hi im using J101 (https://www.seeedstudio.com/Jetson-10-1-A0-p-5336.html) . its is equipped with quectel ec25 module and i connect 2 usb cameras NLK2MP08s with them… everything worked fine but after sometime like after 2 hours my jetson gets off for no reason apparently. i have attached me kernel and system logs. pleas check and help me escaping out of this fuss.
kern.log (163.3 KB)
logs.zip (776.7 KB)
Please use serial console to monitor the log and see what got printed when your “jetson gets off” situation happened.
unfortunately i cant observe any realtime operations. me seeed board is installed in a truck. cna u tell me is there any way to save those serial debug logs
maybe trying to use ramoops.
there are kernel logs and sys logs in log directory, can i use them too see what causes the device to gets off or reboot. ??does these logs store all the events?
If your system got crashed accidentally, then syslog and kernel logs won’t have the logs for crash reason.
That is the purpose I gave you those methods.
ok thankyou… what does kernel and sys logs give then ?
Just common driver info when device is still working.
For example, if you hotplug a usb mouse, the kern.log will record you add and remove a usb device.
The problem here is your system is in the edge of crash. When system crashes, syslog is already dead, it won’t have time to record the panic log in it.
thank you so much, one last question plz
if use jfrog connect to remotely access my nano deployed in a truck. if i jumper serial rx tx wire and deployed it in vehicle , is there any chance my nano works properly with all functions mean time i can see serial console???
Sorry that I have no idea about what is jfrog.
thankyou i will be back with some other findings
im studying ramoops u shared above, can you plz clarify the diff between power cycle and kernel panic. if there is low current / power issues then the file will be available in dmesg-ramoops right? like any power issue will be stored ? what will be actually written? and any kernel issue causing the nano to gets off or crash will fall under kernel panic category?
That’s another question that we would like to ask you too.
Does your system crash mean your board just shutdowns and never come back or it goes off state but still be able to reboot for a while?
A software crash will trigger watchdog timer so your board would still be reboot.
But if this is sudden shutdown and never came back, then it could be power issue and that would be hardware problem.
our nano gets off randomly then either reboot sometimes itself or when truck ignition gets restarted
That is the point to dump log. Your comment sounds like a random one that we cannot guarantee what you hit either.
im trying to implement ramoops, i will share the logs.
Also, if nano gets off randomly and the power is continuous is there a chance that nano gets on again itself???
im unable to clone this
git clone -b toradex_5.4-2.1.x-imx git://git.toradex.com/linux-toradex.git
please give me updated link
You don’t need to follow the whole page. ramoops is already supported on Jetson nano.
What you need to do is just that Console log section
can you please clarify what is the difference among these files?
pmsg-ramoops-0
console-ramoops-0
dmesg-ramoops-0
i have attached logs…can u please check these and give me an illustration…these are in lab device
ramoops logs.zip (42.2 KB)
Where is the log?