We have a batch of NX products, the system will be directly down after a long time of operation, and it can run normally after restart. Could you please help to determine the problem of the block? It is known that the temperature and memory occupation are normal, the attachment is the relevant dmesg information.
dmesg.log (257.5 KB)
Can you share more detail info?
What applications are running on device?
What’s the temperature you saw?
Which JetPack version?
Failure rate?
Thanks
Our client’s software is running on the device.
memory.log (66.0 MB)
The attachment is the value monitored by the tegrastats command.
Jetpack4.5.1 was used，
This happens in about 30 or 40 out of 100 devices.
What kind of client SW?
Have you confirmed there is no memory leakage?
According to the tegrastats test, the memory usage is normal, but the specific software information of the customer is not clear. Do you mean that the customer’s software may cause memory leakage, which may lead to the system shutdown directly?
Hi,
Suggest you to run same software on NV devkit and see if it would repro same error.
The test was carried out on the official base plate before, but the troublesome situation is that the problem can not be repeated at any time, and the time period may be quite long. This batch of equipment has been put into the terminal for testing last year, and the equipment has been in operation until now.
It has been running normally for more than a year. In addition, after the restart of the equipment, it can still run normally.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.