My Vero 2 constantly crashes while in Kodi. Each time I enumerate the list of repository add-ons the system crashes.
I experience the same issues when Kodi is stopped. I have attempted to run a CPU diagnostic cmd and on testing CPU 3 it locks up. CPU 1 & 2 test ok.
Attached are the most recent uploaded logs.
Thanks,
Paul
Hi Paul
I received your email about this a few days ago and made some suggestions.
I don’t believe I received a reply, but if I did, please send it through again.
We’re keen on getting you up and running as soon as possible.
Unfortunately I couldn’t discern anything obvious from your logs. Could you follow my advice in my email (if you have received it) re. what diagnostics you have run?
Sam
Hi Sam.
Apologies, your reply was snagged in my junk mail. I used the stress cmd/sw (stress --cpu 3 --io 4 --vm 2 --vm-bytes 128M --timeout 10s).
I will try to snag a pix when the crash is encountered.
Not that familiar with that utility.
But a photo and check of temps (even with a physical judgement) will be good. Remember some things like cpuburn are meant to stress things to their maximum.
If you have external peripherals attached, in particular USB drives, I’d suggest removing those before proceeding.
Sam
Do I not have any peripherals attached. I am unable to repeat the process that generates error messages, my system just locks up.
I did install lm-sensors and got this reading, the peak value, just before the system rebooted.
osmc@osmc:~$ sensors
aml_thermal-virtual-0
Adapter: Virtual device
temp1: +62.6°C (crit = +60.0°C)
Latest Set of logs
http://paste.osmc.io/beyubusehi
Seeing several addons in your log I would recommend renaming /home/osmc/.kodi and rebooting, this will reset Kodi to defaults. Set up only what needs to be set up and then test if you still have the problem
Thanks for the suggestion, but I end up with the same issues. Although, I’m able to do things (scrolling around) for longer, but it still doesn’t help. I did not install anything after renaming the .kodi directory.
I’ve scaled back the CPU frequency to 1000Mhz for all the cores and the system is stable. I will continue to test, but I think I may have a hardware problem, probably CPU related.
You need to be careful. Manually adjusting the CPU frequency can reduce the life of the device and even damage it if this is done in the long term. In some circumstances you may be overriding thermal governing which exists to prolong the device’s life.
The Vero 2 has thermal governing and we use trip points to ensure that the ARM and Mali are governed accordingly when things get too hot.
This would mean that you hit the first thermal trip point.
Trip points are defined as follows:
trip_point=<60 1536001 1536001 511 511 3 2
65 1488001 1488001 426 426 3 2
70 1320001 1320001 328 328 2 2
75 1200001 1200001 328 328 2 1
80 1000001 1000001 328 328 2 1
90 816001 816001 251 251 1 1
Which translates to:
Temperature / trip | CPU freq highest | CPU freq lowest | Mali freq highest | Mali freq lowest | Mali PP highest | Mali PP lowest
If the device freezes on certain CPU frequencies, this usually means:
BTW: this command does not mean it is stressing the third core; but rather that it will use 3 CPU bound threads in the stress test.
osmc@testvero2:~$ stress --cpu 3 --io 4 --vm 2 --vm-bytes 128M --timeout 10sstress: info: [3237] dispatching hogs: 3 cpu, 4 io, 2 vm, 0 hdd
stress: info: [3237] successful run completed in 10s
osmc@testvero2:~$ cat /sys/class/thermal/thermal_zone0/temp 56062
You are correct, I misrepresented the information
I agree, the device is having issues. How do I go about seeking a replacement unit?
Hi
I’m really sorry to hear you’re having problems with your device. If you send an email to sales@osmc.tv, we can arrange for a return.
We will inspect the device upon receipt.
The issue has been resolved for the customer without the need for a replacement.