you wait for the new images to be made available (couple of hours) and try a fresh one from that. Particularly if you had difficulties updating in the 1st place, as this may have caused some corruption
When the image froze, the Vero2 crashed within about 2 seconds to a full off system. No SSH, no image/sound at all. Solid black. No recognized video by the TV.
Hard booted, uploaded log.
@sam_nazarko
Ok, I will NOT crack open the case then. I will attempt a fresh install tomorrow
2016.04.01 First 2 installs failed and never made it past initial install settings (screen crashing / freezing). Swapped to Vero 1 PSU, used windows installer to prep disc rather than usual linux, swapped to a different Sandisk SD card. First install failed, second succeeded. Manually mapped a single video share (rather than restoring from a backup). Played a different movie, failed within two minutes. First boot back up failed during OSMC splash screen. Third boot up failed playback within 10 minutes.
I have indeed tried the Vero1 PSU Sam with no joy. The current clean install is on the Vero1 PSU and it still glitches up (although the audio pops while navigating the menu are no more!)
When I first received the V2 it did not get past the initial install screen upon the first boot. Attributing this to growing pains I gave it a couple days then installed the updates from the installer… it took a few times locking up along the way before it appeared stable. Over the next two weeks or so we watched a handful of movies – at first we had a couple of lock ups while not in use (I was guessing it to do with the previous CEC implementation)… and then the lockups became more frequent. At this point I came to the forums and found this thread.
Over the last 2 weeks the glitching has become more frequent. About a week ago I thought maybe it had something to do with restoring from a backup, but no joy. Even a stock install with NO restoring from backup, no mappings, etc at all would glitch out.
Installed 4.1 and other than some audio pops everything was again looking really good. Spent a chunk of time restoring from the backup, testing add-ons (pandora, ustvnow, ted), playing music, navigating menus, checking short movie blurbs. All looked OK. Watched the first full movie on the install and a bit over an hour it locked up again.
Now it seams to fail out within a short period of time consistently.
