Tx2 Dev Kit Turns Off Automatically

My TX2 will turn off after a few hours of inactivity. This happens a few times a day hasn’t occurred while I am present. Is this a “feature” with a setting or a malfunction that requires replacement?

Someone else will need to comment, but it is possible you are running into Ubuntu’s sleep/power save features. There are other related features for performance settings as well. I couldn’t tell you how to verify if this is the case or not. In part the answer might depend on whether you’re using an old release or a current release (R28.1 is the most recent…release can be checked with “head -n 1 /etc/nv_tegra_release”).

I just watched it power down during use. Nothing subtle, it just turned off. Most leds go off (3 of 4). CR5 stays on. A press of the power button and it boots right back up. I must not have purchased the extended use license.

I spoke too soon it just crashed again… Time to check the peripherals.

Hi roblight,

Are you using JetPack3.1 on TX2?
You means, when you boot up and stay, it will shutdown or crash automatically?

Thanks!

Hi carolyuu

Yes I am using JetPack3.1 on TX2 and it will crash / shutdown without notice

Is the fan spinning? Is the heat sink warm when this happens? If so, it may be thermal protection.

What power are you using to drive it? Is that power supply warm?

If you check the system logs after you restart the system again, what do they say right before the shutdown?

(“sudo journalctl” is the tool to view various system logs)

Finally, if you install a USB TTL UART listener on the UART0 console, an look at what’s printed from the kernel, does it print anything when it shuts down?

Snarky thank-you for the troubleshooting tips. Fan is on and heat sink cool, no apparent issues with power or peripherals

The kit shut down last night again while idle. I found some talk about the wifi IC being flakey so I disabled wifi and have been processing data non-stop for 14 hours.

At some point when I’m not on a time crunch I’ll validate the problem

Hi roblight,

No idea about this issue.
Could you provide kernel log first?

Thanks!

The system is still crashing. Looking at kern.log at the time of crash

Dec 20 04:35:30 tegra-ubuntu kernel: [ 3782.342276] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:35:31 tegra-ubuntu kernel: [ 3783.342264] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:35:40 tegra-ubuntu kernel: [ 3792.366264] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:35:41 tegra-ubuntu kernel: [ 3793.366265] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:35:50 tegra-ubuntu kernel: [ 3802.390259] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:35:51 tegra-ubuntu kernel: [ 3803.390259] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:36:00 tegra-ubuntu kernel: [ 3812.414274] extcon-gpio-states external-connection:extcon@1: Cable state 2
Dec 20 04:36:01 tegra-ubuntu kernel: [ 3813.414255] extcon-gpio-states external-connection:extcon@1: Cable state 2
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^$
Dec 20 15:40:18 tegra-ubuntu kernel: [ 0.000000] Initializing cgroup subsys cpuset
Dec 20 15:40:18 tegra-ubuntu kernel: [ 0.000000] Initializing cgroup subsys cpu
Dec

What is connected to the Jetson? Is the device tree modified? Perhaps this is a side effect of something external touching the Jetson.

@Carolyuu I am having the similar issue:

I flashed the OS with Jetpack3.1,
After putting the TX2 on the development board, and connecting the power cable, the red LED indicator turned on. And after pressing the poweron button, two green LEDs turned on and the fan started running.
But after several seconds of booting up, it turned off itself and only leaving the red LED on. Looks like it shut down the bootup process.
I tried bootup several times but failed

Cui

May be hdmi monitor affects the boot process?

Earlier this week I had 3 or 4 power off crashes in 2 days. As of 12/20 I have been processing data and have no crashes.

After my latest boot 2 days ago I ran ./jetson_clocks.sh

Not convinced this did the trick because I have had power down crashes with and without running the clock script.

roblight,

Do you use our devkit?

@robtlight has you been able to fix this issue. I ran into the same problem. Thanks.

the same issue[powering off] since a couple of week ago
kern.log (845 KB)

@andrey1984, it seemed like we have an issue with heat.

If we turn off the board, keep it cool, then turn it back on, it will run for about 2 hour or so before crashing. After that, if I turned it back on, it will crash immediately.

The problem is, I had check the CPU temperature it stayed constant at 36C. The fan is turned on through out.

Do you have any comment? What else I should do to diagnose this issue?

sensors
BCPU-therm-virtual-0
Adapter: Virtual device
temp1:        +26.0°C  (crit = +101.0°C)

MCPU-therm-virtual-0
Adapter: Virtual device
temp1:        +26.0°C  (crit = +101.0°C)

GPU-therm-virtual-0
Adapter: Virtual device
temp1:        +24.0°C  (crit = -40.0°C)

Tboard_tegra-virtual-0
Adapter: Virtual device
temp1:        +22.0°C  (crit = +107.0°C)

Tdiode_tegra-virtual-0
Adapter: Virtual device
temp1:        +22.5°C  (crit = +107.0°C)

thermal-fan-est-virtual-0
Adapter: Virtual device
temp1:        +24.9°C
uptime
 22:09:03 up 21:07,  2 users,  load average: 1.51, 1.74, 1.91

There is Autumn coming and the outside temperature got much colder.
I am observing and gathering statistics, as I do not have better ideas.

which command did you use to print this info? thanks.

ok, I got it. Mine is like the following. Seem like higher than yours.

BCPU-therm-virtual-0
Adapter: Virtual device
temp1: +35.0°C (crit = +101.0°C)

MCPU-therm-virtual-0
Adapter: Virtual device
temp1: +35.0°C (crit = +101.0°C)

GPU-therm-virtual-0
Adapter: Virtual device
temp1: +41.0°C

AO-therm-virtual-0
Adapter: Virtual device
temp1: +33.5°C (crit = -40.0°C)

Tboard_tegra-virtual-0
Adapter: Virtual device
temp1: +33.0°C (crit = +107.0°C)

Tdiode_tegra-virtual-0
Adapter: Virtual device
temp1: +33.5°C (crit = +107.0°C)

thermal-fan-est-virtual-0
Adapter: Virtual device
temp1: +34.4°C